CN114185760A - System risk assessment method and device and charging equipment operation and maintenance detection method - Google Patents

Abstract

The application discloses a system risk assessment method and device and a charging equipment operation and maintenance detection method, wherein the system risk assessment method comprises the following steps: acquiring system change data of a target system in target time; the system change data comprises system early warning information and device names corresponding to system changes; based on the system change data, determining early warning information corresponding to the target system change by using a natural language processing technology; acquiring time sequence data of a target system in target time, and obtaining the variation trend of a monitoring index corresponding to the target system by using a machine learning algorithm based on the time sequence data; and carrying out system risk assessment on the target system based on the early warning information and the change trend. According to the method and the device, the risk of the system change is evaluated through the early warning information of the system change obtained based on natural language processing and the monitoring index trend abnormity obtained based on machine learning, and the accuracy of evaluation of the risk of the system change under the micro-service architecture is improved.

Description

System risk assessment method and device, charging equipment operation and maintenance detection method

technical field

The invention relates to the field of micro-service systems, in particular to a system risk assessment method and device, and a charging equipment operation and maintenance detection method.

Background technique

At present, under the impact of imperfect business models and ultra-large-scale traffic, many enterprises have abandoned the traditional monolithic architecture and adopted the micro-service architecture. The micro-service architecture model has independent development, independent deployment, scalability, and reusability. The advantages. However, at the same time, the development, iteration, and operation and maintenance of the microservice system are highly complex. The increase in the number of services increases the probability of one of the services failing, and one service failure may cause the entire system to hang up. It becomes very difficult to locate the fault point. At the same time, due to the large number of services, the workload of deployment and management is very large, and accidents caused by negligence are extremely prone to occur. In addition to some irresistible factors, 80% of the common Internet product accidents are caused by bugs in publishing programs or changes in basic resources. Therefore, it is very important to ensure the original stability and core business of the system in the process of frequent product iteration, release and launch under the complex microservice system. Therefore, how to quickly analyze those anomalies from complex systems that are caused by system changes, and conduct risk assessments on system changes, so as to notify administrators to pay attention and deal with them in a timely manner, is the premise to ensure the stability of the microservice system.

In the prior art, system stability is maintained through grayscale distribution, and grayscale distribution is performed according to traffic or specific data content, so that when a problem occurs, users on the entire network will not be affected. The impact of publishing failures does not fully cover all users or traffic in the production environment, reducing the ability to troubleshoot risks. In the prior art, the version analysis function of Zhuge io is also used to measure the overall effect after the release of this version by analyzing the data of new users and active users after the revision, and comparing the historical data of different versions. However, it relies on the trend of indicators for a period of time after the release to evaluate the program after the release, and cannot find online problems in time, which reduces the ability of system risk investigation.

SUMMARY OF THE INVENTION

In view of this, one aspect of the present invention provides a system risk assessment method, which can improve the accuracy of system risk assessment under a microservice architecture.

Another aspect of the present invention also provides a method for operation and maintenance detection of charging equipment, which can improve the accuracy and detection efficiency of operation and maintenance detection of charging equipment.

Another aspect of the present invention also provides a system risk assessment device, which can improve the accuracy of the system risk assessment under the micro-service architecture.

According to the first aspect of the present invention, the present application discloses a system risk assessment method, which is applied to the system risk assessment under the microservice architecture, including:

Obtain the system change data of the target system within the target time; the system change data includes the system early warning information and the device name corresponding to the system change;

Based on the system change data, use natural language processing technology to determine early warning information corresponding to the target system change;

Acquire the time series data of the target system within the target time, and based on the time series data, use a machine learning algorithm to obtain the change trend of the monitoring indicators corresponding to the target system;

A system risk assessment is performed on the target system based on the early warning information and the change trend.

Optionally, based on the system change data, use natural language processing technology to determine the early warning information corresponding to the target system change, including:

Perform corpus preprocessing on the system early warning information, and extract the first group of keywords from the processed system early warning information;

performing corpus preprocessing on the device name, and extracting a second group of keywords from the processed device name;

Calculate the similarity between the first group of keywords and the keywords in the second group of keywords, and determine early warning information corresponding to the target system change based on the similarity.

Optionally, the target time includes a system change time of the target system and a time within a preset time period after the change.

Optionally, based on the time series data, using a machine learning algorithm to obtain the change trend of the monitoring indicators corresponding to the target system, including:

Classify the time series data according to a preset classification standard to obtain corresponding target type data; the target type data includes volatile data, periodic data and stable data;

Using a preset detection algorithm corresponding to the target type data, anomaly detection is performed on the target type data, and the change trend of the monitoring index corresponding to the target type data is obtained, so as to obtain the change of the monitoring index corresponding to the target system. trend.

Optionally, performing anomaly detection on the target type data by using a preset detection algorithm corresponding to the target type data to obtain a change trend of the monitoring index corresponding to the target type data, including:

Perform anomaly detection on the volatile data based on Turkey detection, and obtain the change trend of the volatile monitoring indicators;

Perform anomaly detection on the periodic data based on the same chain ratio algorithm, and obtain the change trend of the periodic monitoring indicators;

Anomaly detection is performed on the stable data based on the time series ARIMA algorithm, and the change trend of the stable monitoring indicators is obtained.

Optionally, classifying the time series data according to a preset classification standard to obtain corresponding target type data, including:

Periodically detect the time series data based on the similarity of window data, and classify according to a first preset threshold to obtain the volatile data and the non-volatile data;

Stability detection is performed on the non-volatile data based on the STL algorithm, and the periodic data and the stable data are obtained by classifying according to a second preset threshold.

Optionally, the system risk assessment on the target system based on the early warning information and the change trend includes:

Perform a system risk assessment on the target system based on the early warning information, the change trend and the index level of the monitoring index, to obtain the risk level of the target system;

generating a corresponding assessment report based on the risk level;

Wherein, the indicator levels include core business-level indicators, technology-level indicators, and system resource-level indicators.

According to the second aspect of the present invention, the present application discloses a method for detecting operation and maintenance of charging equipment, including:

Obtain the system change data of the target system within the target time; the target system includes the charging equipment system; the system change data includes the operation and maintenance data of the charging equipment, and the operation and maintenance data of the charging equipment includes the system warning information and the corresponding system change name of the device;

Based on the system change data, use natural language processing technology to determine early warning information corresponding to the target system change;

Acquire the time series data of the target system within the target time, and based on the time series data, use a machine learning algorithm to obtain the change trend of the monitoring indicators corresponding to the target system;

A system risk assessment is performed on the target system based on the early warning information and the change trend.

Optionally, the obtaining of the time series data of the target system within the target time, and based on the time series data, using a machine learning algorithm to obtain the change trend of the monitoring indicators corresponding to the target system, including:

Obtain time series data within the target time during the operation and maintenance detection process of the charging equipment;

Periodically detecting the time series data to determine the periodic characteristic value;

In the case that the periodic characteristic value is smaller than the first preset threshold, it is determined that the time series data belongs to volatile data;

In the case that the periodic characteristic value is greater than or equal to the first preset threshold value, further perform stability detection on the time series data to determine the stability characteristic value;

In the case that the stability characteristic value is greater than the second preset threshold, it is determined that the time series data belongs to stable data;

In the case that the stability characteristic value is less than or equal to the second preset threshold, it is determined that the time series data belongs to periodic data;

Perform anomaly detection on the time series data according to the target type data to which the time series data belongs, so as to obtain the change trend of the monitoring index corresponding to the target type data, so as to obtain the change trend of the monitoring index corresponding to the target system;

The target type data includes volatile data, periodic data and stable data.

Optionally, the periodic detection of the time series data includes:

Given a reference period T of the time series data;

Taking the reference period T as a dividing point, dividing the time series data, and dividing the time series data into n/T sub-time series units, where n is the length of the time series data;

Comparing each of the sub-time series units to calculate a similarity coefficient, and determining the similarity system as the periodic feature value;

The further performing stability detection on the time series data includes:

Seasonally decompose the time series data by using the moving average algorithm, and decompose it into a seasonal cycle component, a long-term trend component and a random residual component;

The stability characteristic value is determined according to the variance of the random residual component.

Optionally, performing anomaly detection on the time series data according to the target type data to which the time series data belongs, including:

In the case of determining that the time series data belongs to volatile data, use Turkey,s Test or 3-Sigema algorithm to perform dynamic threshold detection on the time series data, so as to obtain the change trend of the monitoring index corresponding to the target system;

When it is determined that the time series data is stable data, the SARIMAX or moving average algorithm is used to perform baseline threshold detection on the time series data, so as to obtain the change trend of the monitoring index corresponding to the target system;

When it is determined that the time series data belongs to periodic data, a machine learning classification model and a regression model are used to perform anomaly detection on the time series, so as to obtain the change trend of the monitoring indicators corresponding to the target system.

According to a third aspect of the present invention, the present application discloses a system risk assessment device, which is applied to system risk assessment under a micro-service architecture, including:

The data acquisition module is used to acquire the system change data of the target system within the target time; the system change data includes the system early warning information and the name of the change device;

an early warning information determination module, configured to determine early warning information corresponding to the target system change by using natural language processing technology based on the system change data;

A change trend determination module, configured to obtain the time series data of the target system within the target time, and based on the time series data, use a machine learning algorithm to obtain the change trend of the monitoring indicators corresponding to the target system;

A risk assessment module, configured to perform system risk assessment on the target system based on the early warning information and the change trend.

In a fourth aspect, the present application also discloses an electronic device, comprising:

memory for storing computer programs;

A processor for executing the computer program to implement the aforementioned system risk assessment method.

In a fifth aspect, the present application further discloses a computer-readable storage medium for storing a computer program; wherein the computer program implements the aforementioned system risk assessment method when executed by a processor.

In this application, by acquiring the system change data of the target system within the target time, the system change data includes the system early warning information and the device name corresponding to the system change; and then based on the system change data, natural language processing technology is used to determine the The target system changes the corresponding early warning information; and obtains the time series data of the target system within the target time, and then uses the machine learning algorithm to obtain the monitoring indicators corresponding to the target system based on the time series data. change trend; finally, perform a system risk assessment on the target system based on the early warning information and the change trend. It can be seen that the natural language processing technology is used to determine the early warning information corresponding to the system change through the system early warning information within the time range of the system change and the names of the devices affected by the system change within the spatial range, and then combined with the machine learning algorithm to pass the time The change trend of the monitoring indicators corresponding to the system determined by the sequence data, the current risk of the system can be assessed, and the risks brought by software patch release, configuration change, and basic resource change can be quickly and accurately identified. Accuracy of change risk assessments. The method can be applied to the operation and maintenance detection of charging equipment, the operation and maintenance detection of power grids, and/or the integration of vehicle pile and network, as well as the operation and maintenance detection of other detection systems under the micro-service architecture built on the Internet of Things, and has strong versatility.

Description of drawings

In order to explain the embodiments of the present invention or the technical solutions in the prior art more clearly, the following briefly introduces the accompanying drawings that need to be used in the description of the embodiments or the prior art. Obviously, the accompanying drawings in the following description are only It is an embodiment of the present invention. For those of ordinary skill in the art, other drawings can also be obtained according to the provided drawings without creative work.

1 is a flowchart of a system risk assessment method provided in Embodiment 1 of the present application;

2 is a flowchart of a specific system risk assessment method provided in Embodiment 1 of the present application;

3 is a flowchart of a specific system risk assessment method provided in Embodiment 2 of the present application;

4 is a flowchart of a specific system risk assessment method provided in Embodiment 3 of the present application;

Fig. 5 is a kind of time series data classification flowchart provided by this application;

FIG. 6 is a flowchart of a charging equipment operation and maintenance detection method provided in Embodiment 4 of the present application;

Figure 7 is a schematic diagram of the operation and maintenance system of the charging equipment;

FIG. 8 is a specific flowchart of step S43 of the charging equipment operation and maintenance detection method;

9 is a schematic structural diagram of a system risk assessment device provided by the present application;

FIG. 10 is a structural diagram of an electronic device provided by the application.

Detailed ways

In the prior art, the system stability is maintained by means of grayscale distribution, but it cannot completely cover all users or traffic in the production environment, which reduces the capability of risk investigation. It also uses Zhuge io's version analysis function to analyze the data of new users and active users after the revision, compare the historical data of different versions, and measure the overall effect after the release; but it can't find online problems in time. Reduced ability to troubleshoot systemic risks. In order to overcome the above technical problems, the present application provides a system-based risk assessment method, which can improve the accuracy of system change risk assessment under the micro-service architecture.

Embodiment 1 of the present application discloses a system risk assessment method, which is applied to system risk assessment under a microservice architecture. Referring to FIG. 1 , the method may include the following steps:

Step S11: Obtain system change data of the target system within the target time; the system change data includes system warning information and device names corresponding to the system change.

In this embodiment, the system change data of the target system within the target time is obtained first. The system change data includes system warning information, and the device name corresponding to the system change, that is, the name of the device affected by the system change; the device name includes But not limited to data centers, business units, functional modules, service components, nodes and hosts. Wherein, the above-mentioned target time includes the system change time of the target system and the time within the preset time period after the change; it can be understood that the above-mentioned target time includes the period from the start of the change to the end of the change of the target system, and the time when the target system is changed. After the end, the duration is a preset period of time. Since the impact of the system change will not stop immediately after the system change is completed, by obtaining the system change data within a period of time after the change, the integrity of the change data can be improved. The above-mentioned system changes may include, but are not limited to, software release, configuration changes, basic setting adjustments, database publishing and subscription adjustments, and middleware operation and maintenance adjustments. Specifically, it may include host expansion, machine restart, and network adjustments.

Step S12: Based on the system change data, use natural language processing technology to determine early warning information corresponding to the target system change.

In this embodiment, after the system change data is obtained, the natural language processing technology is used to determine the early warning information corresponding to the change of the target system from the system change data.

It is understandable that the microservice architecture generally has a relatively complete monitoring and early warning system, which can detect faults as much as possible and issue system early warning information. However, due to the numerous components in the microservice architecture, early warning information often occurs, and abnormality occurs when the system changes. When there is an accident or an accident, the system warning information explodes, and it is difficult to find the warning information that is really related to the system change. Therefore, by combining the system early warning information with the device name corresponding to the system change, and using corpus preprocessing and feature extraction, the early warning information associated with the target system change can be determined from the system change data.

Step S13: Acquire time series data of the target system within the target time, and based on the time series data, use a machine learning algorithm to obtain a change trend of the monitoring indicators corresponding to the target system.

In this embodiment, while the early warning information corresponding to the target system change is determined through the system change data, the time series data of the target system within the target time is obtained, and then the above target system is obtained based on the time series data using a machine learning algorithm The change trend of the monitoring indicators corresponding to the system. It is understandable that there are many components (that is, services) in the microservice architecture, and each component needs to monitor different indicators. However, when an abnormality occurs in a component, the trend of the monitored indicators will undergo abnormal mutation, and the overall increase or decrease. , so you can determine whether there is any abnormality in the component or service by trend detection of the monitoring indicators. Specifically, different types of monitoring indicators can be detected through different algorithms to obtain corresponding change trends.

Step S14: Perform system risk assessment on the target system based on the early warning information and the change trend.

In this embodiment, after obtaining the warning information corresponding to the change of the target system and the change trend of the monitoring indicators of the target system, the system risk assessment is performed on the target system according to the warning information and the change trend, and the risks existing in the target system are determined. and risk level. It can be understood that, for example, as shown in FIG. 2 , in this embodiment, after the system is changed, the early warning information associated with the change is determined by monitoring the relevant data, and the abnormality detection of the monitoring index trend of the target system, and finally the system risk is checked for the target system. Through comprehensive evaluation, system problems related to system changes can be understood through early warning information, and whether components or services have abnormal occurrences can be judged by monitoring the change trend of indicators, so that system risks can be accurately obtained through comprehensive analysis.

It can be seen from the above that in this embodiment, the natural language processing technology is used to determine the warning information corresponding to the system change through the system early warning information within the time range of the system change and the names of the devices affected by the system change within the space range, and then combine Using the machine learning algorithm to determine the change trend of the monitoring indicators corresponding to the system through the time series data, to evaluate the current risk of the system, it can quickly and accurately identify the risks brought about by software patch releases, configuration changes, and basic resource changes. Accuracy of system change risk assessment under microservice architecture.

The second embodiment of the present application discloses a specific system risk assessment method, which is applied to the system risk assessment under the microservice architecture. Referring to FIG. 3 , the method may include the following steps:

Step S21: Obtain system change data of the target system within the target time; the system change data includes system warning information and device names corresponding to the system change.

Step S22: perform corpus preprocessing on the system early warning information, and extract the first group of keywords from the processed system early warning information; perform corpus preprocessing on the device name, and extract the first group of keywords from the processed system early warning information; Extract the second set of keywords.

In this embodiment, after obtaining the above-mentioned system warning information and the device name corresponding to the system change, the team performs corpus preprocessing on the above-mentioned system warning information and the device name, and then can extract the corresponding data from the processed data according to the specific business logic. keywords to obtain the corresponding first group of keywords and the second group of keywords. Among them, the above-mentioned corpus preprocessing includes but is not limited to word segmentation, part-of-speech tagging, named entity recognition and removal of stop words; specifically, the system warning information text is firstly divided into word-based data structures by word segmentation tools, and then The part-of-speech tagging tool is used to tag the results of the above word segmentation, and mark them as verbs, nouns or adjectives, etc.; and then use the named entity recognition tool to identify the proper names of the tagged words to identify proper nouns, such as: data center , business unit, host, node, and finally remove the stop words in the text according to the Chinese stop word list. Among them, the above-mentioned processing tools for expected preprocessing include but are not limited to jieba, Hanlp, NLTK and StandfordCoreNLP. The above Chinese stop word lists include but are not limited to Harbin Institute of Technology stop word list, Baidu stop word list, and Sichuan University Machine Intelligence Laboratory stop word database.

Step S23: Calculate the similarity between the first group of keywords and the keywords in the second group of keywords, and determine early warning information corresponding to the target system change based on the similarity.

In this embodiment, after obtaining the first group of keywords and the second group of keywords, the similarity of the keywords in the first group of keywords and the second group of keywords is calculated, that is, the key obtained based on the early warning information is calculated. word, the similarity with the keyword obtained based on the device name corresponding to the system change, and then filter out the keywords whose similarity is greater than the preset threshold as the early warning information corresponding to the target system change. The calculation algorithm of the above similarity may include, but is not limited to, Euclidean distance, McHatton distance, cosine similarity, Pearson acquaintance, K-means and DBSACN.

Step S24: Acquire the time series data of the target system within the target time, classify the time series data according to a preset classification standard, and obtain corresponding target type data.

In this embodiment, the time series data of the target system within the above target time is obtained, and then the above time series data is classified according to a preset classification standard to obtain corresponding target type data; the above target type data may include volatile data, Periodic data and stable data, that is, classifying time series data according to data characteristics to obtain corresponding volatile data, periodic data and stable data. It is understandable that there are many types of monitoring indicators and their relationships are complex, but the indicators themselves have the characteristics of periodicity, regular spikes, overall rise or fall, and low peak periods. The influencing factors of indicators include holidays, temporary activities, weather, epidemics and other factors. Different types of indicators can be detected by different algorithms to obtain better detection results, so the time series data can be automatically classified according to the characteristics of the indicators.

Step S25: Use a preset detection algorithm corresponding to the target type data to perform anomaly detection on the target type data, and obtain the change trend of the monitoring index corresponding to the target type data, so as to obtain the monitoring corresponding to the target system. trend of the indicator.

In this embodiment, after the target type data is obtained, a preset detection algorithm corresponding to the target type data is used to perform anomaly detection on the target type data, and the change trend of the monitoring index corresponding to the target type data is obtained, so as to obtain the corresponding target system change trend of monitoring indicators. It can be understood that using different detection algorithms for different types of data can improve the accuracy of the detection results.

Step S26: Perform a system risk assessment on the target system based on the early warning information, the change trend and the index level of the monitoring index, and obtain the risk level of the target system.

In this embodiment, after obtaining the above-mentioned early warning information and the change trend, the system risk assessment is performed on the target system according to the early warning information, the change trend and the index level of the above-mentioned monitoring index, and the risk level of the target system is obtained, wherein the above-mentioned index level may include Core business-level indicators, technical-level indicators and system resource-level indicators.

It can be understood that the importance of the system status reflected by different monitoring indicators to the system operation is different. The monitoring indicators can be divided into three levels of influence, namely core business-level indicators, technical-level indicators, and system resource-level indicators. The degree decreases successively. Among them, the core business-level indicators can be the real-time overall health status of the business. From the core business-level indicators, you can intuitively see the degree of damage to the business, the market, and the impact. When the core business-level indicators are abnormal after the system is changed, especially When it is damaged, it can reflect that the risk of system changes is extremely high, which has affected the stability of the core business. Among them, the technical-level indicators can be the real-time health status of components or services, and the technical-level indicators can reflect the upstream and downstream call volume (TPS), delay, success rate, and number of threads. When the trend is abnormal, such as increased latency, decreased success rate, sudden drop in TPS, sudden increase or sudden drop in the number of threads, etc., technical-level indicators can determine that the component or service has been affected by system changes. System resource-level indicators can be real-time middleware, host or system-level health status, such as CPU usage, memory usage, network traffic, etc. When a system failure occurs, system resource-level indicators usually appear abnormal one after another.

The system risk assessment of the target system is carried out according to the early warning information, the change trend of the monitoring indicators and the indicator level of the monitoring indicators, and the risk level of the target system is obtained. The above risk levels can include no risk, low risk, medium risk and high risk. And according to the index level of the monitoring index, the monitoring index with the change trend is weighted and calculated, and the risk level is finally determined; for example, according to the result, no system-related early warning information and no monitoring abnormal change trend can be determined as no. Risk; the situation where the system-related early warning information and the abnormal change trend are detected is determined as medium risk; the situation where the abnormal change trend of the core business-level indicators is monitored is determined as high risk.

Step S27: Generate a corresponding assessment report based on the risk level.

In this embodiment, a corresponding assessment report is generated according to the determined risk level. For example, the assessment report corresponding to no risk can be a successful change, the assessment report corresponding to medium risk can be a recommendation to observe closely, and the assessment report corresponding to a high risk can be a recommendation to roll back immediately, so that the administrator can be accurately notified of the current status of the target system. status, and how to handle it accordingly.

For the specific process of the foregoing step S21, reference may be made to the corresponding content disclosed in the foregoing embodiments, which will not be repeated here.

It can be seen from the above that the first group of keywords is extracted from the processed system early warning information by corpus preprocessing on the system early warning information; The second set of keywords, and then the similarity between the first set of keywords and the keywords in the second set of keywords is calculated, and early warning information corresponding to the target system change is determined based on the similarity. Based on corpus preprocessing and feature extraction to determine the warning information associated with the change, the key words associated with the system change can be extracted from the system warning information that has skyrocketed in the event of a failure, and then the device name can be calculated with the change space range. The degree of linkage determines the warning information corresponding to the change of the target system. And by acquiring the time series data of the target system within the target time, classifying the time series data according to the preset classification criteria, to obtain the corresponding target type data, and then using the preset detection algorithm corresponding to the target type data to classify the target type data. Anomaly detection is performed to obtain the change trend of the monitoring index corresponding to the target type data, so as to obtain the change trend of the monitoring index corresponding to the target system. The trend detection of monitoring indicators based on machine learning can intelligently select a classification model according to the data characteristics of the indicators, and make abnormal judgments based on the characteristics of historical data, and then judge whether the changes have an impact on components and services, and whether the core business is damaged.

The third embodiment of the present application discloses a specific system risk assessment method, which is applied to the system risk assessment under the microservice architecture. Referring to FIG. 4 , the method may include the following steps:

Step S31: Obtain system change data of the target system within the target time; the system change data includes system warning information and device names corresponding to the system change.

Step S32: Based on the system change data, use natural language processing technology to determine early warning information corresponding to the target system change.

Step S33: Acquire time series data of the target system within the target time.

Step S34: Periodically detect the time series data based on the similarity of the window data, and classify according to a first preset threshold to obtain the volatile data and the non-volatile data.

In this embodiment, for example, as shown in FIG. 5 , firstly, periodic detection is performed on time series data based on the similarity of window data, and volatile data and non-volatile data are obtained by classification according to a first preset threshold. Specifically, by calculating the similarity of the time series windows, the similarity calculation can be calculated based on the Pearson correlation and the dynamic programming distance. The above-mentioned first preset threshold may be 0.8.

Step S35: Perform stability detection on the non-volatile data based on the STL algorithm, and classify the periodic data and the stable data according to a second preset threshold.

In this embodiment, as shown in FIG. 5 , the stability detection is performed on the non-volatile data based on the STL algorithm, and periodic data and stable data are obtained by classifying according to the second preset threshold. Specifically, the non-volatile time series data is decomposed by the STL algorithm of the additive model, that is, time series = periodic component + trend component + residual (residual), and whether it is stable data is determined according to the variance of the residual. The above-mentioned second preset threshold may be 0.002.

Step S36: Perform anomaly detection on the volatile data based on Turkey detection to obtain a change trend of the volatile monitoring index.

In this embodiment, anomaly detection is performed on the above-mentioned volatile data based on Turkey detection, and a change trend of the volatile monitoring index is obtained. Specifically, the anomaly detection algorithm based on Turkey's TEST may include: firstly calculating the first quartile (Q1), median and third quartile (Q3) of the change trend data, and let IQR=Q3-Q1 , if IQR is between Q3+k(IQR) and Q1-k(IQR), it can be considered that there is no abnormality in the volatile data, otherwise it is determined that the corresponding volatile monitoring index has an abnormal trend, where k is the preset threshold, 1.5 or 3 are desirable.

Step S37: Perform anomaly detection on the periodic data based on the same chain ratio algorithm to obtain a change trend of the periodic monitoring index.

In this embodiment, anomaly detection is performed on periodic data based on the same-chain ratio algorithm to obtain a change trend of periodic monitoring indicators. Specifically, the anomaly detection algorithm based on the year-on-year algorithm may include: extracting the 7-day month-on-month data at the time, removing outliers, and completing default values; then calculating the mean and standard deviation of the year-on-year data; and finally passing the periodic abnormality judgment formula:| t-m|>b×σ, to judge whether there is an abnormal trend in the corresponding periodic monitoring index; wherein, the above t is the current value, m is the year-on-year average, b is the preset threshold, and σ is the standard deviation, if |t-m| is greater than b× σ means that there is an abnormal trend in the corresponding periodic monitoring index.

Step S38: Perform anomaly detection on the stable data based on the time series ARIMA algorithm to obtain a change trend of the stable monitoring index, so as to obtain a change trend of the monitoring index corresponding to the target system.

In this embodiment, anomaly detection is performed on the stable data based on the time series ARIMA (Autoregressive Integrated Moving Average model, differential integrated moving average autoregression) algorithm to obtain the change trend of the stable monitoring index, and finally the change trend of the variable monitoring index is obtained according to the , the change trend of the periodic monitoring index and the change trend of the stable monitoring index, so as to obtain the change trend of the monitoring index corresponding to the above target system. Specifically, the anomaly detection algorithm based on the time series ARIMA algorithm may include: first, according to the preset stability threshold and white noise threshold, the first-order difference of the time series is subjected to stability and white noise tests, and a stable time series that meets the conditions is obtained. Data, the stable time series data is divided into: trend component + period component + residual (residual) according to the STL algorithm, and then the ARIMA model is trained on the trend component of the historical data, and the grid search method is used to automatically optimize parameters. , get the predicted trend, and finally judge whether there is an abnormal trend in the corresponding stable monitoring index according to the stable abnormal judgment formula: |r-p|>c×l. Among them, r is the actual data, p is the predicted data, c is the preset threshold, and l is the historical residual; wherein the predicted data=predicted trend+historical period. If |r-p| is greater than c×l, it is determined that the corresponding stable monitoring index has an abnormal trend.

Step S39: Perform system risk assessment on the target system based on the early warning information and the change trend.

Wherein, for the specific processes of the above-mentioned steps S31 to S33 and S39, reference may be made to the corresponding contents disclosed in the foregoing embodiments, which will not be repeated here.

It can be seen from the above that anomaly detection is performed on the volatile data based on Turkey detection, and the change trend of the volatile monitoring index is obtained; anomaly detection is performed on the periodic data based on the same chain ratio algorithm to obtain the change trend of the periodic monitoring index; The time series ARIMA algorithm performs anomaly detection on the stable data, and obtains the change trend of the stable monitoring index, so as to obtain the change trend of the monitoring index corresponding to the target system. By using different anomaly detection algorithms for different types of data, the accuracy of index anomaly detection is improved; combined with the early warning information of system changes, this embodiment can quickly and accurately identify system change zones such as software patch release, configuration change, and basic resource change. This can effectively reduce the impact of system changes on the stability of the microservice system.

The operation and maintenance management of charging equipment is an important link in the power system. The daily inspection of charging piles is directly related to the stability, reliability and failure rate of charging equipment. Safety hazards, and repairs to ensure the stable and stable operation of charging facilities. The high incidence of charging pile faults is mainly in the energy path, among which charging modules, DC contactors, fuses, gun wires, gun heads, etc. are the main ones. The damage to the gun wire and gun head is mainly caused by mechanical damage and electrical faults caused by mechanical damage. When hidden dangers occur, inspectors are easy to find. The hidden dangers of charging modules, DC contactors, and fuses can only be realized through data acquisition and data analysis. By installing monitoring and measuring points inside the charging pile, the voltage across the DC contactor is measured when the charging pile is charging, and the data of each voltage drop is stored in the memory. At present, in the operation and maintenance of charging facilities, fault diagnosis and safety operation and maintenance service systems of charging stations are generally not perfect. There is a lack of intelligent fault diagnosis and safety warning for the charging process, and potential equipment defects and hidden dangers cannot be discovered in advance.

According to the embodiments of the present disclosure, for the intelligent operation and maintenance of charging equipment, Embodiment 4 of the present application provides a method for detecting operation and maintenance of charging equipment, as shown in FIG. 6 .

As an example, in this embodiment, for the operation and maintenance detection of charging equipment, in the operation and maintenance of charging equipment, the electric vehicle is used as the carrier, and the electric vehicle is the vehicle terminal, and the interactive communication between the AC and DC charging piles and the charging network. At present, 3G LTE, 4G LTE, GPRS and other wireless communication methods are mainly used between the electric vehicle on-board terminal, the AC and DC charging piles and the charging network, and the data is uploaded to the cloud platform management center in real time. The system architecture of the vehicle pile network is shown in Figure 7. , the system mainly includes on-board terminal 11, charging pile 12, and charging network 13, energy storage equipment 14, and power grid 15. Wireless communication, wireless communication between charging network 13, energy storage device 14 and power grid 15, wireless communication between multiple vehicle terminals 11, and wireless communication between multiple charging piles 12 constitute a charging sharing network intelligent operation and maintenance system. The intelligent operation and maintenance system collects charging full data information for the power distribution system, charging system and charging vehicle of the charging station through the charging pile probe, DTU, image recognition and collection, and portable diagnostic tools. Data and station probe data support the health status assessment of charging piles and charging stations, comprehensively consider equipment alarms and station health status assessment, and generate operation and maintenance inspection plans on demand.

Referring to Figure 6, a method for detecting operation and maintenance of charging equipment includes:

Step S41: Obtain system change data of the target system within the target time; the target system includes a charging equipment system; the system change data includes the operation and maintenance data of the charging equipment, and the operation and maintenance data of the charging equipment includes system warning information and The device name corresponding to the system change;

In this embodiment, the system change data of the target charging equipment system within the target time of the operation and maintenance process is first obtained, the system change data includes the operation and maintenance data of the charging equipment, and the operation and maintenance data of the charging equipment includes system warning information and The name of the device corresponding to the system change, that is, the name of the device affected by the system change. Wherein, the above-mentioned target time includes the system change time of the target system and the time within the preset time period after the change; it can be understood that the above-mentioned target time includes the period from the start of the change to the end of the change of the target system, and the time when the target system is changed. After the end, the duration is a preset period of time. Since the impact of the system change will not stop immediately after the system change is completed, by obtaining the system change data within a period of time after the change, the integrity of the change data can be improved.

Step S42: Based on the system change data, use natural language processing technology to determine early warning information corresponding to the target system change;

In this embodiment, after the system change data is obtained, the natural language processing technology is used to determine the early warning information corresponding to the change of the target system from the system change data.

It is understandable that the intelligent operation and maintenance system of charging equipment has a relatively complete monitoring and early warning system, which can detect faults as much as possible and issue system early warning information. When system changes cause anomalies or accidents, system early warning information explodes, and it is difficult to find early warning information that is truly related to system changes. Therefore, by combining the system early warning information with the device name corresponding to the system change, and using corpus preprocessing and feature extraction, the early warning information associated with the target system change can be determined from the system change data.

Step S43: Acquire time series data of the target system within the target time, and based on the time series data, use a machine learning algorithm to obtain a change trend of the monitoring indicators corresponding to the target system.

In a specific implementation, as shown in Figure 8, this step specifically includes:

S431: Acquire time-series data within the target time during the operation and maintenance detection process of the charging equipment;

S432: Periodically detect the time series data to determine a periodic feature value;

In a specific implementation, the periodic detection of the time series data includes:

Given a reference period T of the time series data;

Taking the reference period T as a dividing point, dividing the time series data, and dividing the time series data into n/T sub-time series units, where n is the length of the time series data;

A similarity coefficient is calculated for each of the sub-time series units by a pairwise comparison, and the periodic feature value is determined.

In some embodiments, optionally, using Pearson correlation to compare each of the sub-time series units pairwise to calculate a similarity coefficient to determine the periodic feature value, including:

Pearson correlation is used to calculate the similarity coefficient for each of the sub-time series units, and the Pearson similarity coefficient is expressed as:

X_T＝{x₁,...,x_T}，Y_T＝{y₁,...,y_T}表示两个子时间序列单元；若X_T＝Y_T，则COR(X_T,Y_T)＝1，表示两个所述子时间序列单元完全一致；若X_T＝-Y_T，则COR(X_T,Y_T)＝-1，表示两个所述子时间序列单元负相关；-1≤COR(X_T,Y_T)≤1。in,

X _T ={x ₁ ,...,x _T }, Y _T ={y ₁ ,...,y _T } represents two sub-time series units; if X _T =Y _T , then COR(X _T ,Y _T )=1, indicating that the two sub-time series units are completely consistent; if X _T =-Y _T , then COR(X _T , Y _T )=-1, indicating that the two sub-time series units are negatively correlated; -1≤COR(X _T ,Y _T )≤1.

Then, the calculated similarity coefficient is determined as the periodic feature value.

S433: In the case that the periodic feature value is less than the first preset threshold, determine that the time series data belongs to volatile data;

When it is determined that the time series data belongs to volatile data, anomaly detection is performed on the time series data to obtain the change trend of the monitoring indicators corresponding to the target type data, so as to obtain the monitoring indicators corresponding to the target system. changing trend.

In some embodiments, when it is determined that the time series data belongs to volatile data, dynamic threshold detection is performed on the time series data by using Turkey,s Test or 3-Sigema algorithm, so as to obtain the corresponding data of the target type. The change trend of the monitoring index is obtained to obtain the change trend of the monitoring index corresponding to the target system.

S434: In the case that the periodic characteristic value is greater than or equal to the first preset threshold, further perform stability detection on the time series data to determine the stability characteristic value;

In some embodiments, optionally, the further performing stability detection on the time series data includes:

Seasonally decompose the time series data by using the moving average algorithm, and decompose it into a seasonal cycle component, a long-term trend component and a random residual component;

The stability characteristic value is determined according to the variance of the random residual component.

S435: In the case that the stability characteristic value is greater than the second preset threshold, determine that the time series data belongs to stable data;

When it is determined that the time series data is stable data, anomaly detection is performed on the time series data to obtain the change trend of the monitoring index corresponding to the target type data, so as to obtain the monitoring index corresponding to the target system changing trend.

In some embodiments, when it is determined that the time series data belongs to stable data, the SARIMAX or moving average algorithm is used to perform baseline threshold detection on the time series data, so as to obtain the monitoring indicators corresponding to the target type data. The change trend is to obtain the change trend of the monitoring index corresponding to the target system.

S436: In the case that the stability characteristic value is less than or equal to the second preset threshold, determine that the time series data belongs to periodic data;

When it is determined that the time series data belongs to periodic data, anomaly detection is performed on the time series data to obtain the change trend of the monitoring indicators corresponding to the target type data, so as to obtain the monitoring indicators corresponding to the target system. changing trend.

In some embodiments, when it is determined that the time series data belongs to periodic data, a machine learning classification model and a regression model are used to perform anomaly detection on the time series to obtain monitoring indicators corresponding to the target type data. The change trend of the target system can be obtained to obtain the change trend of the monitoring index corresponding to the target system. Specifically include:

Preprocessing the time series data; wherein, the time series data includes historical time series data and current time series data;

Obtain a first abnormality detection result corresponding to the current time series data; wherein, the first abnormality detection result is to extract various types of features from the preprocessed time series data, and input the various types of features into into a pre-built machine learning classification model;

Obtain a second abnormality detection result corresponding to the current time series data; wherein, the second abnormality detection result is obtained by comparing the current time series data with a prediction baseline, and the prediction baseline is a preprocessed The historical time series data obtained by inputting the pre-built regression model;

Vote on the first anomaly detection result and the second anomaly detection result to obtain the anomaly detection result of the time series data, that is, obtain the change trend of the monitoring index corresponding to the target type data, and obtain the target The change trend of the monitoring indicators corresponding to the system.

In this embodiment, on the basis of the above-mentioned embodiment, when it is determined that the time-series data belongs to volatile data, the Turkey,s Test or 3-Sigema algorithm is used to perform dynamic threshold detection on the time-series data, so as to obtain the The abnormality detection result of the time series data; when it is determined that the time series data is stable data, use SARIMAX or moving average algorithm to perform baseline threshold detection on the time series data to obtain the abnormality of the time series data. Detection result; in the case of determining that the time series data belongs to periodic data, perform anomaly detection on the time series data based on a machine learning classification model and a regression model. This method improves the flexibility and accuracy of anomaly detection of time series data by performing periodic detection and stability detection on time series data. sex, easy to operate.

Step S44: Perform system risk assessment on the target system based on the early warning information and the change trend.

In the embodiment of the present application, the fault diagnosis and safety operation and maintenance service system of charging station is generally not perfect in the operation and maintenance of charging facilities, and there is a lack of intelligent fault diagnosis and safety warning for the charging process, and potential equipment defects cannot be discovered in advance. To solve the problem of hidden dangers and accidents, improve the abnormal detection of the operation and maintenance detection data during the operation and maintenance of the charging equipment. The characteristic value is compared with the first preset threshold to determine whether it is volatile data; after it is determined to be non-volatile data, further stability detection is performed, the stability characteristic value is compared with the second preset threshold, and the stable data is distinguished from the second preset threshold. Periodic data. According to the category of the time series data, the abnormal detection of the operation and maintenance detection data of the charging equipment is performed on the periodic data, the volatile data, and the stable data according to different abnormal detection methods, so as to obtain the abnormal detection result of the time series data. . The method performs periodic detection and stability detection on the time series data in the operation and maintenance detection process of the charging equipment. After distinguishing the categories of the time series data, an appropriate anomaly detection method is selected according to the categories of the different data to perform anomaly detection. The flexibility and accuracy of time-series data anomaly detection in the process of charging equipment operation and maintenance detection, and the operation is simple. The charging equipment operation and maintenance detection method comprehensively considers charging equipment alarms, equipment abnormality detection, and equipment health status evaluation, and can generate an operation and maintenance inspection plan on demand. The charging equipment operation and maintenance detection method is not only suitable for the operation and maintenance of charging equipment, power grid operation and maintenance, energy storage equipment operation and maintenance, and/or the operation and maintenance of vehicle pile network integration, etc., but also the operation and maintenance of other detection systems based on the Internet of Things. The detection can be applied to the operation and maintenance detection method of the charging equipment, and the versatility is strong.

Meanwhile, for the method embodiments, for the sake of simple description, they are all expressed as a series of action combinations, but those skilled in the art should know that the embodiments of the present application are not limited by the described action sequence, because according to the present application In some embodiments, certain steps may be performed in other orders or concurrently. Secondly, those skilled in the art should also know that the embodiments described in the specification are all preferred embodiments, and the actions involved are not necessarily required by the embodiments of the present application.

Correspondingly, the embodiment of the present application also discloses a system risk assessment device, which is applied to the system risk assessment under the micro-service architecture as shown in FIG. 9 , and the device includes:

The data acquisition module 21 is used to acquire the system change data of the target system within the target time; the system change data includes the system early warning information and the name of the change device;

an early warning information determination module 22, configured to determine early warning information corresponding to the target system change by using natural language processing technology based on the system change data;

The change trend determination module 23 is used to obtain the time series data of the target system within the target time, and based on the time series data, use a machine learning algorithm to obtain the change trend of the monitoring index corresponding to the target system;

The risk assessment module 24 is configured to perform system risk assessment on the target system based on the early warning information and the change trend.

In this application, by acquiring the system change data of the target system within the target time, the system change data includes the system early warning information and the device name corresponding to the system change; and then based on the system change data, natural language processing technology is used to determine the The target system changes the corresponding early warning information; and obtains the time series data of the target system within the target time, and then uses the machine learning algorithm to obtain the monitoring indicators corresponding to the target system based on the time series data. change trend; finally, perform a system risk assessment on the target system based on the early warning information and the change trend. It can be seen that the natural language processing technology is used to determine the early warning information corresponding to the system change through the system early warning information within the time range of the system change and the names of the devices affected by the system change within the spatial range, and then combined with the machine learning algorithm to pass the time The change trend of the monitoring indicators corresponding to the system determined by the sequence data, the current risk of the system can be assessed, and the risks brought by software patch release, configuration change, and basic resource change can be quickly and accurately identified. Accuracy of change risk assessments.

In some specific embodiments, the early warning information determination module 22 may specifically include:

A first group of keyword determination unit, used for corpus preprocessing on the system early warning information, and extracting the first group of keywords from the processed system early warning information;

The second group of keyword determination unit is used for corpus preprocessing on the device name, and extracts the second group of keywords from the processed changed device name;

A similarity determination unit, configured to calculate the similarity between the first group of keywords and the keywords in the second group of keywords, and determine early warning information corresponding to the target system change based on the similarity.

In some specific embodiments, the change trend determination module 23 may specifically include:

a classification unit, configured to classify the time series data according to a preset classification standard to obtain corresponding target type data; the target type data includes volatile data, periodic data and stable data;

A trend determination unit, configured to perform anomaly detection on the target type data by using a preset detection algorithm corresponding to the target type data, and obtain the change trend of the monitoring index corresponding to the target type data, so as to obtain the target system The change trend of the corresponding monitoring indicators;

The first trend determination subunit is used to perform anomaly detection on the variable data based on Turkey detection, and obtain the change trend of the variable monitoring index;

The second trend determination subunit is configured to perform abnormality detection on the periodic data based on the same chain ratio algorithm, and obtain the change trend of the periodic monitoring index;

A third trend determination subunit, configured to perform anomaly detection on the stable data based on the time-series ARIMA algorithm, to obtain the change trend of the stable monitoring indicators;

a first classification subunit, configured to perform periodic detection on the time series data based on the similarity of window data, and classify according to a first preset threshold to obtain the volatile data and the non-volatile data;

The second classification subunit is configured to perform stability detection on the non-volatile data based on the STL algorithm, and classify the periodic data and the stable data according to a second preset threshold.

In some specific embodiments, the risk assessment module 24 may specifically include:

a risk level determination unit, configured to perform a system risk assessment on the target system based on the early warning information, the change trend and the index level of the monitoring index, and obtain the risk level of the target system; wherein, the index level Including core business-level indicators, technical-level indicators and system resource-level indicators;

An assessment report generating unit, configured to generate a corresponding assessment report based on the risk level.

Further, the embodiment of the present application also discloses an electronic device, as shown in FIG. 10 , the content in the figure should not be considered as any limitation on the scope of use of the present application.

FIG. 10 is a schematic structural diagram of an electronic device 30 according to an embodiment of the present application. The electronic device 30 may specifically include: at least one processor 31 , at least one memory 32 , a power supply 33 , a communication interface 34 , an input and output interface 35 and a communication bus 36 . Wherein, the memory 32 is used for storing a computer program, and the computer program is loaded and executed by the processor 31 to implement the relevant steps in the system risk assessment method disclosed in any of the foregoing embodiments.

In this embodiment, the power supply 33 is used to provide working voltage for each hardware device on the electronic device 30; the communication interface 34 can create a data transmission channel between the electronic device 30 and external devices, and the communication protocol it follows is applicable Any communication protocol in the technical solution of the present application is not specifically limited here; the input and output interface 35 is used to obtain external input data or output data to the outside world, and its specific interface type can be selected according to specific application needs, here No specific limitation is made.

In addition, the memory 32, as a carrier for resource storage, can be a read-only memory, a random access memory, a magnetic disk or an optical disk, etc., and the resources stored on it include an operating system 321, a computer program 322, and data 323 including system change data, etc., The storage method can be short-term storage or permanent storage.

The operating system 321 is used to manage and control each hardware device and computer program 322 on the electronic device 30, so as to realize the operation and processing of the massive data 323 in the memory 32 by the processor 31, which can be Windows Server, Netware, Unix, Linux etc. In addition to the computer program that can be used to complete the system risk assessment method performed by the electronic device 30 disclosed in any of the foregoing embodiments, the computer program 322 may further include a computer program that can be used to complete other specific tasks. The data 323 may include system modification data acquired by the electronic device 30 .

Further, an embodiment of the present application further discloses a computer storage medium, where computer-executable instructions are stored in the computer storage medium, and when the computer-executable instructions are loaded and executed by a processor, the disclosure in any of the foregoing embodiments is realized. The systematic risk assessment method steps.

The various embodiments in this specification are described in a progressive manner, and each embodiment focuses on the differences from other embodiments, and the same or similar parts between the various embodiments may be referred to each other. As for the device disclosed in the embodiment, since it corresponds to the method disclosed in the embodiment, the description is relatively simple, and the relevant part can be referred to the description of the method.

The steps of a method or algorithm described in conjunction with the embodiments disclosed herein may be directly implemented in hardware, a software module executed by a processor, or a combination of the two. A software module can be placed in random access memory (RAM), internal memory, read only memory (ROM), electrically programmable ROM, electrically erasable programmable ROM, registers, hard disk, removable disk, CD-ROM, or any other in the technical field. in any other known form of storage medium.

Finally, it should also be noted that in this document, relational terms such as first and second are used only to distinguish one entity or operation from another, and do not necessarily require or imply these entities or that there is any such actual relationship or sequence between operations. Moreover, the terms "comprising", "comprising" or any other variation thereof are intended to encompass a non-exclusive inclusion such that a process, method, article or device that includes a list of elements includes not only those elements, but also includes not explicitly listed or other elements inherent to such a process, method, article or apparatus. Without further limitation, an element qualified by the phrase "comprising a..." does not preclude the presence of additional identical elements in a process, method, article or apparatus that includes the element.

A system risk assessment method, device, equipment and medium provided by the present invention have been introduced in detail above. Specific examples are used in this paper to illustrate the principles and implementations of the present invention. The descriptions of the above examples are only used to help Understand the method of the present invention and its core idea; at the same time, for those skilled in the art, according to the idea of the present invention, there will be changes in the specific implementation and application scope. In summary, the content of this specification does not It should be understood as a limitation of the present invention.

Claims (12)

1. A system risk assessment method is applied to system risk assessment under a micro-service architecture, and is characterized by comprising the following steps:

acquiring system change data of a target system in target time; the system change data comprises system early warning information and a device name corresponding to the system change;

based on the system change data, determining early warning information corresponding to the target system change by using a natural language processing technology;

acquiring time sequence data of the target system in the target time, and obtaining a variation trend of a monitoring index corresponding to the target system by using a machine learning algorithm based on the time sequence data;

and performing system risk assessment on the target system based on the early warning information and the change trend.

2. The system risk assessment method according to claim 1, wherein the determining early warning information corresponding to the target system change based on the system change data by using a natural language processing technique comprises:

performing corpus preprocessing on the system early warning information, and extracting a first group of keywords from the processed system early warning information;

performing corpus preprocessing on the device name, and extracting a second group of keywords from the processed changed device name;

and calculating the similarity of the keywords in the first group of keywords and the second group of keywords, and determining early warning information corresponding to the target system change based on the similarity.

3. The system risk assessment method according to claim 1, wherein the target time comprises a system change time of the target system and a time within a preset time period after the change.

4. The system risk assessment method according to claim 1, wherein the obtaining of the variation trend of the monitoring index corresponding to the target system by using a machine learning algorithm based on the time series data comprises:

classifying the time sequence data according to a preset classification standard to obtain corresponding target type data; the target type data comprises volatile data, periodic data and stable data;

and carrying out abnormal detection on the target type data by using a preset detection algorithm corresponding to the target type data to obtain the change trend of the monitoring index corresponding to the target type data so as to obtain the change trend of the monitoring index corresponding to the target system.

5. The system risk assessment method according to claim 4, wherein the performing anomaly detection on the target type data by using a preset detection algorithm corresponding to the target type data to obtain a variation trend of a monitoring index corresponding to the target type data comprises:

performing abnormity detection on the easy-to-deform data based on Turkey detection to obtain the variation trend of the easy-to-deform monitoring index;

performing anomaly detection on the periodic data based on a homocyclic ratio algorithm to obtain the variation trend of the periodic monitoring index;

and carrying out abnormal detection on the stable data based on a time sequence ARIMA algorithm to obtain the change trend of the stable monitoring index.

6. The system risk assessment method according to claim 4, wherein the classifying the time-series data according to a preset classification standard to obtain corresponding target type data comprises:

carrying out periodic detection on the time sequence data based on window data similarity, and classifying according to a first preset threshold value to obtain the volatile data and the non-volatile data;

and carrying out stability detection on the non-volatile data based on an STL algorithm, and classifying according to a second preset threshold value to obtain the periodic data and the stable data.

7. The system risk assessment method according to any one of claims 1 to 6, wherein the performing the system risk assessment on the target system based on the early warning information and the trend of change comprises:

performing system risk assessment on the target system based on the early warning information, the change trend and the index grade of the monitoring index to obtain the risk grade of the target system;

generating a corresponding assessment report based on the risk level;

the index grade comprises a core service grade index, a technical grade index and a system resource grade index.

8. A charging device operation and maintenance detection method is characterized by comprising the following steps:

acquiring system change data of a target system in target time; the target system comprises a charging device system; the system change data comprises the operation and maintenance data of the charging equipment, and the operation and maintenance data of the charging equipment comprises system early warning information and a device name corresponding to system change;

based on the system change data, determining early warning information corresponding to the target system change by using a natural language processing technology;

acquiring time sequence data of the target system in the target time, and obtaining a variation trend of a monitoring index corresponding to the target system by using a machine learning algorithm based on the time sequence data;

and performing system risk assessment on the target system based on the early warning information and the change trend.

9. The method according to claim 8, wherein the obtaining of time-series data of the target system in the target time and obtaining a variation trend of a corresponding monitoring index of the target system based on the time-series data by using a machine learning algorithm comprises:

acquiring time sequence data in the target time in the operation and maintenance detection process of the charging equipment;

determining a periodic characteristic value by periodically detecting the time sequence data;

under the condition that the periodic characteristic value is smaller than a first preset threshold value, judging that the time sequence data belong to the volatile data;

under the condition that the periodic characteristic value is greater than or equal to the first preset threshold value, further performing stability detection on the time series data to determine a stability characteristic value;

under the condition that the stability characteristic value is larger than a second preset threshold value, judging that the time sequence data belong to stable data;

when the stability characteristic value is less than or equal to the second preset threshold value, judging that the time series data belong to periodic data;

according to the target type data to which the time sequence data belong, carrying out abnormity detection on the time sequence data to obtain a variation trend of a monitoring index corresponding to the target type data so as to obtain a variation trend of a monitoring index corresponding to the target system;

the target type data includes volatile data, periodic data, and stable data.

10. The method of claim 9,

the passing the time-series data through a periodic detection includes:

a reference period T given to the time series data;

dividing the time series data into n/T sub-time series units by taking the reference period T as a dividing point, wherein n is the length of the time series data;

comparing every two of each sub-time sequence unit to calculate a similarity coefficient, and determining the similarity system as the periodic characteristic value;

the further performing stability detection on the time-series data comprises:

carrying out seasonal decomposition on the time sequence data by adopting a moving average algorithm to obtain a seasonal periodic component, a long-term trend component and a random residual error component;

and determining the stability characteristic value according to the variance of the random residual error component.

11. The method according to claim 9 or 10, wherein the detecting the abnormality of the time-series data according to the target type data to which the time-series data belongs comprises:

under the condition that the time sequence data belong to the volatile data, performing dynamic threshold detection on the time sequence data by adopting a Turkey, s Test or 3-Sigema algorithm to obtain the variation trend of the monitoring index corresponding to the target system;

under the condition that the time sequence data are judged to be stable data, performing baseline threshold detection on the time sequence data by adopting SARIMAX or a moving average algorithm to obtain a variation trend of a monitoring index corresponding to the target system;

and under the condition that the time sequence data belong to periodic data, performing abnormity detection on the time sequence by adopting a machine learning classification model and a regression model to obtain the variation trend of the monitoring index corresponding to the target system.

12. A system risk assessment device is applied to system risk assessment under a micro-service architecture, and is characterized by comprising:

the data acquisition module is used for acquiring system change data of the target system in target time; the system change data comprises system early warning information and a change device name;

the early warning information determining module is used for determining early warning information corresponding to the target system change by utilizing a natural language processing technology based on the system change data;

the variation trend determining module is used for acquiring time sequence data of the target system in the target time and obtaining the variation trend of the monitoring index corresponding to the target system by utilizing a machine learning algorithm based on the time sequence data;

and the risk evaluation module is used for carrying out system risk evaluation on the target system based on the early warning information and the change trend.

Images