CN115186754A - Unit energy efficiency monitoring and diagnosing method based on information entropy and auto-associative regression model - Google Patents
Unit energy efficiency monitoring and diagnosing method based on information entropy and auto-associative regression model Download PDFInfo
- Publication number
- CN115186754A CN115186754A CN202210839653.3A CN202210839653A CN115186754A CN 115186754 A CN115186754 A CN 115186754A CN 202210839653 A CN202210839653 A CN 202210839653A CN 115186754 A CN115186754 A CN 115186754A
- Authority
- CN
- China
- Prior art keywords
- condition
- energy efficiency
- value
- condition parameter
- deviation degree
- 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.)
- Pending
Links
Images
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
- G06Q10/00—Administration; Management
- G06Q10/06—Resources, workflows, human or project management; Enterprise or organisation planning; Enterprise or organisation modelling
- G06Q10/063—Operations research, analysis or management
- G06Q10/0639—Performance analysis of employees; Performance analysis of enterprise or organisation operations
- G06Q10/06393—Score-carding, benchmarking or key performance indicator [KPI] analysis
Abstract
The invention discloses a unit energy efficiency monitoring and diagnosing method based on information entropy and auto-associative regression model, which comprises the following steps: collecting historical data of the unit in all-working-condition operation interval, and preprocessing the historical data; clustering the condition parameters by taking decision variable minimization as a target, and establishing a state estimation model of an optimal operation condition area; determining the weight value of each condition parameter according to the information entropy weight of the condition parameter; inputting the condition parameters in the unit operation process into a state estimation model of an optimal operation condition area in real time, predicting an estimation value of the condition parameters, calculating a deviation combination weighted value of the estimation value and a measured value, constructing an energy efficiency deviation degree, and obtaining a sliding average deviation degree after filtering; and comparing the sliding average deviation with a low-energy-efficiency deviation threshold and a high-energy-efficiency deviation threshold, and giving a unit energy efficiency monitoring diagnosis scheme. The method is based on a fusion data mining model of auto-associative regression estimation, data clustering and information entropy, and is used for monitoring the energy efficiency of the unit.
Description
Technical Field
The invention relates to the technical field of unit energy efficiency state monitoring, in particular to a turbine energy efficiency monitoring and diagnosing method based on information entropy and an auto-associative regression model.
Background
At present, the energy efficiency monitoring and diagnosis of most units still stay at the qualitative diagnosis level, but with the deepening of national energy conservation and emission reduction work, the traditional mode of judging the energy efficiency state of the unit by means of operation experience cannot meet the current development requirement.
With the continuous improvement of the automation informatization level of the domestic unit, a data-driven unit energy efficiency information extraction and analysis technology is developed to a certain extent, but the judgment of the energy efficiency still adopts more theoretical values or design values and is lack of dynamics. In addition, the data-driven energy efficiency monitoring evaluation also has the problems of not considering influence difference between variables, combining energy efficiency monitoring and fault diagnosis and the like.
The invention provides a novel information entropy weight method distribution attribute weight calculation method for determining the weight of a variable, and a comprehensive evaluation method combining energy efficiency monitoring and fault diagnosis by monitoring the current energy efficiency state by adopting a multivariate auto-associative regression technology based on massive historical data and obtaining a space with better energy efficiency of the actual operation of a unit through cluster analysis.
Disclosure of Invention
The invention aims to provide a unit energy efficiency monitoring and diagnosing method based on information entropy and an auto-associative regression model, which is a fusion data mining model based on auto-associative regression estimation, data clustering and information entropy and is used for unit energy efficiency monitoring.
In order to achieve the purpose, the invention adopts the following technical scheme: a unit energy efficiency monitoring and diagnosing method based on information entropy and auto-associative regression models specifically comprises the following steps:
step 5, inputting the condition parameters in the unit operation process into a state estimation model of an optimal operation condition area in real time, predicting the estimation value of the condition parameters, calculating the deviation of the estimation value and the measured value, combining the weight value in the step 4, constructing an energy efficiency deviation degree, filtering the energy efficiency deviation degree to obtain a moving average deviation degree, and determining a low energy efficiency deviation degree threshold and a high energy efficiency deviation degree threshold;
and 6, comparing the sliding average deviation with a low-energy-efficiency deviation threshold and a high-energy-efficiency deviation threshold, and giving a unit energy efficiency monitoring diagnosis scheme.
Further, the process of preprocessing the historical data in step 1 specifically includes:
step 11, respectively judging whether each condition parameter exceeds a threshold value according to a normal operation threshold value of each condition parameter in historical data, and deleting a state vector corresponding to the condition parameter if the condition parameter exceeds the threshold value;
and step 12, judging whether the fluctuation rate of the adjacent time of each condition parameter is smaller than a set value or not for the condition parameters in the reserved state vectors, and if not, deleting the state vectors corresponding to the condition parameters.
Further, the state estimation model of the optimal operating condition region in step 3 is:
wherein x is est，j For the estimated value of the j-th condition parameter in the test set, x kj Is the measured value of the jth condition parameter in the test set under the kth working condition, n is the total quantity of the condition parameters under each working condition in the test set, and q (k) is x kj Q (k) = ker (x, x '), wherein the ker () is a gaussian kernel function, x is a test set, and x' is a working condition library.
Further, step 4 comprises the following sub-steps:
step 41, calculating the initial entropy weight of each condition parameter according to the information entropy of the condition parameter
Step 42, calculating the sub-entropy weight as
Step 43, determining the weight value of each condition parameter as
Wherein H j For the information entropy of the jth condition parameter,
is the mean value of the entropy of information, H k For the information entropy of the kth condition parameter, k ≠ j.
Further, the energy efficiency deviation degree is constructed by the following process:
wherein, dist i Represents the energy efficiency deviation degree at the ith time,
weight value, x, representing the jth condition parameter est，j For estimated value of j-th condition parameter, x j Is the measured value of the jth condition parameter.
Further, the energy-inefficient deviation threshold is represented as δ 1 =3.5 σ, the energy-efficient deviation threshold is denoted δ 2 =6 σ, where σ represents a standard deviation of the energy efficiency deviation degree.
Further, the process of obtaining the moving average deviation degree in the step 5 is as follows: filtering the energy efficiency deviation degree through moving average filtering, and setting the number of moving averages to be l, so that the moving average deviation degree
Wherein, dist i And (4) the energy efficiency deviation degree at the ith moment.
Further, the specific process of step 6 is: when the sliding average deviation degree is lower than the low-energy-efficiency deviation degree threshold value, representing that the condition parameters run reasonably; when the sliding average deviation degree is between the low energy efficiency deviation degree threshold value and the high energy efficiency deviation degree threshold value, a corresponding adjusting method is given by combining corresponding condition parameters with an operation adjusting principle and experience; and when the sliding average deviation exceeds the high-energy-efficiency deviation threshold, carrying out operation adjustment on corresponding condition parameters, carrying out operation adjustment if the operation adjustment can improve the energy efficiency, otherwise, considering the condition of equipment failure, carrying out corresponding reasoning according to corresponding sign parameters and combining expert diagnosis knowledge, and positioning the failed equipment.
Further, when the moving average deviation exceeds the high energy efficiency deviation threshold, the precondition for performing the operation adjustment on the corresponding condition parameter is as follows: performing moving average filtering processing on the deviation of the estimated value and the measured value, setting the number of the moving averages to be l, and obtaining the moving average deviation
If the moving average deviation exceeds the control threshold of the corresponding condition parameter, the corresponding condition parameter deviates from the optimal operation interval and needs to be adjusted; wherein, err k Indicates the deviation of the estimated value from the measured value, and i indicates the ith time.
Further, the control threshold of the condition parameter is determined by inputting the test set into the state estimation model of the optimal operating condition area to obtain an estimated value of the condition parameter, and counting the deviation between the estimated value and the measured value of the condition parameter.
Compared with the prior art, the invention has the following beneficial effects:
1. according to the steam turbine energy efficiency monitoring and diagnosing method based on the information entropy and the auto-associative regression model, the heat consumption of the steam turbine is minimized, an optimal condition parameter set is clustered, the influence of artificial subjectivity is avoided, and the scientific theories are stronger;
2. the invention provides a steam turbine energy efficiency monitoring and diagnosing method based on information entropy and auto-associative regression model, and provides a weight calculation method of condition parameters based on the information entropy theory, so that the weight difference between important parameters and secondary parameters in the condition parameters is more obvious, and the prediction of the model is more reliable by combining with the auto-associative regression estimation;
3. the turbine energy efficiency monitoring and diagnosing method based on the information entropy and the auto-associative regression model is used for designing the moving average deviation index by utilizing the auto-associative regression estimation model to meet the requirement of real-time monitoring, is used for monitoring the energy efficiency index in real time and improves the monitoring reliability.
Drawings
FIG. 1 is a flow chart of a unit energy efficiency monitoring and diagnosing method based on information entropy and auto-associative regression models;
fig. 2 is a graph showing the effect of parameter estimation in an optimal operating space of a unit according to an index of heat consumption and energy efficiency of a steam turbine, wherein (a) in fig. 2 is a graph showing a comparison between an estimated value and an actually measured value of a main steam pressure, (b) in fig. 2 is a graph showing a comparison between an estimated value and an actually measured value of a feedwater temperature, and (c) in fig. 2 is a graph showing a comparison between an estimated value and an actually measured value of a condenser vacuum;
fig. 3 is a diagram illustrating the effect of monitoring the energy efficiency index of the steam turbine, wherein (a) in fig. 3 is a diagram illustrating deviation degree early warning, and (b) in fig. 3 is a diagram illustrating condenser vacuum early warning.
Detailed Description
The technical solution of the present invention is further explained below with reference to the accompanying drawings.
Fig. 1 is a flowchart of the unit energy efficiency monitoring and diagnosing method based on the information entropy and the auto-associative regression model, and the unit energy efficiency monitoring and diagnosing method specifically includes the following steps:
The process of preprocessing the historical data specifically comprises the following steps:
step 11, respectively judging whether each condition parameter exceeds a threshold value according to a normal operation threshold value of each condition parameter in historical data, and deleting a state vector corresponding to the condition parameter if the condition parameter exceeds the threshold value;
and step 12, judging whether the fluctuation rate of the adjacent time of each condition parameter is smaller than a set value or not for the condition parameters in the reserved state vectors, and if not, deleting the state vectors corresponding to the condition parameters.
Due to the fact that a unit system is complex, the number of measuring point parameter types is large, the field production environment with high temperature and high pressure is provided with electromagnetic interference, communication failure and the like, and the condition that a few measured values are abnormal can be caused.
And 3, establishing a state estimation model of the optimal operation working condition area by the test set through an auto-associative regression estimation method, wherein the auto-associative regression estimation method is a non-parameter estimation method, so that hyper-parameters needing to be manually selected do not exist in the state estimation model, and the calculation process is scientific and reasonable. The auto-associative regression estimation method is characterized in that an optimal condition parameter set obtained by clustering is used as a reference working condition set for monitoring the running state, and the closer the state estimation value is to a state vector in the working condition set, the closer the current state is, and even the current state belongs to the optimal working condition set, otherwise, the energy efficiency state deviates from a better space and needs to be adjusted.
The establishment process of the state estimation model of the optimal operation condition area specifically comprises the following steps:
wherein x is est，j For the estimated value of the jth condition parameter in the test set, x kj Is the measured value of the jth condition parameter in the test set under the kth working condition, n is the total quantity of the condition parameters under each working condition in the test set, and q (k) is x kj Q (k) = ker (x, x '), wherein the ker () is a gaussian kernel function, x is a test set, and x' is a working condition library.
Taking a steam turbine as an example, fig. 2 is a result obtained by calculating a test set by an auto-associative regression estimation method, fig. 2 (a) is a comparison graph of a main steam pressure estimation value and an actual measurement value, fig. 2 (b) is a comparison graph of a feedwater temperature estimation value and an actual measurement value, fig. 2 (c) is a comparison graph of a condenser vacuum estimation value and an actual measurement value, and as can be seen from fig. 2 (a) - (c), the estimation values of the three condition parameters are the result obtained by calculating the auto-associative regression estimation pair, the estimation values are very close to the actual measurement value, the deviation is small, the test data set belongs to a better working condition, and the condition estimation model is verified to be effective in accordance with the selected result.
And 4, considering the difference of the influence degrees of the condition parameters on the energy efficiency, objectively considering the importance degrees of different parameters when calculating the energy efficiency related indexes, and calculating the information entropy of each condition parameter to determine the corresponding weight, wherein the unreasonable situation that the entropy value difference is different and the entropy weight calculation value is the same occurs when the information entropy is calculated by using the traditional entropy weight calculation formula. Therefore, a two-step method for determining the entropy weight is designed, the initial entropy weight and the sub-entropy weight are sequentially calculated according to the information entropy weight of the condition parameter, and the sum of the initial entropy weight and the sub-entropy weight forms the total weight value of the parameter, so that the importance degree of the parameter can be favorably distinguished. The method specifically comprises the following substeps:
step 41, calculating the initial entropy weight of each condition parameter according to the information entropy of the condition parameter
Step 42, calculating the sub-entropy weight as
Step 43, determining the weight value of each condition parameter as
Wherein H j For the information entropy of the jth condition parameter,
is the mean value of the entropy of information, H k The information entropy of the kth condition parameter is k ≠ j.
And 5, inputting the condition parameters in the running process of the unit into a state estimation model of the optimal running working condition area in real time, predicting the estimation value of the condition parameters, calculating the deviation of the estimation value and the measured value, combining the weighted value in the step 4, constructing an energy efficiency deviation degree, wherein the energy efficiency deviation degree covers the information of all the condition parameters, the energy efficiency deviation degree in a normal threshold value range substantially indicates that the current state vector belongs to a better energy efficiency state space, and once the deviation degree exceeds the threshold value, the condition parameters are necessarily deviated to a certain degree, so that the analysis and adjustment of operators are facilitated. In the real-time monitoring process, in order to improve the reliability of the deviation index, filtering processing is carried out on the energy deviation degree to obtain a moving average deviation degree, and a low energy efficiency deviation degree threshold value and a high energy efficiency deviation degree threshold value are determined.
The construction process of the energy efficiency deviation degree in the invention is as follows:
wherein, dist i Represents the energy efficiency deviation degree at the ith time,
weight value, x, representing the jth condition parameter est，j For an estimated value of the jth condition parameter, x j Is the measured value of the jth condition parameter.
Degree of moving average deviation in the inventionThe process is as follows: filtering the energy efficiency deviation degree through moving average filtering, and setting the number of moving averages to be l, so that the moving average deviation degree
Wherein, dist i And the energy efficiency deviation degree at the ith moment.
The threshold value of the low energy efficiency deviation degree in the invention is expressed as delta 1 =3.5 σ, and the energy-efficient offset threshold is denoted δ 2 =6 σ, where σ represents a standard deviation of the energy efficiency deviation degree. The reduction of the energy efficiency in the operation of the unit is divided into two conditions, wherein the first condition is that the energy efficiency is reduced to a certain degree, the caused reasons comprise the change, disturbance and the like of external conditions, the conditions are normal, the reduction amplitude and fluctuation are generally not too large, and the energy efficiency can be improved as much as possible through operation adjustment; the second reason for the reduction of energy efficiency is caused by equipment or system failure, the energy efficiency may be reduced to a large extent or suddenly, and comprehensive judgment needs to be performed to determine the failure reason, so that the situation is prevented from being enlarged. Therefore, the low-energy-efficiency deviation threshold and the high-energy-efficiency deviation threshold are respectively set to preliminarily judge the reason of the energy efficiency reduction, so that different conditions can be distinguished and timely processed.
When the sliding average deviation degree exceeds the high energy efficiency deviation degree threshold value, corresponding strips are processedThe precondition for adjusting the operation of the part parameters is as follows: performing moving average filtering processing on the deviation of the estimated value and the measured value, setting the number of the moving averages to be l, and obtaining the moving average deviation
If the moving average deviation exceeds the control threshold of the corresponding condition parameter, the corresponding condition parameter deviates from the optimal operation interval and needs to be adjusted; wherein, err k Indicating the deviation of the estimated value from the measured value, and i indicates the ith time. The control threshold of the condition parameters is input into the state estimation model of the optimal operation condition area through the test set to obtain the estimation value of the condition parameters, and the deviation between the estimation value and the measured value of the condition parameters is counted and determined.
Fig. 3 is an effect diagram of monitoring an energy efficiency index of a steam turbine by using the energy efficiency monitoring and diagnosing method of the present invention, and it can be seen that a monitoring process of heat consumption increase caused by condenser vacuum deterioration is performed, and it can be seen from (a) in fig. 3 that sensitivity of an energy efficiency deviation index is high, and an abnormal rising trend appears in a value after a certain time, which indicates that energy efficiency is low in a current operating state, and a condenser pressure parameter calculated by using a model also appears an abnormal rising condition, as shown in (b) in fig. 3, but at this time, a scene operator does not prompt a condenser vacuum alarm, and after a period of time, a DCS operator displays a prompt of DCS alarm, thereby verifying that monitoring of the present invention is effective and that the present invention has an early warning effect.
The above are only preferred embodiments of the present invention, and the scope of the present invention is not limited to the above embodiments, and all technical solutions that fall under the spirit of the present invention belong to the scope of the present invention. It should be noted that modifications and embellishments within the scope of the invention may be made by those skilled in the art without departing from the principle of the invention.
Claims (10)
1. A unit energy efficiency monitoring and diagnosing method based on information entropy and auto-associative regression models is characterized by comprising the following steps:
step 1, collecting historical data of a unit full-working-condition operation interval, and preprocessing the historical data; the historical data is composed of a state vector [ x ] j ,x m [ composition of a set, wherein x j Is the measured value of the condition parameter, j is the index of the condition parameter, j belongs to [1, m-1 ]]，x m M is the total number of condition parameters and decision variables as decision variables;
step 2, dividing the preprocessed full-working-condition operation interval into a plurality of local working condition areas, clustering condition parameters in each local working condition area by taking decision variable minimization as a target, and obtaining an optimal condition parameter set; dividing the condition parameter set into a test set and a working condition library;
step 3, establishing a state estimation model of the optimal operation condition area by the test set through an auto-associative regression estimation method;
step 4, calculating an initial entropy weight and a secondary entropy weight according to the information entropy weight of the condition parameters, and determining the weight value of each condition parameter;
step 5, inputting the condition parameters in the unit operation process into a state estimation model of an optimal operation condition area in real time, predicting the estimation value of the condition parameters, calculating the deviation of the estimation value and the measured value, combining the weight value in the step 4, constructing an energy efficiency deviation degree, filtering the energy efficiency deviation degree to obtain a moving average deviation degree, and determining a low energy efficiency deviation degree threshold and a high energy efficiency deviation degree threshold;
and 6, comparing the sliding average deviation degree with a low-energy-efficiency deviation degree threshold value and a high-energy-efficiency deviation degree threshold value, and giving a unit energy efficiency monitoring and diagnosing scheme.
2. The unit energy efficiency monitoring and diagnosing method based on the information entropy and the auto-associative regression model as claimed in claim 1, wherein the process of preprocessing the historical data in the step 1 specifically comprises:
step 11, respectively judging whether each condition parameter exceeds a threshold value according to a normal operation threshold value of each condition parameter in historical data, and deleting a state vector corresponding to the condition parameter if the condition parameter exceeds the threshold value;
and step 12, judging whether the fluctuation rate of the adjacent time of each condition parameter is less than a set value or not for the condition parameters in the reserved state vectors, and if not, deleting the state vectors corresponding to the condition parameters.
3. The unit energy efficiency monitoring and diagnosing method based on the information entropy and the auto-associative regression model as claimed in claim 1, wherein the state estimation model of the optimal operation condition region in the step 3 is:
wherein x is est,j For the estimated value of the j-th condition parameter in the test set, x kj Is the measured value of j condition parameter in the test set under k condition, n is the total quantity of condition parameters under each condition in the test set, q (k) is x kj Q (k) = ker (x, x '), where the ker () is a gaussian kernel function, x is a test set, and x' is a working condition library.
4. The method for monitoring and diagnosing the energy efficiency of the steam turbine based on the information entropy and the auto-associative regression model according to claim 1, wherein the step 4 comprises the following substeps:
step 41, calculating the initial entropy weight of each condition parameter according to the information entropy of the condition parameter
Step 42, calculating the sub-entropy weight as
Step 43, determining the weight value of each condition parameter as
Wherein H j For the information entropy of the jth condition parameter,
is the mean value of entropy of information, H k For the information entropy of the kth condition parameter, k ≠ j.
5. The method for monitoring and diagnosing the unit energy efficiency based on the information entropy and the auto-associative regression model is characterized in that the energy efficiency deviation degree is constructed by the following steps:
wherein dist i Indicates the energy efficiency deviation degree at the ith moment,
weight value, x, representing the jth condition parameter est,j For an estimated value of the jth condition parameter, x j Is the measured value of the jth condition parameter.
6. The unit energy efficiency monitoring and diagnosing method based on the information entropy and auto-associative regression model as claimed in claim 1, wherein the threshold value of the low energy efficiency deviation degree is represented as delta 1 =3.5 sigma, said energy-efficient deviation threshold is denoted delta 2 =6 σ, where σ represents a standard deviation of the energy efficiency deviation degree.
7. The method for monitoring and diagnosing the energy efficiency of the unit based on the information entropy and the auto-associative regression model as claimed in claim 1, wherein the process of obtaining the moving average deviation degree in the step 5 is as follows: filtering the energy efficiency deviation degree through moving average filtering, and setting the number of the moving averages to be l, so as to obtain the moving average deviation degree
Wherein, dist i And the energy efficiency deviation degree at the ith moment.
8. The unit energy efficiency monitoring and diagnosing method based on the information entropy and the auto-associative regression model according to claim 1, wherein the specific process of the step 6 is as follows: when the sliding average deviation degree is lower than the low-energy-efficiency deviation degree threshold value, representing that the condition parameters run reasonably; when the sliding average deviation degree is between the low-energy-efficiency deviation degree threshold value and the high-energy-efficiency deviation degree threshold value, a corresponding adjusting method is given by combining corresponding condition parameters with an operation adjusting principle and experience; and when the moving average deviation exceeds the high-energy-efficiency deviation threshold, carrying out operation adjustment on corresponding condition parameters, if the operation adjustment can improve the energy efficiency, carrying out operation adjustment, otherwise, considering the condition of equipment failure, carrying out corresponding reasoning according to corresponding sign parameters and combining expert diagnosis knowledge, and positioning the failed equipment.
9. The unit energy efficiency monitoring and diagnosing method based on the information entropy and the auto-associative regression model according to claim 8, wherein when the moving average deviation exceeds the high energy efficiency deviation threshold, the precondition for performing operation adjustment on the corresponding condition parameter is: performing moving average filtering processing on the deviation between the estimated value and the measured value, setting the number of the moving averages to be l, and obtaining the moving average deviation
If the moving average deviation exceeds the control threshold of the corresponding condition parameter, the corresponding condition parameter deviates from the optimal operation interval and needs to be adjusted; wherein, err k Indicating the deviation of the estimated value from the measured value, and i indicates the ith time.
10. The unit energy efficiency monitoring and diagnosing method based on the information entropy and auto-associative regression model according to claim 9, wherein the control threshold of the condition parameter is obtained by inputting a test set into a state estimation model of an optimal operation condition area to obtain an estimated value of the condition parameter, and a deviation between the estimated value and a measured value of the condition parameter is counted to determine.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202210839653.3A CN115186754A (en) | 2022-07-18 | 2022-07-18 | Unit energy efficiency monitoring and diagnosing method based on information entropy and auto-associative regression model |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202210839653.3A CN115186754A (en) | 2022-07-18 | 2022-07-18 | Unit energy efficiency monitoring and diagnosing method based on information entropy and auto-associative regression model |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN115186754A true CN115186754A (en) | 2022-10-14 |
Family
ID=83520304
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202210839653.3A Pending CN115186754A (en) | 2022-07-18 | 2022-07-18 | Unit energy efficiency monitoring and diagnosing method based on information entropy and auto-associative regression model |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN115186754A (en) |
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN117408787A (en) * | 2023-12-15 | 2024-01-16 | 江西求是高等研究院 | Root cause mining analysis method and system based on decision tree |
-
2022
- 2022-07-18 CN CN202210839653.3A patent/CN115186754A/en active Pending
Cited By (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN117408787A (en) * | 2023-12-15 | 2024-01-16 | 江西求是高等研究院 | Root cause mining analysis method and system based on decision tree |
| CN117408787B (en) * | 2023-12-15 | 2024-03-05 | 江西求是高等研究院 | Root cause mining analysis method and system based on decision tree |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN107146004B (en) | A kind of slag milling system health status identifying system and method based on data mining | |
| US11840998B2 (en) | Hydraulic turbine cavitation acoustic signal identification method based on big data machine learning | |
| CN108681633B (en) | Condensate pump fault early warning method based on state parameters | |
| CN104714537B (en) | A kind of failure prediction method based on the relative mutation analysis of joint and autoregression model | |
| CN110738274A (en) | nuclear power device fault diagnosis method based on data driving | |
| CN110689075A (en) | Fault prediction method of self-adaptive threshold of refrigeration equipment based on multi-algorithm fusion | |
| CN111090939B (en) | Early warning method and system for abnormal working condition of petrochemical device | |
| CN109538311B (en) | Real-time monitoring method for control performance of steam turbine in high-end power generation equipment | |
| CN112598144B (en) | CNN-LSTM burst fault early warning method based on correlation analysis | |
| CN114386312A (en) | Equipment fault diagnosis method | |
| CN112836941A (en) | Online health condition evaluation method for high-pressure steam turbine system of coal-electric unit | |
| CN115186754A (en) | Unit energy efficiency monitoring and diagnosing method based on information entropy and auto-associative regression model | |
| CN112700085A (en) | Association rule based method, system and medium for optimizing steady-state operation parameters of complex system | |
| Chen et al. | Adaptive transfer learning for multimode process monitoring and unsupervised anomaly detection in steam turbines | |
| CN114326486A (en) | Process monitoring method based on probability slow feature analysis and elastic weight consolidation | |
| Chesterman et al. | Condition monitoring of wind turbines and extraction of healthy training data using an ensemble of advanced statistical anomaly detection models | |
| CN117032120A (en) | Integrated intelligent cloud control system and control method for air compression station | |
| CN105302476B (en) | A kind of reliability data online acquisition for nuclear power plant equipment analyzes storage system and its storage method | |
| CN106439199A (en) | Monitoring method for control valve failure based on DCS data | |
| CN115496188A (en) | Coal mill fault early warning method based on deep learning convolutional neural network | |
| CN114997309A (en) | Water feed pump fault early warning method and device | |
| KR102561062B1 (en) | Monitoring system and method for nuclear power plant | |
| WO2022165792A1 (en) | Method and system of sensor fault management | |
| CN115212996A (en) | Fault diagnosis system for coal mill | |
| Song et al. | Anomaly detection of wind turbine generator based on temporal information |
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 |