CN106567964B - A kind of flow control valve monitoring method based on DCS data - Google Patents
A kind of flow control valve monitoring method based on DCS data Download PDFInfo
- Publication number
- CN106567964B CN106567964B CN201610984487.0A CN201610984487A CN106567964B CN 106567964 B CN106567964 B CN 106567964B CN 201610984487 A CN201610984487 A CN 201610984487A CN 106567964 B CN106567964 B CN 106567964B
- Authority
- CN
- China
- Prior art keywords
- valve
- value
- working condition
- data
- score
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Active
Links
- 238000000034 method Methods 0.000 title claims abstract description 23
- 238000012544 monitoring process Methods 0.000 title claims abstract description 22
- 238000012549 training Methods 0.000 claims abstract description 25
- 239000002245 particle Substances 0.000 claims description 16
- 208000001750 Endoleak Diseases 0.000 claims description 6
- 206010064396 Stent-graft endoleak Diseases 0.000 claims description 6
- 238000005457 optimization Methods 0.000 claims description 6
- 238000005070 sampling Methods 0.000 claims description 5
- 230000015556 catabolic process Effects 0.000 abstract 1
- 230000009278 visceral effect Effects 0.000 description 6
- 230000000694 effects Effects 0.000 description 3
- 238000012423 maintenance Methods 0.000 description 3
- 238000004886 process control Methods 0.000 description 3
- 238000004364 calculation method Methods 0.000 description 2
- 238000003745 diagnosis Methods 0.000 description 2
- 238000004519 manufacturing process Methods 0.000 description 2
- 230000004044 response Effects 0.000 description 2
- 230000006978 adaptation Effects 0.000 description 1
- 230000003044 adaptive effect Effects 0.000 description 1
- 238000004458 analytical method Methods 0.000 description 1
- 238000013459 approach Methods 0.000 description 1
- 230000009286 beneficial effect Effects 0.000 description 1
- 238000012824 chemical production Methods 0.000 description 1
- 238000005516 engineering process Methods 0.000 description 1
- 230000006870 function Effects 0.000 description 1
- 238000009776 industrial production Methods 0.000 description 1
- 238000009434 installation Methods 0.000 description 1
- 230000003993 interaction Effects 0.000 description 1
- 230000003449 preventive effect Effects 0.000 description 1
- 238000006467 substitution reaction Methods 0.000 description 1
Classifications
-
- F—MECHANICAL ENGINEERING; LIGHTING; HEATING; WEAPONS; BLASTING
- F16—ENGINEERING ELEMENTS AND UNITS; GENERAL MEASURES FOR PRODUCING AND MAINTAINING EFFECTIVE FUNCTIONING OF MACHINES OR INSTALLATIONS; THERMAL INSULATION IN GENERAL
- F16K—VALVES; TAPS; COCKS; ACTUATING-FLOATS; DEVICES FOR VENTING OR AERATING
- F16K37/00—Special means in or on valves or other cut-off apparatus for indicating or recording operation thereof, or for enabling an alarm to be given
Landscapes
- Engineering & Computer Science (AREA)
- General Engineering & Computer Science (AREA)
- Mechanical Engineering (AREA)
- Flow Control (AREA)
- Feedback Control In General (AREA)
- Testing And Monitoring For Control Systems (AREA)
Abstract
The present invention relates to a kind of flow control valve monitoring methods based on DCS data, it is characterised in that：Include the following steps, step 1：Valve DCS data are obtained as training data, training data is classified according to different operating modes, establishes a variety of valve models；Step 2：Based on training data, corresponding valve model under different operating modes is determined；Step 3：It identifies current working, determines corresponding valve model under current working；Step 4：According to the valve model that step 3 determines, valve output flow value is predicted；Step 5：According to valve actual flow value and its correspondence estimated value, breakdown judge is carried out.
Description
Technical Field
The invention relates to a flow control valve monitoring method based on DCS data.
Background
The flow control valve is one of the most terminal control elements adopted in the industrial automatic production process, particularly the chemical production process, determines the timeliness and effectiveness of process control, and is an important link in the whole control loop but a weak link in the technology for a long time. Once the control valve fails, the control circuit will be forced to interrupt the control operation, possibly causing a larger system failure, resulting in plant shutdown and immeasurable economic loss. However, in the actual production process, the operator cannot monitor the control valve in real time, and only can perform preventive maintenance on important valves when the maintenance is planned. The maintenance mode is long in period, blind, untimely, low in efficiency and high in time and personnel cost.
The digital valve positioner becomes a main approach for designing a valve intelligent diagnosis system at home and abroad at present, and main control valve manufacturers such as FISER, SAMSON, INVENSYS and the like are all concerned with developing the digital valve positioner and perfecting respective diagnosis software. However, these software are tightly integrated with the manufacturer's hardware, and require the installation of additional valve positioners to obtain the corresponding sensing signals, so that failure analysis can be performed. Installing additional digital valve positioners presents process difficulties and is costly.
Distributed Control Systems (DCS) are comprehensive control systems which are developed along with continuous rise of automation of modern large-scale industrial production and increasingly complex process control requirements, are modern equipment for completing process control and process management, and most factories establish complete DCS databases for recording corresponding DCS data, and realize monitoring of flow control valves based on the DCS data is still a technical blank at present.
Disclosure of Invention
The invention aims to provide a flow control valve monitoring method based on DCS data, which can realize real-time monitoring on a flow control valve according to DSC data and effectively reduce the cost.
The technical scheme for realizing the purpose of the invention is as follows:
a flow control valve monitoring method based on DCS data is characterized in that: comprises the following steps of (a) carrying out,
step 1: acquiring valve DCS data as training data, classifying the training data according to different working conditions, and establishing various valve models;
step 2: determining corresponding valve models under different working conditions based on the training data;
and step 3: identifying the current working condition, and determining a corresponding valve model under the current working condition;
and 4, step 4: predicting the valve output flow value according to the valve model determined in the step 3;
and 5: and judging the fault according to the actual flow value of the valve and the corresponding estimated value.
In step 1, the training data are measured values of the valve controllerOutput value of。
In the step 2, parameters of each valve model are identified by a particle swarm optimization algorithm, the deviation of each valve model is calculated, and the valve model with the minimum deviation is selected as the valve model under the working condition.
And 3, estimating the flow value of the valve by using the valve models under different working conditions, solving the relative error between the current actual flow value and the estimated value of the current actual flow value, and selecting the working condition with the minimum relative error as the current working condition.
In step 5, calculating the actual flow value of the valveAnd its corresponding estimated valueAnd (4) using the deviation value to rank the degree of control valve failure.
In step 5, according to the actual flow value of the valveAnd its corresponding estimated valueThe internal leakage and filth blockage faults can be diagnosed, the specific formula is as follows,
if it is notIf yes, the result is judged as visceral obstruction, and the visceral obstruction score isInner leakage is 0; otherwise, visceral obstruction score is 0 and endoleak score is 0In the formulaMThe number of the selected real-time data is selected.
In step 5, according to the actual flow value of the valveAnd its corresponding estimated valueViscous faults can be diagnosed; viscous fault scoring is carried out by using least square method, and viscous fault scoringscoreThe specific formula of _2is as follows:
in the formulaT S Is the sampling period of the system and is,Mthe number of the selected real-time data is selected.
In step 5, according to the actual flow value of the valveAnd its corresponding estimated valueThe stuck fault can be diagnosed; stuck fault scoringscoreEquation of 3The following were used:
in the formulaMThe number of the selected real-time data is selected.
According to the output value of the valve controllerThe total stroke of the valve can be monitoredThe concrete formula is as follows,
and when a new working condition occurs to the valve, training by taking the current valve DCS data as new working condition data to obtain a valve model under the new working condition.
The invention has the following beneficial effects:
the method comprises the steps of obtaining valve DCS data as training data, classifying the training data according to different working conditions, and establishing various valve models; determining corresponding valve models under different working conditions based on the training data; identifying the current working condition, and determining a corresponding valve model under the current working condition; predicting the valve output flow value according to the determined valve model; and judging the fault according to the actual flow value of the valve and the corresponding estimated value. The invention can realize the online fault monitoring of the flow control valve only by using the DCS data of the control valve in the existing DCS database, and does not need to additionally install a valve positioner, thereby greatly reducing the monitoring cost; the method takes DCS data as training data, determines the corresponding valve models under different working conditions, increases man-machine interaction and learning capacity, is applicable to complex working conditions, and has more reliable monitoring effect.
The training data of the invention are measured values of the valve controllerOutput value ofAnd identifying parameters of each valve model by using a particle swarm optimization algorithm, calculating the deviation of each valve model, and selecting the valve model with the minimum deviation as the valve model under the working condition, so that the obtained valve model is more accurate.
The invention estimates the flow value of the valve by utilizing the valve models under different working conditions, obtains the relative error between the current actual flow value and the estimated value thereof, selects the working condition with the minimum relative error as the current working condition, can realize the accurate judgment of the current working condition and further ensures the monitoring effect on the flow control valve.
The invention is based on the actual flow value of the valveAnd its corresponding estimated valueThe error of the control valve can be graded according to the fault degree of the control valve and the actual flow value of the valveAnd its corresponding estimated valueRelative error between the two parts can diagnose internal leakage and filth blockage faults of the valve; according to the actual flow value of the valveAnd its corresponding estimated valueUsing least squares to estimate the valveReaction lag time, which can diagnose the viscous fault of the valve; according to the actual flow value of the valveAnd its corresponding estimated valueThe covariance between the two can diagnose the dead-locking fault of the valve, namely, the invention can realize the monitoring of various fault types of the flow control valve.
The invention can not only realize the fault monitoring of the flow control valve, but also can monitor the fault according to the output value of the valve controllerMonitoring the total stroke of the valveAnd the use degree of the valve is monitored.
Drawings
FIG. 1 is a flow chart of the present invention.
Detailed Description
As shown in fig. 1, the flow control valve monitoring method based on DCS data of the present invention includes the following steps,
step 1: and acquiring valve DCS data as training data, classifying the training data according to different working conditions, and establishing various valve models.
The training data being measured values of a valve controllerOutput value of，The representative is the valve opening degree,the representative is the flow value of the valve, and the data are obtained by a DCS database.
The valve model may be obtained from the flow characteristics of the valve, for example: the relationship of the logarithmic valve is
，
The relationship of the linear valve isMore complex hierarchical relationships may also be set, e.g.。
Step 2: and determining corresponding valve models under different working conditions based on the training data.
And identifying parameters of each valve model by using a particle swarm optimization algorithm, calculating the deviation of each valve model, and selecting the valve model with the minimum deviation as the valve model under the working condition.
Obtaining from DCS databaseOutput value of valve controllerAnd valve flow valueSeparately forming training data setsAnd. Calculating training data under each working condition by Particle Swarm Optimization (PSO)Andthe corresponding valve model type and corresponding parametersAndthe specific process is as follows:
(1) selecting particle numbers for particle swarm optimization(recommended value = 20) and number of evolutions(suggested value = 200);
(2) order toEach particle is assigned a random value，P m The value in (A) represents the parameter soughtAndwill beSubstitution function. Calculating an adaptation value for each particle
Finding out the adaptive value of 20 particlesThe smallest particle, the value of which is set as the global optimum positionWhile simultaneously adjusting the value of each particleSet to the historical optimum position of the particle itself. Setting the initial velocity of each particle。
(3) Order to(ii) a The positions of the 20 particles are updated by the following method:
wherein,andis a random number between 0 and 1 that is reassigned each time it evolves. Recalculating the fitness value for each particle ifThen, then(ii) a If it is notThen, then。
(4) If it is notThen return to (3), otherwise willValue of (5) to the parameterAndand the deviation of the model is。
Then comparing the modelsThe minimum model, which is recorded as the model of the valve at the current operating condition. When a control valve is presentIn different working conditionsShould be established for the control valveAn individual model。
And step 3: and identifying the current working condition and determining the corresponding valve model under the current working condition. And estimating the flow value of the valve by using the valve models under different working conditions, solving the relative error between the current actual flow value and the estimated value of the current actual flow value, and selecting the working condition with the minimum relative error as the current working condition.
Obtaining from DCS databaseReal-time data of each valveAndwhen on-line monitoring is performed, the data can be substitutedValve modelCalculating the data atRelative model deviation under different working condition modelsSelecting the working condition with the minimum deviation as the current working condition of the valve, and using the corresponding working condition modelPerforming output prediction and fault diagnosis。
And 4, step 4: predicting the valve output flow value according to the valve model determined in the step 3; calculating the predicted value under the current working conditionAnd simultaneously reading the minimum model deviation calculated in the step 3 as the model deviation of the current monitoring data.
And 5: and judging the fault according to the actual flow value of the valve and the corresponding estimated value.
(one) failure scoring
According to the actual flow value of the valveAnd its corresponding estimated valueAnd (3) calculating the fault score of the valve by combining the model deviation obtained in the step (3), wherein the calculation method comprises the following steps:
the fault score is found to be 0 to 100 and reflects the relative deviation of the current valve output (input) and the theoretical output (input). Using the score to perform fault classification:
the specific standard of grading can be adjusted according to the actual situation on site.
(II) diagnosing endoleak and filth blockage faults
According to the actual flow value of the valveAnd its corresponding estimated valueThe internal leakage and filth blockage faults can be diagnosed, the specific formula is as follows,
if it is notIf yes, the result is judged as visceral obstruction, and the visceral obstruction score isInner leakage is 0; otherwise, visceral obstruction score is 0 and endoleak score is 0In the formulaMThe number of the selected real-time data is selected. The physical meaning of the endoleak and filth failure score is the relative deviation of the current valve output (input) and the theoretical output (input).
(III) diagnosing viscous faults
According to the actual flow value of the valveAnd its corresponding estimated valueViscous faults can be diagnosed; viscous fault scoring is carried out by using least square method, and viscous fault scoringscoreThe specific formula of _2is as follows:
in the formulaT S Is the sampling period of the system and is,Mthe number of the selected real-time data is selected. The viscous fault score represents the response time of the valve, and a larger value indicates a longer response time of the valve.
(IV) diagnosing stuck-at faults
According to the actual flow value of the valveAnd its corresponding estimated valueThe stuck fault can be diagnosed; stuck fault scoringscoreThe calculation formula of _3is as follows:
in the formulaMThe number of the selected real-time data is selected. The physical meaning of the stuck fault score is a percentage probability between 0 and 100, with the greater the probability, the higher the likelihood of stuck.
The invention can not only realize the fault monitoring of the flow control valve, but also can monitor the fault according to the output value of the valve controllerMonitoring the total stroke of the valveAnd the use degree of the valve is monitored. The method comprises the following specific steps:
calculating the absolute value of the opening difference between the front sampling point and the rear sampling point of each valve, and accumulating:
similar to an automobile odometer, the valve usage degree can be reflected to a certain degree, and the usage degree of the valve at any two time pointsThe difference can represent the frequency of valve opening and closing in the period, and the value has great utilization value in industry, and can be used for monitoring the operation stability of the valve and evaluating the use degree of the valve, and also can be used for evaluating the control effect of the valve controller.
And when a new working condition occurs to the valve, training by taking the current valve DCS data as new working condition data to obtain a valve model under the new working condition.
Claims (5)
1. A flow control valve monitoring method based on DCS data is characterized in that: comprises the following steps of (a) carrying out,
step 1: acquiring valve DCS data as training data, classifying the training data according to different working conditions, and establishing various valve models;
step 2: determining corresponding valve models under different working conditions based on the training data;
and step 3: identifying the current working condition, and determining a corresponding valve model under the current working condition;
and 4, step 4: predicting the valve output flow value according to the valve model determined in the step 3;
and 5: according to the actual flow value of the valve and the corresponding estimated value, fault judgment is carried out;
in the step 1, training data are a measured value PV and an output value OP of a valve controller;
in step 2, identifying parameters of each valve model by using a particle swarm optimization algorithm, calculating the deviation of each valve model, and selecting the valve model with the minimum deviation as the valve model under the working condition;
in step 3, estimating the flow value of the valve by using the valve models under different working conditions, solving the relative error between the current actual flow value and the estimated value of the current actual flow value, and selecting the working condition with the minimum relative error as the current working condition;
in step 5, calculating the actual flow value of the valveAnd its corresponding estimated valueThe degree of control valve failure is graded by using the deviation value;
in step 5, according to the actual flow value of the valveAnd its corresponding estimated valueCan diagnose internal leakage and filth blockage faults, and has the following specific formula,
if result _1 is more than or equal to 0, judging that the filth is filth blockage, wherein the filth blockage score is score _1, and the endoleak score is 0; otherwise, the dirty block score is 0, the endoleak score is score _1, and M is the number of the selected real-time data.
2. The method of claim 1, wherein: in step 5, according to the actual flow value of the valveAnd its corresponding estimated valueViscous faults can be diagnosed; and performing viscous fault scoring by using a least square method, wherein a specific formula of the viscous fault scoring score 2 is as follows:
Θ＝(HTH)-1HTZ
in the formula TsThe sampling period of the system is M, and M is the number of the selected real-time data.
3. The method of claim 2, wherein: in step 5, according to the actual flow value of the valveAnd its corresponding estimated valueThe stuck fault can be diagnosed; the seizure fault score _3 is calculated as follows:
wherein M is the number of the selected real-time data.
4. The method of claim 3, wherein: the total stroke sum _ OP of the valve can be monitored based on the output OP of the valve controller, as follows,
ΔOP(l)＝OP(l)-OP(l-1)
sum_OP(l)＝sum_OP(l-1)+|ΔOP(l)| 。
5. the method of claim 4, wherein: and when a new working condition occurs to the valve, training by taking the current valve DCS data as new working condition data to obtain a valve model under the new working condition.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201610984487.0A CN106567964B (en) | 2016-11-09 | 2016-11-09 | A kind of flow control valve monitoring method based on DCS data |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201610984487.0A CN106567964B (en) | 2016-11-09 | 2016-11-09 | A kind of flow control valve monitoring method based on DCS data |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN106567964A CN106567964A (en) | 2017-04-19 |
| CN106567964B true CN106567964B (en) | 2018-10-23 |
Family
ID=58540746
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201610984487.0A Active CN106567964B (en) | 2016-11-09 | 2016-11-09 | A kind of flow control valve monitoring method based on DCS data |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN106567964B (en) |
Families Citing this family (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN106439199B (en) * | 2016-11-09 | 2019-04-09 | 北京化工大学 | A kind of control valve failure monitoring method based on DCS data |
| CN107728599B (en) * | 2017-09-01 | 2020-12-18 | 北京中燕信息技术有限公司 | Method and device for determining state of refining device valve |
| CN109222189B (en) * | 2018-09-07 | 2021-04-27 | 龙岩烟草工业有限责任公司 | Valve through-flow capacity mutation detection device and method and tobacco moisture regaining equipment |
| CN112924115A (en) * | 2021-03-16 | 2021-06-08 | 中电华创(苏州)电力技术研究有限公司 | Device and method for monitoring internal leakage of high-temperature and high-pressure pipeline valve |
| CN117553161B (en) * | 2023-11-25 | 2024-10-01 | 北京嘉联优控科技有限公司 | Valve fault diagnosis method and system based on positioner setting and feedback |
Citations (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN103488906A (en) * | 2013-09-30 | 2014-01-01 | 中国石油大学(华东) | Valve inner leakage defect type recognition and inner leakage rate calculation method |
| CN105095962A (en) * | 2015-07-27 | 2015-11-25 | 中国汽车工程研究院股份有限公司 | Method for predicting dynamic mechanical property of material based on BP artificial neural network |
| CN105930302A (en) * | 2016-04-06 | 2016-09-07 | 东南大学 | Rapid genetic algorithm and ash bin model based coal mill fault diagnosis method |
| CN105974327A (en) * | 2016-06-12 | 2016-09-28 | 广州市香港科大霍英东研究院 | Lithium battery pack SOC prediction method based on neural network and UKF |
| CN106054104A (en) * | 2016-05-20 | 2016-10-26 | 国网新疆电力公司电力科学研究院 | Intelligent ammeter fault real time prediction method based on decision-making tree |
| CN106439199A (en) * | 2016-11-09 | 2017-02-22 | 北京化工大学 | Monitoring method for control valve failure based on DCS data |
-
2016
- 2016-11-09 CN CN201610984487.0A patent/CN106567964B/en active Active
Patent Citations (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN103488906A (en) * | 2013-09-30 | 2014-01-01 | 中国石油大学(华东) | Valve inner leakage defect type recognition and inner leakage rate calculation method |
| CN105095962A (en) * | 2015-07-27 | 2015-11-25 | 中国汽车工程研究院股份有限公司 | Method for predicting dynamic mechanical property of material based on BP artificial neural network |
| CN105930302A (en) * | 2016-04-06 | 2016-09-07 | 东南大学 | Rapid genetic algorithm and ash bin model based coal mill fault diagnosis method |
| CN106054104A (en) * | 2016-05-20 | 2016-10-26 | 国网新疆电力公司电力科学研究院 | Intelligent ammeter fault real time prediction method based on decision-making tree |
| CN105974327A (en) * | 2016-06-12 | 2016-09-28 | 广州市香港科大霍英东研究院 | Lithium battery pack SOC prediction method based on neural network and UKF |
| CN106439199A (en) * | 2016-11-09 | 2017-02-22 | 北京化工大学 | Monitoring method for control valve failure based on DCS data |
Non-Patent Citations (1)
| Title |
|---|
| Nonlinear coal mill modeling and its application to model predictive control;Andrea Cortinovis et al.;《Control Engineering Practice》;20130331;第21卷(第3期);308-320 * |
Also Published As
| Publication number | Publication date |
|---|---|
| CN106567964A (en) | 2017-04-19 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN106567964B (en) | A kind of flow control valve monitoring method based on DCS data | |
| US10839115B2 (en) | Cleansing system for a feed composition based on environmental factors | |
| US20170315543A1 (en) | Evaluating petrochemical plant errors to determine equipment changes for optimized operations | |
| US20160292325A1 (en) | Advanced data cleansing system and method | |
| CN103064340B (en) | Failure prediction method facing to numerically-controlled machine tool | |
| US20160292188A1 (en) | Data cleansing system and method for inferring a feed composition | |
| US7526405B2 (en) | Statistical signatures used with multivariate statistical analysis for fault detection and isolation and abnormal condition prevention in a process | |
| CN109524139A (en) | A kind of real-time device performance monitoring method based on equipment working condition variation | |
| CN108363836B (en) | Multi-working-condition self-adaptive industrial robot health degree assessment method and system | |
| CN109213127A (en) | A kind of HVAC system gradual failure diagnostic method based on deep learning | |
| US10261480B2 (en) | Computer-implemented method and system for automatically monitoring and determining the status of entire process segments in a process unit | |
| CN116658492B (en) | Intelligent power catwalk and method thereof | |
| CN106439199B (en) | A kind of control valve failure monitoring method based on DCS data | |
| CN109084971B (en) | A kind of pneumatic control valve method for diagnosing faults based on particle group optimizing | |
| CN115427767A (en) | Improved pattern recognition techniques for data-driven fault detection within a process plant | |
| CN112560339B (en) | Method for predicting guide bearing bush temperature of hydroelectric generating set by utilizing machine learning | |
| CN117128162B (en) | Intelligent energy air compression station energy-saving control system and control method | |
| CN118133470A (en) | Configuration modeling method of simulation software interface | |
| CN108709426B (en) | Sintering machine air leakage fault online diagnosis method based on frequency spectrum characteristic bilateral detection method | |
| CN115186754A (en) | Unit energy efficiency monitoring and diagnosing method based on information entropy and auto-associative regression model | |
| CN108268017A (en) | The appraisal procedure and optimization method and its apparatus for evaluating of tank production line efficiency processed | |
| WO2019005541A1 (en) | Evaluating petrochemical plant errors to determine equipment changes for optimized operations | |
| CN113159148B (en) | Extended Kalman filtering method for FDD variable memory length of water chilling unit | |
| CN118884887A (en) | PLC intelligent fault diagnosis and prevention system based on machine learning | |
| CN114371686A (en) | Multi-working-condition process fault detection method and system based on local neighborhood standardization |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |