CN117422200A - Intelligent monitoring and early warning method and system for power plant - Google Patents

Abstract

The invention discloses an intelligent monitoring and early warning method and system for a power plant, which belong to the field of power systems, wherein the method comprises the following steps: collecting the concentration of particulate matters and the concentration of carbon dioxide in a power plant; performing compensation analysis on the plurality of particle concentrations to obtain a plurality of compensation particle concentrations; obtaining a particulate matter concentration characteristic field and a carbon dioxide characteristic field according to the plurality of compensated particulate matter concentrations, the plurality of carbon dioxide concentrations and the plurality of position coordinates; acquiring an operation abnormality analysis result of the power plant according to the characteristic field; constructing a power generation emission data matrix according to the concentration field, and acquiring a plurality of emission scores of a plurality of positions; and generating an operation monitoring result of the target power plant according to the operation abnormality analysis result and the emission scores, and carrying out early warning according to the operation monitoring result based on the early warning station. The technical problems of low monitoring and early warning precision and poor effect of the operation of the power plant in the prior art are solved, and the technical effects of improving the monitoring and early warning precision and effect of the operation of the power plant are achieved.

Description

An intelligent monitoring and early warning method and system for power plants

Technical field

The invention relates to the field of power systems, and in particular to an intelligent monitoring and early warning method and system for a power plant.

Background technique

With the development of society, the demand for electricity continues to grow, and the proportion of thermal power generation continues to increase. During the thermal power generation process, a large amount of emissions are produced, which reflects the operating status of the thermal power plant. Therefore, monitoring and early warning of emissions from thermal power plants is crucial. In the existing technology, various types of monitoring equipment are used to monitor power plant emissions. However, these equipment are often used in isolation and scattered, making it difficult to effectively monitor the overall operation of the power plant and to accurately identify the location of abnormal power plant operations. In addition, the early warning function of existing monitoring equipment is relatively simple, and the early warning effect on power plant emergencies is still limited. Therefore, the accuracy of monitoring and early warning of power plant operations in the existing technology is low, the early warning effect is poor, and it cannot meet the needs of safe production of power plants.

Contents of the invention

This application provides an intelligent monitoring and early warning method and system for a power plant, aiming to solve the technical problems of low accuracy and poor effectiveness in monitoring and early warning of power plant operations in the existing technology.

In view of the above problems, this application provides an intelligent monitoring and early warning method and system for a power plant.

The first aspect disclosed in this application provides an intelligent monitoring and early warning method for a power plant. The method includes: collecting particulate matter concentration and carbon dioxide at multiple locations through testing devices arranged at multiple locations in the target power plant in the emission data collection station. Concentration, the target power plant is a thermal power plant; collect water vapor concentrations at multiple locations through multiple test devices, perform compensation analysis on multiple particulate matter concentrations based on multiple water vapor concentrations, and obtain multiple compensated particulate matter concentrations; based on multiple compensated particulate matter concentrations concentration, multiple carbon dioxide concentrations and the coordinates of multiple locations, construct the particle concentration field and carbon dioxide concentration field respectively, and perform processing and discrimination to obtain the particle concentration characteristic field and carbon dioxide characteristic field; through the power plant operation analysis station, according to the particle concentration characteristic field and carbon dioxide characteristic fields to identify and obtain the operation abnormality analysis results of the target power plant. The operation abnormality analysis results include whether an abnormality occurs and the abnormal location when an abnormality occurs; through the emission analysis station, power generation emission data is constructed based on the particulate matter concentration field and carbon dioxide concentration field. Matrix, and calculate multiple emission scores at multiple locations based on the power generation emission data matrix; generate operation monitoring results of the target power plant based on the operation abnormality analysis results and multiple emission scores, and conduct early warning based on the operation monitoring results based on the early warning station. Among them, the job monitoring results include event description information of job abnormal events.

Another aspect disclosed in this application provides an intelligent monitoring and early warning system for a power plant. The system includes: a power plant data acquisition module, which is used to collect multiple Particulate matter concentration and carbon dioxide concentration at the location, the target power plant is a thermal power plant; the concentration compensation analysis module is used to collect water vapor concentrations at multiple locations through multiple test devices, and perform compensation analysis on multiple particulate matter concentrations based on multiple water vapor concentrations. , obtain multiple compensated particulate matter concentrations; the data processing and discrimination module is used to construct a particulate matter concentration field and a carbon dioxide concentration field respectively based on multiple compensated particulate matter concentrations, multiple carbon dioxide concentrations, and coordinates of multiple locations, and perform processing and discrimination to obtain particulate matter Concentration characteristic field and carbon dioxide characteristic field; the operation abnormality analysis module is used to identify and obtain the operation abnormality analysis results of the target power plant based on the particulate matter concentration characteristic field and carbon dioxide characteristic field through the power plant operation analysis station. The operation abnormality analysis results include whether an abnormality occurs and the abnormal location when an abnormality occurs; the location emission scoring module is used to construct a power generation emission data matrix based on the particulate matter concentration field and the carbon dioxide concentration field through the emission analysis station, and calculate and obtain multiple locations at multiple locations based on the power generation emission data matrix. Emission scoring; monitoring results early warning module is used to generate operation monitoring results of the target power plant based on operation abnormality analysis results and multiple emission scores. Based on the early warning station, early warning is carried out based on the operation monitoring results. Among them, the operation monitoring results include operation abnormalities. Event description information for the event.

One or more technical solutions provided in this application have at least the following technical effects or advantages:

Due to the use of testing devices deployed at multiple points in the target power plant, multi-source data are collected to achieve comprehensive monitoring of the overall operation of the power plant; particulate matter concentration, carbon dioxide concentration and water vapor concentration data are collected, and through compensation analysis of the particulate matter concentration, it is possible to Improve the accuracy of data collection; use the collected data to construct particulate matter concentration fields and carbon dioxide concentration fields, and process and identify to obtain characteristic fields to accurately locate abnormal conditions in power plant operations; identify abnormal operation conditions in power plants based on characteristic field analysis, and improve the understanding of power plants Monitoring accuracy for abnormal operations; calculate emission scores at multiple locations to intuitively reflect the emission status of different parts of the power plant; conduct early warnings based on monitoring results, and generate operational monitoring results to promptly warn power plant emergencies, improve monitoring effects, and solve technical solutions It solves the technical problems of low accuracy and poor effect of power plant operation monitoring and early warning in the existing technology, and achieves the technical effect of improving the accuracy and effect of power plant operation monitoring and early warning.

The above description is only an overview of the technical solutions of the present application. In order to have a clearer understanding of the technical means of the present application, they can be implemented according to the content of the description, and in order to make the above and other purposes, features and advantages of the present application more obvious and understandable. , the specific implementation methods of the present application are specifically listed below.

Description of the drawings

Figure 1 is a schematic flow chart of an intelligent monitoring and early warning method for a power plant provided by an embodiment of the present application;

Figure 2 is a schematic flowchart of generating operation monitoring results in an intelligent monitoring and early warning method for power plants provided by an embodiment of the present application;

Figure 3 is a schematic structural diagram of an intelligent monitoring and early warning system for a power plant provided by an embodiment of the present application.

Explanation of reference signs: power plant data acquisition module 11, concentration compensation analysis module 12, data processing and identification module 13, operation abnormality analysis module 14, location emission scoring module 15, monitoring result early warning module 16.

Implementation

The overall idea of the technical solution provided by this application is as follows:

The embodiment of the present application provides an intelligent monitoring and early warning method and system for a power plant. Multi-source data is collected by arranging test devices at multiple points in the target power plant. The collected data includes particulate matter concentration, carbon dioxide concentration and water vapor concentration. In order to improve the accuracy of data collection, compensation analysis is performed on the concentration of particulate matter. Then, the collected data are used to construct the particle concentration field and the carbon dioxide concentration field respectively, and the respective characteristic fields are obtained through processing and discrimination to accurately locate the abnormal operation of the power plant, thereby improving the monitoring accuracy of the abnormal operation of the power plant. At the same time, emission scores for multiple locations are calculated to reflect the emission status of different parts of the power plant. Finally, based on the monitoring results, early warning is given to the power plant and operation monitoring results are generated to achieve intelligent monitoring and early warning of the power plant operation.

After introducing the basic principles of the present application, various non-limiting implementations of the present application will be specifically introduced below in conjunction with the accompanying drawings.

Example

As shown in Figure 1, the embodiment of the present application provides a power plant intelligent monitoring and early warning method, which is applied to a power plant intelligent monitoring and early warning equipment. The equipment includes an emission data collection station, a power plant operation analysis station, and an emission analysis station. and early warning stations.

Specifically, the embodiment of the present application is an intelligent monitoring and early warning method for a power plant, which realizes intelligent monitoring and early warning of the power plant's operation process by collecting and analyzing the emission data of the power plant. The method in the embodiment of this application is applied to a power plant's intelligent monitoring and early warning equipment. The equipment includes an emission data collection station, a power plant operation analysis station, an emission analysis station and an early warning station.

Among them, the emission data collection station is a device used to collect emission monitoring data at multiple locations in the power plant, including test devices set up at different locations in the power plant to collect emission data such as particulate matter concentration and carbon dioxide concentration; the power plant operation analysis station is Based on the collected emission data, it is a device that analyzes and judges the operating conditions of the power plant, and identifies abnormal conditions during the operation of the power plant; the emission analysis station is based on the data of the emission concentration field to score and evaluate the emission conditions at different locations in the power plant. The analysis device evaluates the emission quality level of the power plant; the early warning station is a device that provides early warning for abnormal situations that occur during the operation of the power plant based on the monitoring results of the power plant operation, and issues early warning information to avoid or reduce the impact of abnormal operations on the power plant. The emission data collection station, power plant operation analysis station, emission analysis station and early warning station together form the power plant's intelligent monitoring and early warning equipment, which realizes comprehensive monitoring, intelligent analysis and early warning of the power plant's emission data, effectively improving the safe operation of the power plant. level.

Monitoring and early warning methods include:

Through the testing devices deployed at multiple locations in the target power plant in the emission data collection station, particulate matter concentrations and carbon dioxide concentrations at multiple locations are collected. The target power plant is a thermal power plant;

In the embodiment of the present application, the emission data collection station includes multiple test devices, and these test devices are arranged at multiple locations in the target thermal power plant to collect emission data at different locations, for example. Among them, the test device uses a light scattering sensor to monitor and record the concentration of particulate matter at the target location in real time; at the same time, a non-dispersive infrared gas analyzer is set up to monitor and record the carbon dioxide concentration at the target location.

Among them, multiple test devices are deployed in different locations of the thermal power plant, such as the area around the chimney, the entrance and exit of the power plant, etc., to comprehensively monitor the emissions of the thermal power plant. These multiple test devices are connected to the emission data collection station through data lines, and send the detected particulate matter concentration and carbon dioxide concentration data to the emission data collection station in real time to provide basic data for subsequent emission analysis.

By obtaining the particulate matter and carbon dioxide concentrations at multiple locations in the thermal power plant, it can comprehensively reflect the emission levels in different areas of the thermal power plant, providing a basis for concentration field analysis and anomaly identification.

Through multiple testing devices, water vapor concentrations at multiple locations are collected, and multiple particulate matter concentrations are compensated and analyzed based on multiple water vapor concentrations to obtain multiple compensated particulate matter concentrations;

Further steps include:

Obtain sample particle concentration information records based on test data records tested by light scattering measurement methods;

Obtain sample compensation particle concentration information records based on test data records measured by sampling and weighing, and obtain sample water vapor concentration records based on water vapor concentration monitoring records;

According to the sample particulate matter concentration information record and the sample compensation particulate matter concentration information record, calculate and obtain the sample compensation coefficient record;

Sampling sample water vapor concentration record and sample compensation coefficient record, training to obtain compensation correction analysis channel;

Obtain multiple compensation coefficients and perform compensation and correction on multiple particulate matter concentration information. Multiple compensation coefficients are obtained by inputting multiple water vapor concentrations into the compensation correction analysis channel for processing.

In a preferred embodiment, in order to improve the accuracy of the particle concentration and compensate for the influence of water vapor, a compensation correction analysis channel needs to be established. First, a large number of space particle matter samples are detected by using light scattering sensors, and the resulting particle concentration data is recorded as a sample particle concentration information record, reflecting the particle concentration level in the sample that has not been compensated for the influence of water vapor, and is used for standard sampling weighing. It lays the foundation for comparison and calibration with the accurate concentration obtained by the method. Secondly, for the same batch of samples, in addition to light scattering measurements, a sampler is used to collect spatial particle samples corresponding to the sample particle concentration. The particles are weighed using a precision electronic balance to calculate and obtain the particle mass concentration in the sample volume, which is recorded as a sample. Compensated particle concentration information recording. At the same time, for the same batch of samples, the temperature and humidity sensor is used to measure and record the water vapor concentration monitoring record corresponding to the sample as the sample water vapor concentration record. Then, by comparing and calculating the sample particulate matter concentration information record and the sample compensated particulate matter concentration information record, the proportional relationship between the original particulate matter concentration and the compensated particulate matter concentration under the same sample conditions is determined. This set of proportional relationships is the sample compensation coefficient record. Then, a machine learning algorithm is used, using the sample water vapor concentration record as input and the sample compensation coefficient record as output, to train and establish a nonlinear mapping model between water vapor concentration and compensation coefficient, that is, the compensation correction analysis channel.

In addition to detecting particulate matter concentration and carbon dioxide concentration, the test device is also equipped with temperature and humidity sensors to simultaneously collect water vapor concentration data from multiple locations. After obtaining the water vapor concentration data at multiple locations, input the multiple water vapor concentrations into the compensation correction analysis channel to obtain the corresponding multiple compensation coefficients, and then use multiple compensation coefficients to multiply corresponding to multiple particle concentrations to achieve concentration Compensation eliminates the impact of water vapor on the particle concentration test results, thereby obtaining a more accurate compensated particle concentration, that is, multiple compensated particle concentrations.

Through the compensation of water vapor concentration, the monitoring accuracy of particulate matter emissions from thermal power plants is improved, and more accurate particulate matter concentration data are provided for subsequent concentration field analysis to improve early warning accuracy.

According to the coordinates of multiple compensated particulate matter concentrations, multiple carbon dioxide concentrations and multiple locations, the particulate matter concentration field and the carbon dioxide concentration field are constructed respectively, and processed and distinguished to obtain the particulate matter concentration characteristic field and the carbon dioxide characteristic field;

Further, this step specifically includes:

Obtain the coordinate information of multiple locations, combine the multiple coordinate information with multiple carbon dioxide concentrations and multiple compensated particle concentrations to construct a particle concentration field and a carbon dioxide concentration field;

Use local processing operators to divide the particle concentration field and carbon dioxide concentration field to obtain multiple first local areas and multiple second local areas;

In each first local area, the particle concentration at the center position is used as the threshold, and the particle concentration at other locations is judged to be greater than, less than, or within the error range of the threshold, and marked as 1, -1, and 0 respectively to generate multiple local areas. Vector to obtain the particle concentration characteristic field;

Discriminate each second local area to obtain the carbon dioxide characteristic field.

In a preferred embodiment, a particulate matter concentration field and a carbon dioxide concentration field are constructed based on the collected compensated particulate matter concentration, carbon dioxide concentration and corresponding position coordinates at each location, and the corresponding feature fields are extracted through processing.

First, conduct a 3D model of the target power plant, generate a coordinate map of the target power plant, mark the location of the test device on the map, and map the multiple position coordinates of the test device with its corresponding compensated particulate matter concentration and carbon dioxide concentration. And use the spatial interpolation algorithm to generate continuous distribution scenes of particulate matter concentration and carbon dioxide concentration within the coordinate map of the target power plant, that is, particulate matter concentration field and carbon dioxide concentration field. Among them, the particulate matter concentration field represents the distribution and change of particulate matter concentration in the target power plant; the carbon dioxide concentration field represents the distribution of carbon dioxide concentration. Secondly, a local processing operator is used to define a local window on the coordinate map, and based on this window, the particle concentration field and the carbon dioxide concentration field are divided into local areas. The particle concentration field is divided into multiple first local areas, and the carbon dioxide concentration field is divided into multiple first local areas. The concentration field is divided to obtain multiple second local areas. These local areas reflect the characteristic information of the concentration field in the local area.

Then, for each first local area, the particle concentration value at the center of the local area is used as the threshold, and the relationship between the particle concentration at other locations in the local area and the threshold is determined based on the error range that is greater than, less than, or equal to the threshold. distinguish. If the particle concentration at a certain location is higher than the positive error range of the central location concentration, it is marked as 1; if it is lower than the negative error range of the central location concentration, it is marked as -1; if it is within the positive and negative error range of the central location concentration, then Marked as 0. This process is performed on each first local area in turn, and the characteristics of the relationship between the particle concentration in the first local area and the particle concentration in the central position are extracted to form a local vector for each local area. These multiple local vectors constitute the particulate matter. The characteristic expression of the concentration field, that is, the particle concentration characteristic field. At the same time, for each second local area, the average carbon dioxide concentration in the local area is calculated, and the average carbon dioxide concentration is compared with the carbon dioxide concentration at the center of the local area. If the average carbon dioxide concentration is higher than the central concentration, mark the local area as 1; if it is lower than the central concentration, mark it as -1; if it is close to the central concentration, mark it as 0. The degree of proximity can be set based on the error range for carbon dioxide concentration testing by those skilled in the art. Repeat the above process for all second local areas, extract the carbon dioxide concentration distribution characteristics in each local area, and form multiple local vectors. These local vectors constitute the characteristic expression of the carbon dioxide concentration field, that is, the carbon dioxide characteristic field.

Through the power plant operation analysis station, based on the particulate matter concentration characteristic field and the carbon dioxide characteristic field, the operation abnormality analysis results of the target power plant are identified and obtained. The operation abnormality analysis results include whether an abnormality occurs and the abnormal location when the abnormality occurs;

Further, this step specifically includes:

Based on the historical emission monitoring data of the target power plant, process and obtain multiple sample particulate matter concentration characteristic fields and multiple sample carbon dioxide characteristic fields, and obtain multiple sample operation abnormality analysis results through the operation abnormality monitoring data;

Multiple sample particulate matter concentration characteristic fields and multiple sample carbon dioxide characteristic fields are used as input training data, and multiple sample operation abnormality analysis results are used as output training data. Based on machine learning, in the power plant operation analysis station, the operation abnormality identification channel is trained and obtained ;

The operation abnormality analysis results are obtained by inputting the particulate matter concentration characteristic field and the carbon dioxide characteristic field into the operation abnormality identification channel for operation abnormality analysis.

In an optimal implementation, firstly, the emission monitoring data of the target power plant in the past period of time are collected, including the concentration monitoring results of the power plant in different periods; for these historical monitoring data, concentration field construction, feature extraction, and processing are performed Corresponding multiple sample particulate matter concentration characteristic fields and multiple sample carbon dioxide characteristic fields are obtained. At the same time, the operation abnormality analysis results corresponding to these historical samples are obtained, that is, the marking results of whether there are abnormal power plant operations during the period in which these historical data are located, as the operation abnormality analysis results of multiple samples. Then, the obtained sample particulate matter concentration characteristic field and sample carbon dioxide characteristic field are used as input data, and the corresponding sample operation abnormality analysis results are used as output data. In the power plant operation analysis station, supervised learning algorithms, such as SVM and random forest algorithms, are used Wait for model training. Through the training of a large number of historical samples, a model that can judge job anomalies in new scenarios is obtained, that is, the job anomaly identification channel.

Subsequently, after obtaining the particulate matter concentration characteristic field and carbon dioxide characteristic field of the target power plant, the power plant operation analysis station uses these two characteristic fields as new inputs and inputs them into the operation anomaly identification channel, which analyzes the input particulate matter concentration characteristic field and carbon dioxide characteristic field. Analyze and identify whether there are abnormal operation conditions on site, and output corresponding operation abnormality analysis results to provide the operation status of each location of the current target power plant, thereby determining which location has operation abnormality and realizing intelligent analysis of power plant emission data.

Through the emission analysis station, a power generation emission data matrix is constructed based on the particulate matter concentration field and carbon dioxide concentration field, and multiple emission scores for multiple locations are calculated and obtained based on the power generation emission data matrix;

Further, this step specifically includes:

Maximize the concentration data in the particulate matter concentration field and carbon dioxide concentration field, as follows:

;

Among them, y is the concentration data after maximization, and x is the original concentration data;

Based on the data in the maximized particulate matter concentration field and carbon dioxide concentration field, a power generation emission data matrix is constructed, as follows:

;

Among them, P is the power generation emission data matrix, is the maximized particle concentration data at the first position,/> is the maximized particle concentration data at the n-th location, n is the number of multiple locations,/> is the maximized carbon dioxide concentration data at the first position,/> is the maximized carbon dioxide concentration data at the nth position.

Further, this step also includes:

According to the power generation emission data matrix, multiple emission scores for multiple locations are calculated and obtained as follows:

;

in, Score the emissions at the i-th location,/> ,/> and/> is the weight,/> and/> is the data after maximizing the particulate matter concentration standard and carbon dioxide concentration standard of the target power plant,/> and/> is the maximum particulate matter concentration data and maximum carbon dioxide concentration data at the i-th position,/> is the data in column j and row i in the power generation emission data matrix,/> is the minimum value of the jth column in the power generation emission data matrix,/> The weight assigned based on the impact of particulate matter concentration and carbon dioxide concentration on the emission quality of the target power plant.

In a preferred embodiment, in the emission analysis station, a power generation emission data matrix is constructed based on the particulate matter concentration field and the carbon dioxide concentration field, and the emission score of each location of the target power plant is obtained based on the power generation emission data matrix, and the different power plants are quantitatively evaluated. The emission status of the area provides support for early warning of subsequent abnormal operations. First, for each position of the original concentration data of the concentration data in the particulate matter concentration field and carbon dioxide concentration field, according to the formula , perform maximization processing, where y is the concentration data after maximization, and x is the original concentration data. When the original concentration x is small, the calculated maximum concentration y is large; when the original concentration x is large, the maximum concentration y is small, which plays the role of compressing the dynamic range of the background concentration, and reducing the background concentration value has a negative impact on subsequent Computational impact. Secondly, the data in the particulate matter concentration field and carbon dioxide concentration field after maximization are processed according to the matrix/> The position correspondences are combined to construct a power generation emission data matrix. Each row of the matrix represents a position, the first column represents the maximum particulate matter concentration data, and the second column represents the maximum carbon dioxide concentration data. Among them,/> is the power generation emission data matrix,/> is the maximized particle concentration data at the first position,/> is the maximized particle concentration data at the n-th location, n is the number of multiple locations,/> is the maximized carbon dioxide concentration data at the first position,/> is the maximized carbon dioxide concentration data at the nth position. The data of particulate matter and carbon dioxide indicators at all locations in the target power plant are integrated in the form of a matrix to facilitate the subsequent calculation of emission scores based on the matrix.

Then, according to the constructed power generation emission data matrix, according to Calculate the emissions score for each monitoring location in the target power plant. Among them,/> Score the emissions at the i-th location,/> ,/> and/> is the weight,/> and/> is the data after maximizing the particulate matter concentration standard and carbon dioxide concentration standard of the target power plant,/> and/> is the maximum particulate matter concentration data and maximum carbon dioxide concentration data at the i-th position,/> is the data in column j and row i in the power generation emission data matrix,/> is the minimum value of the jth column in the power generation emission data matrix,/> The weight assigned based on the impact of particulate matter concentration and carbon dioxide concentration on the emission quality of the target power plant. Quantitatively score the emission status of each location through the emission score calculation method, obtain multiple emission scores for multiple locations, clarify the emission status of each location, and provide support for subsequent assessment and warning.

Based on the operation abnormality analysis results and multiple emission scores, the operation monitoring results of the target power plant are generated. Based on the early warning station, early warning is carried out based on the operation monitoring results. The operation monitoring results include event description information of the operation abnormal events.

Further, as shown in Figure 2, this step specifically includes:

According to the emission monitoring data of the target power plant, extract the data and calculate to obtain sample emission score records and sample emission quality grade records;

Use sample emission score records and sample emission quality grade records to construct an emission quality comparison table;

Match based on multiple emission scores to obtain multiple emission quality levels;

When an abnormality occurs in the operation abnormality analysis results or any emission quality level is lower than the qualified emission quality level, event description information is generated as the operation monitoring result.

In a feasible implementation, first, the emission monitoring data of the target power plant in the past time period are collected, including the emission monitoring results of different locations of the power plant under normal and abnormal operating conditions. These historical monitoring data are processed and calculated to obtain the emission score corresponding to each sample data, and form a sample emission score record. At the same time, the corresponding emission quality levels are determined based on the scoring results of these samples, and a sample emission quality level record is formed. For example, they are graded according to excellent, good, slightly excessive, etc. Then, these sample emission score records and emission quality grade records are used to establish a mapping relationship between emission scores and quality grades to form an emission quality comparison table. For example, multiple scoring intervals are preset based on expert experience, and then the quality levels corresponding to different intervals are determined, so that the emission quality level can be corresponding to the emission score. Subsequently, based on the currently acquired multiple emission scores of multiple locations, they are matched with the emission quality comparison table to obtain multiple emission quality levels corresponding to the multiple emission scores.

Then, risk warning is carried out based on operation abnormality analysis and multiple emission quality levels. When the operation abnormality analysis results show that there are operation abnormalities in the power plant area, or the emission quality level at any one of the multiple emission quality levels is lower than the qualified level, it is determined that there is an operation abnormality in the target power plant, and the operation monitoring results are generated, which include event description information , such as abnormal type, abnormal location, degree of exceedance, occurrence time and other details. When the operation monitoring results show that an abnormality is detected, the early warning station will display the operation monitoring results as early warning information, and at the same time issue an early warning notice to the power plant supervisors in the form of sound and light. The supervisors will check the event description details in the early warning information based on the early warning station. , and judge the degree of risk accordingly, and take countermeasures such as adjusting operating parameters and repairing equipment failures to reduce or avoid the impact of abnormalities on the power plant, achieve intelligent monitoring and early warning of the power plant process, improve the reliability and efficiency of the power plant, and ensure the stability of the power plant Run efficiently.

To sum up, the intelligent monitoring and early warning method for power plants provided by the embodiments of this application has the following technical effects:

Through the emission data collection station, test devices arranged at multiple locations in the target power plant are used to collect particulate matter concentration and carbon dioxide concentration at multiple locations. The target power plant is a thermal power plant, thereby achieving comprehensive monitoring of the overall operation of the power plant. Through multiple testing devices, water vapor concentrations at multiple locations are collected, and multiple particulate matter concentrations are compensated and analyzed based on multiple water vapor concentrations to obtain multiple compensated particulate matter concentrations to improve the accuracy of data collection. Based on the coordinates of multiple compensated particulate matter concentrations, multiple carbon dioxide concentrations and multiple locations, the particulate matter concentration field and carbon dioxide concentration field are constructed respectively, and processed and discriminated to obtain the particulate matter concentration characteristic field and carbon dioxide characteristic field to accurately locate abnormal conditions in power plant operations. . Through the power plant operation analysis station, based on the particle concentration characteristic field and carbon dioxide characteristic field, the operation abnormality analysis results of the target power plant are identified and obtained. The operation abnormality analysis results include whether an abnormality occurs and the abnormal location when the abnormality occurs, so as to realize the monitoring of power plant operation abnormality. . Through the emission analysis station, a power generation emission data matrix is constructed based on the particulate matter concentration field and carbon dioxide concentration field, and multiple emission scores for multiple locations are calculated and obtained based on the power generation emission data matrix to quantify the emission status of different parts of the power plant. Based on the operation abnormality analysis results and multiple emission scores, the operation monitoring results of the target power plant are generated. Based on the early warning station, early warning is carried out according to the operation monitoring results. Among them, the operation monitoring results include event description information of operation abnormal events to realize the operation of the power plant. Intelligent monitoring and early warning of the situation can comprehensively improve the accuracy and effectiveness of monitoring and early warning.

Example

Based on the same inventive concept as the power plant intelligent monitoring and early warning method in the previous embodiment, as shown in Figure 3, the embodiment of the present application provides an intelligent power plant monitoring and early warning system, which is applied to a power plant intelligent monitoring and early warning equipment. It includes an emission data collection station, a power plant operation analysis station, an emission analysis station and an early warning station. The system includes:

The power plant data collection module 11 is used to collect particulate matter concentration and carbon dioxide concentration at multiple locations through test devices arranged in multiple locations in the target power plant in the emission data collection station. The target power plant is a thermal power plant;

The concentration compensation analysis module 12 is used to collect water vapor concentrations at multiple locations through multiple testing devices, perform compensation analysis on multiple particulate matter concentrations based on the multiple water vapor concentrations, and obtain multiple compensated particulate matter concentrations;

The data processing and identification module 13 is used to construct a particulate matter concentration field and a carbon dioxide concentration field respectively based on the coordinates of multiple compensated particulate matter concentrations, multiple carbon dioxide concentrations and multiple locations, and perform processing and identification to obtain a particulate matter concentration characteristic field and a carbon dioxide characteristic field. ;

The operation abnormality analysis module 14 is used to identify and obtain the operation abnormality analysis results of the target power plant according to the particulate matter concentration characteristic field and the carbon dioxide characteristic field through the power plant operation analysis station. The operation abnormality analysis results include whether an abnormality occurs and whether an abnormality occurs. abnormal position;

The location emission scoring module 15 is used to construct a power generation emission data matrix based on the particulate matter concentration field and the carbon dioxide concentration field through the emission analysis station, and calculate and obtain multiple emission scores for multiple locations based on the power generation emission data matrix;

The monitoring result early warning module 16 is used to generate operation monitoring results of the target power plant based on the operation abnormality analysis results and multiple emission scores, and based on the early warning station, perform early warning according to the operation monitoring results, where the operation monitoring results include operation abnormal events. Event description information.

Further, the concentration compensation analysis module 12 includes the following execution steps:

Obtain sample particle concentration information records based on test data records tested by light scattering measurement methods;

Obtain sample compensation particle concentration information records based on test data records measured by sampling and weighing, and obtain sample water vapor concentration records based on water vapor concentration monitoring records;

According to the sample particulate matter concentration information record and the sample compensation particulate matter concentration information record, calculate and obtain the sample compensation coefficient record;

Sampling sample water vapor concentration record and sample compensation coefficient record, training to obtain compensation correction analysis channel;

Obtain multiple compensation coefficients and perform compensation and correction on multiple particulate matter concentration information. Multiple compensation coefficients are obtained by inputting multiple water vapor concentrations into the compensation correction analysis channel for processing.

Further, the job anomaly analysis module 14 includes the following execution steps:

Obtain the coordinate information of multiple locations, combine the multiple coordinate information with multiple carbon dioxide concentrations and multiple compensated particle concentrations to construct a particle concentration field and a carbon dioxide concentration field;

Use local processing operators to divide the particle concentration field and carbon dioxide concentration field to obtain multiple first local areas and multiple second local areas;

In each first local area, the particle concentration at the center position is used as the threshold, and the particle concentration at other locations is judged to be greater than, less than, or within the error range of the threshold, and marked as 1, -1, and 0 respectively to generate multiple local areas. Vector to obtain the particle concentration characteristic field;

Discriminate each second local area to obtain the carbon dioxide characteristic field.

Further, the job anomaly analysis module 14 also includes the following execution steps:

Based on the historical emission monitoring data of the target power plant, process and obtain multiple sample particulate matter concentration characteristic fields and multiple sample carbon dioxide characteristic fields, and obtain multiple sample operation abnormality analysis results through the operation abnormality monitoring data;

Multiple sample particulate matter concentration characteristic fields and multiple sample carbon dioxide characteristic fields are used as input training data, and multiple sample operation abnormality analysis results are used as output training data. Based on machine learning, in the power plant operation analysis station, the operation abnormality identification channel is trained and obtained ;

The operation abnormality analysis results are obtained by inputting the particulate matter concentration characteristic field and the carbon dioxide characteristic field into the operation abnormality identification channel for operation abnormality analysis.

Further, the location emission scoring module 15 includes the following execution steps:

Maximize the concentration data in the particulate matter concentration field and carbon dioxide concentration field, as follows:

;

Among them, y is the concentration data after maximization, and x is the original concentration data;

Based on the data in the maximized particulate matter concentration field and carbon dioxide concentration field, a power generation emission data matrix is constructed, as follows:

;

Among them, P is the power generation emission data matrix, is the maximized particle concentration data at the first position,/> is the maximized particle concentration data at the n-th location, n is the number of multiple locations,/> is the maximized carbon dioxide concentration data at the first position,/> is the maximized carbon dioxide concentration data at the nth position.

Further, the location emission scoring module 15 also includes the following execution steps:

According to the power generation emission data matrix, multiple emission scores for multiple locations are calculated and obtained as follows:

;

in, Score the emissions at the i-th location,/> ,/> and/> is the weight,/> and/> is the data after maximizing the particulate matter concentration standard and carbon dioxide concentration standard of the target power plant,/> and/> is the maximum particulate matter concentration data and maximum carbon dioxide concentration data at the i-th position,/> is the data in column j and row i in the power generation emission data matrix,/> is the minimum value of the jth column in the power generation emission data matrix,/> The weight assigned based on the impact of particulate matter concentration and carbon dioxide concentration on the emission quality of the target power plant.

Further, the monitoring result early warning module 16 includes the following execution steps:

According to the emission monitoring data of the target power plant, extract the data and calculate to obtain sample emission score records and sample emission quality grade records;

Use sample emission score records and sample emission quality grade records to construct an emission quality comparison table;

Match based on multiple emission scores to obtain multiple emission quality levels;

When an abnormality occurs in the operation abnormality analysis results or any emission quality level is lower than the qualified emission quality level, event description information is generated as the operation monitoring result.

In summary, any steps of the method described above can be stored in an unlimited computer memory as computer instructions or programs, and can be called and recognized by an unlimited computer processor to implement any of the embodiments of the present application. Method, no unnecessary restrictions are made here.

Furthermore, the first or second mentioned above may not only represent a sequence relationship, but may also represent a specific concept, and/or refer to multiple elements that can be selected individually or in full. Obviously, those skilled in the art can make various changes and modifications to the present application without departing from the scope of the present application. In this way, if these modifications and variations of the present application fall within the scope of the present application and its equivalent technology, the present application is intended to include these modifications and variations.

Claims (8)

1. A power plant intelligent monitoring and early warning method, characterized in that the method is applied to a power plant intelligent monitoring and early warning equipment, and the equipment includes an emission data collection station, a power plant operation analysis station, an emission analysis station and an early warning station , the method includes: Through the testing devices deployed at multiple locations in the target power plant in the emission data collection station, particulate matter concentrations and carbon dioxide concentrations at multiple locations are collected. The target power plant is a thermal power plant; Through multiple testing devices, water vapor concentrations at multiple locations are collected, and multiple particulate matter concentrations are compensated and analyzed based on multiple water vapor concentrations to obtain multiple compensated particulate matter concentrations; According to the coordinates of multiple compensated particulate matter concentrations, multiple carbon dioxide concentrations and multiple locations, the particulate matter concentration field and the carbon dioxide concentration field are constructed respectively, and processed and distinguished to obtain the particulate matter concentration characteristic field and the carbon dioxide characteristic field; Through the power plant operation analysis station, based on the particulate matter concentration characteristic field and the carbon dioxide characteristic field, the operation abnormality analysis results of the target power plant are identified and obtained. The operation abnormality analysis results include whether an abnormality occurs and the abnormal location when the abnormality occurs; Through the emission analysis station, a power generation emission data matrix is constructed based on the particulate matter concentration field and carbon dioxide concentration field, and multiple emission scores for multiple locations are calculated and obtained based on the power generation emission data matrix; Based on the operation abnormality analysis results and multiple emission scores, the operation monitoring results of the target power plant are generated. Based on the early warning station, early warning is carried out based on the operation monitoring results. The operation monitoring results include event description information of the operation abnormal events. 2. The method according to claim 1, characterized in that the method includes: Obtain sample particle concentration information records based on test data records tested by light scattering measurement methods; Obtain sample compensation particle concentration information records based on test data records measured by sampling and weighing, and obtain sample water vapor concentration records based on water vapor concentration monitoring records; According to the sample particulate matter concentration information record and the sample compensation particulate matter concentration information record, calculate and obtain the sample compensation coefficient record; Sampling sample water vapor concentration record and sample compensation coefficient record, training to obtain compensation correction analysis channel; Obtain multiple compensation coefficients and perform compensation and correction on multiple particulate matter concentration information. Multiple compensation coefficients are obtained by inputting multiple water vapor concentrations into the compensation correction analysis channel for processing. 3. The method according to claim 1, characterized in that the method includes: Obtain the coordinate information of multiple locations, combine the multiple coordinate information with multiple carbon dioxide concentrations and multiple compensated particle concentrations to construct a particle concentration field and a carbon dioxide concentration field; Use local processing operators to divide the particle concentration field and carbon dioxide concentration field to obtain multiple first local areas and multiple second local areas; In each first local area, the particle concentration at the center position is used as the threshold, and the particle concentration at other locations is judged to be greater than, less than, or within the error range of the threshold, and marked as 1, -1, and 0 respectively to generate multiple local areas. Vector to obtain the particle concentration characteristic field; Discriminate each second local area to obtain the carbon dioxide characteristic field. 4. The method according to claim 1, characterized in that the method includes: Based on the historical emission monitoring data of the target power plant, process and obtain multiple sample particulate matter concentration characteristic fields and multiple sample carbon dioxide characteristic fields, and obtain multiple sample operation abnormality analysis results through the operation abnormality monitoring data; Multiple sample particulate matter concentration characteristic fields and multiple sample carbon dioxide characteristic fields are used as input training data, and multiple sample operation abnormality analysis results are used as output training data. Based on machine learning, in the power plant operation analysis station, the operation abnormality identification channel is trained and obtained ; The operation abnormality analysis results are obtained by inputting the particulate matter concentration characteristic field and the carbon dioxide characteristic field into the operation abnormality identification channel for operation abnormality analysis. 5. The method according to claim 1, characterized in that, the method includes: Maximize the concentration data in the particulate matter concentration field and carbon dioxide concentration field, as follows: ; Among them, y is the concentration data after maximization, and x is the original concentration data; Based on the data in the maximized particulate matter concentration field and carbon dioxide concentration field, a power generation emission data matrix is constructed, as follows: ; Among them, P is the power generation emission data matrix, is the maximized particle concentration data at the first position,/> is the maximized particle concentration data at the n-th location, n is the number of multiple locations,/> is the maximized carbon dioxide concentration data at the first position,/> is the maximized carbon dioxide concentration data at the nth position. 6. The method according to claim 5, characterized in that the method includes: According to the power generation emission data matrix, multiple emission scores for multiple locations are calculated and obtained as follows: ; in, Score the emissions at the i-th location,/> ,/> and/> is the weight,/> and/> is the data after maximizing the particulate matter concentration standard and carbon dioxide concentration standard of the target power plant,/> and/> is the maximum particulate matter concentration data and maximum carbon dioxide concentration data at the i-th position,/> is the data in column j and row i in the power generation emission data matrix,/> is the minimum value of the jth column in the power generation emission data matrix,/> The weight assigned based on the impact of particulate matter concentration and carbon dioxide concentration on the emission quality of the target power plant. 7. The method according to claim 1, characterized in that the method includes: According to the emission monitoring data of the target power plant, extract the data and calculate to obtain sample emission score records and sample emission quality grade records; Use sample emission score records and sample emission quality grade records to construct an emission quality comparison table; Match based on multiple emission scores to obtain multiple emission quality levels; When an abnormality occurs in the operation abnormality analysis results or any emission quality level is lower than the qualified emission quality level, event description information is generated as the operation monitoring result. 8. A power plant intelligent monitoring and early warning system, characterized in that it is used to implement an intelligent power plant monitoring and early warning method according to any one of claims 1 to 7, and the system is applied to a power plant intelligent monitoring and early warning equipment, and the The equipment includes an emission data collection station, a power plant operation analysis station, an emission analysis station and an early warning station. The system includes: A power plant data collection module. The power plant data collection module is used to collect particulate matter concentration and carbon dioxide concentration at multiple locations through testing devices arranged in multiple locations in the target power plant in the emission data collection station. The target power plant is a thermal power plant; Concentration compensation analysis module, the concentration compensation analysis module is used to collect water vapor concentrations at multiple locations through multiple testing devices, perform compensation analysis on multiple particulate matter concentrations based on multiple water vapor concentrations, and obtain multiple compensated particulate matter concentrations; Data processing and discrimination module, the data processing and discrimination module is used to construct a particle concentration field and a carbon dioxide concentration field respectively based on multiple compensated particle concentrations, multiple carbon dioxide concentrations and coordinates of multiple locations, and perform processing and discrimination to obtain particle concentration characteristics. field and carbon dioxide characteristic field; An operation abnormality analysis module. The operation abnormality analysis module is used to identify and obtain the operation abnormality analysis results of the target power plant according to the particulate matter concentration characteristic field and the carbon dioxide characteristic field through the power plant operation analysis station. The operation abnormality analysis results include whether Abnormality occurs and the abnormal location when the abnormality occurs; The location emission scoring module is used to construct a power generation emission data matrix based on the particulate matter concentration field and the carbon dioxide concentration field through the emission analysis station, and calculate and obtain multiple emission scores for multiple locations based on the power generation emission data matrix. ; Monitoring result early warning module. The monitoring result early warning module is used to generate operation monitoring results of the target power plant based on operation abnormality analysis results and multiple emission scores. Based on the early warning station, early warning is performed according to the operation monitoring results, wherein the operation monitoring results are within Contains event description information for job exception events.