CN109470946B - Power generation equipment fault detection method and system - Google Patents
Power generation equipment fault detection method and system Download PDFInfo
- Publication number
- CN109470946B CN109470946B CN201811088218.1A CN201811088218A CN109470946B CN 109470946 B CN109470946 B CN 109470946B CN 201811088218 A CN201811088218 A CN 201811088218A CN 109470946 B CN109470946 B CN 109470946B
- Authority
- CN
- China
- Prior art keywords
- power generation
- generation equipment
- model
- voltage
- value
- 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
- 238000010248 power generation Methods 0.000 title claims abstract description 158
- 238000001514 detection method Methods 0.000 title claims abstract description 58
- 238000012544 monitoring process Methods 0.000 claims abstract description 54
- 238000000034 method Methods 0.000 claims abstract description 24
- 238000012549 training Methods 0.000 claims description 54
- 238000000611 regression analysis Methods 0.000 claims description 38
- 238000012360 testing method Methods 0.000 claims description 35
- 238000012706 support-vector machine Methods 0.000 claims description 28
- 230000000694 effects Effects 0.000 claims description 6
- 238000011156 evaluation Methods 0.000 claims description 6
- 238000010801 machine learning Methods 0.000 claims description 6
- 238000013507 mapping Methods 0.000 claims description 6
- 230000003111 delayed effect Effects 0.000 claims description 3
- 238000000638 solvent extraction Methods 0.000 claims 1
- 238000012512 characterization method Methods 0.000 abstract description 8
- 230000008569 process Effects 0.000 abstract description 6
- 238000010586 diagram Methods 0.000 description 8
- 238000004590 computer program Methods 0.000 description 7
- 230000006870 function Effects 0.000 description 6
- 238000007781 pre-processing Methods 0.000 description 5
- 238000012545 processing Methods 0.000 description 4
- 238000003860 storage Methods 0.000 description 3
- 230000009286 beneficial effect Effects 0.000 description 2
- 238000011161 development Methods 0.000 description 2
- 238000004458 analytical method Methods 0.000 description 1
- 230000008859 change Effects 0.000 description 1
- 238000004140 cleaning Methods 0.000 description 1
- 238000010276 construction Methods 0.000 description 1
- 238000013075 data extraction Methods 0.000 description 1
- 238000009826 distribution Methods 0.000 description 1
- 238000005516 engineering process Methods 0.000 description 1
- 238000012423 maintenance Methods 0.000 description 1
- 238000004519 manufacturing process Methods 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 230000003287 optical effect Effects 0.000 description 1
- 238000012805 post-processing Methods 0.000 description 1
- 230000009467 reduction Effects 0.000 description 1
Images
Classifications
-
- G—PHYSICS
- G01—MEASURING; TESTING
- G01R—MEASURING ELECTRIC VARIABLES; MEASURING MAGNETIC VARIABLES
- G01R31/00—Arrangements for testing electric properties; Arrangements for locating electric faults; Arrangements for electrical testing characterised by what is being tested not provided for elsewhere
Abstract
A method and system for detecting faults of power generation equipment comprises the following steps: substituting the current generating power of the generating equipment into a pre-established generating equipment fault detection fitting model to obtain a current voltage fitting value; determining whether the power generation equipment is in fault or not at present based on the current voltage fitting value and the voltage monitoring value; substituting the prediction time into a pre-established power generation equipment operation situation prediction model to obtain a voltage prediction value of the power generation equipment at the prediction time; whether the power generation equipment has faults in the prediction time is determined based on the voltage prediction value, and fitting data and prediction data of key characterization features related to the power generation equipment are provided by detecting whether the power generation equipment has faults or not in real time and the operation situation after a certain time, so that the detection result is more convincing, and the safety of the power generation equipment in the normal operation process is improved.
Description
The technical field is as follows:
the invention belongs to the field of power generation equipment fault detection, and particularly relates to a power generation equipment fault detection method and system.
Background art:
since the industrial revolution, the functional requirements for the detection of equipment failure have arisen, and generally, workers judge whether equipment has failed or not, mainly through experience accumulated for a long time or observation of the appearance of the equipment. Along with the rapid advance of industrial development, industrial equipment is increasingly developed and develops towards the direction of intelligence, large size, high speed and distribution. Compared with the traditional method, the structure of the equipment is more complex, and meanwhile, the functions are also more powerful, so that the maintenance difficulty is increased. Then, the probability of equipment failure is greatly improved compared with the prior art, the information data of the equipment failure is exponentially increased, and the generated mass data cannot be loaded manually, so that correct and effective failure analysis cannot be completed.
With the rapid development of information technology, particularly the large-scale construction of various information systems such as a power generation enterprise control system and the like, the data volume accumulated by a power plant is increased rapidly, a new generation of equipment state online monitoring system can help a user to realize the intelligent management of equipment states, the professional efficiency of an equipment manager is brought into full play, fault post-processing is changed into real-time control of the whole dynamic change of the equipment in operation, the operation safety level and efficiency of the equipment are greatly improved in the life cycle of each equipment, the unplanned shutdown and accidents caused by the equipment are reduced, and more benefits are created for power generation enterprises.
Currently, fault detection methods for power generation equipment are threshold-based detection methods. Setting alarm, early warning and fault upper and lower limit thresholds for each key characterization monitoring node of the power generation equipment manually according to experience, and comparing real-time monitoring data with the thresholds to confirm whether the working state of the current equipment is normal or not; the method separates each key characteristic of the power generation equipment, cannot comprehensively consider the operation condition of the equipment, and is not beneficial to analyzing the whole equipment when the number of the key characteristics of the equipment is large.
The invention content is as follows:
in order to overcome the above-mentioned drawbacks, the present invention provides a method for detecting a fault of a power generation device, the method comprising:
substituting the current generating power of the generating equipment into a pre-established generating equipment fault detection fitting model to obtain a current voltage fitting value;
determining whether the power generation equipment is in fault or not at present based on the current voltage fitting value and the voltage monitoring value;
substituting the prediction time into a pre-established power generation equipment operation situation prediction model to obtain a voltage prediction value of the power generation equipment at the prediction time;
determining whether the power generation equipment predicted time is out of order based on the voltage predicted value.
Preferably, the building of the power generation equipment fault detection fitting model includes:
obtaining corresponding voltage based on the historical current and the generated power of the power generation equipment;
dividing the power generation equipment historical temperature, voltage and power generation data into a training set and a test set;
determining the high-dimensional space distance relation between the historical temperature, the historical voltage and the generated power by adopting a support vector machine regression analysis fitting model based on a training set;
and testing based on the test set, and adjusting the high-dimensional space distance relation parameters of the historical temperature, the historical voltage and the generated power to obtain a power generation equipment fault detection fitting model.
Preferably, the establishing of the power generation equipment operation situation prediction model includes:
determining a high-dimensional space distance relation among the historical temperature, the historical voltage and the generated power by adopting a support vector machine regression analysis prediction model based on a training set;
and testing based on the test set, and adjusting the high-dimensional space distance relation parameters of the historical temperature, the historical voltage and the generated power to obtain a power generation equipment operation situation prediction model.
Preferably, the dividing into a training set and a test set based on the historical temperature, voltage and power generation data of the power generation equipment comprises:
acquiring historical temperature, current and power generation monitoring data samples of each monitoring point of the power generation equipment;
performing vector mapping based on the monitoring data samples to obtain a time sequence vector set corresponding to each monitoring point;
dividing the time sequence vector set according to a time sequence;
and dividing the divided time sequence vector set into a training set and a testing set according to a proportion.
Preferably, the determining whether the power generation equipment is currently in fault based on the current voltage fitting value and the voltage monitoring value includes:
and comparing the current voltage fitting value with the current voltage monitoring value, and if the difference value between the current voltage fitting value and the current voltage monitoring value exceeds a preset range, judging that the power generation equipment has a current operation fault.
Preferably, the determining whether the power generation equipment predicted time is failed based on the voltage predicted value includes:
and comparing the predicted voltage value with a preset normal voltage range, and if the predicted voltage value is not in the normal voltage range, carrying out operation failure on the power generation equipment within the prediction time.
Preferably, the determining the operation fault of the power generation equipment at the current and predicted time based on the current voltage fitting value, the current voltage monitoring value, the voltage predicted value and the voltage in the normal operation state further comprises:
and when the operation fault of the power generation equipment is determined, warning the power generation equipment based on the current voltage fitting value and the voltage prediction value.
A power generation equipment fault detection system, the system comprising:
a first drive-in module: the system comprises a power generation device, a fault detection fitting model, a voltage value fitting model and a voltage value fitting model, wherein the power generation device is used for generating power for the power generation device;
a first determination module: the current voltage fitting value and the voltage monitoring value are used for determining whether the power generation equipment is in fault currently;
a second bringing module: the device is used for substituting the prediction time into a pre-established power generation equipment operation situation prediction model to obtain a voltage prediction value of the power generation equipment at the prediction time;
a second determination module: for determining whether the power plant predicted time failed based on the voltage predicted value.
Preferably, the first bring-in module further includes: a power generation equipment fault detection fitting model module;
obtaining corresponding voltage based on the historical current and the generated power of the power generation equipment;
dividing the historical temperature, voltage and power generation power data of the power generation equipment into a training set and a test set;
determining the high-dimensional space distance relation between the historical temperature, the historical voltage and the generated power by adopting a support vector machine regression analysis fitting model based on a training set;
and testing based on the test set, and adjusting the high-dimensional space distance relation parameters of the historical temperature, the historical voltage and the generated power to obtain a power generation equipment fault detection fitting model.
Preferably, the second bringing module further includes: establishing a power generation equipment operation situation prediction model module;
determining a high-dimensional space distance relation among the historical temperature, the historical voltage and the generated power by adopting a support vector machine regression analysis prediction model based on a training set;
and testing based on the test set, and adjusting the high-dimensional space distance relation parameters of the historical temperature, the historical voltage and the generated power to obtain a power generation equipment operation situation prediction model.
Compared with the prior art, the invention has the following beneficial effects:
1. according to the method and the system for detecting the fault of the power generation equipment, the current power generation power of the power generation equipment is brought into a pre-established power generation equipment fault detection fitting model to obtain a current voltage fitting value; determining whether the power generation equipment is in fault or not at present based on the current voltage fitting value and the voltage monitoring value; substituting the prediction time into a pre-established power generation equipment operation situation prediction model to obtain a voltage prediction value of the power generation equipment at the prediction time; whether the power generation equipment has faults or not is determined based on the voltage predicted value, and all key features of the power generation equipment are comprehensively analyzed, so that the problem of fault detection of the equipment is solved more comprehensively and better.
2. According to the method and the system for detecting the power generation equipment fault, whether the power generation equipment has the fault or not and the operation situation after a certain time are detected in real time, fitting data and prediction data of key characterization features related to the power generation equipment are provided, the detection result is more convincing, and the safety of the power generation equipment in the normal operation process is improved.
Description of the drawings:
FIG. 1 is a flow chart of support vector machine based power plant fault detection of the present invention;
FIG. 2 is a flow chart of a power plant fault detection/prediction model set up of the present invention.
The specific implementation mode is as follows:
for better understanding of the present invention, the technical solutions in the embodiments of the present invention will be clearly and completely described below with reference to the drawings in the embodiments of the present invention, and it is obvious that the described embodiments are some embodiments of the present invention, but not all embodiments.
Example 1
As shown in fig. 1, the specific steps are as follows:
the method comprises the following steps: substituting the current generating power of the generating equipment into a pre-established generating equipment fault detection fitting model to obtain a current voltage fitting value;
step two: determining whether the power generation equipment is in fault or not at present based on the current voltage fitting value and the voltage monitoring value;
step three: substituting the prediction time into a pre-established power generation equipment operation situation prediction model to obtain a voltage prediction value of the power generation equipment at the prediction time;
step four: determining whether the power generation equipment predicted time is out of order based on the voltage predicted value.
The invention provides a fault detection method of power generation equipment, which realizes the function of detecting whether the power generation equipment has faults in the operation process, detects whether the power generation equipment normally operates, gives the operation state of the power generation equipment after a certain time and improves the safety of the working environment of the power generation equipment.
The invention provides a generating equipment fault detection method based on a support vector machine, aiming at realizing the problem of accurately detecting whether generating equipment is in fault or normal operation after a certain time. The following describes in further detail embodiments of the present invention.
The method comprises the following steps: substituting the current generating power of the generating equipment into a pre-established generating equipment fault detection fitting model to obtain a current voltage fitting value;
s1. power generation equipment real-time monitoring data extraction: acquiring real-time monitoring data of power generation equipment in batch in a normal operation state;
s2, device feature preprocessing and vectorization: preprocessing real-time monitoring data of the equipment, mapping the preprocessed real-time monitoring data to a vector space, and acquiring a time sequence vector set corresponding to each key measuring point of the equipment;
s3. Power plant failure detection modeling: training and establishing an equipment fault detection model based on a support vector machine regression analysis method;
s1. the real-time monitoring data acquisition method comprises the following steps:
s11, acquiring the analog quantity and state quantity real-time monitoring data of key feature nodes related to the power generation equipment from the real-time monitoring database, and forming a feature data set;
and S12, acquiring real-time monitoring data of key characterization analog quantities related to the power generation equipment from the real-time monitoring database, and forming a label characterization data set.
S2, the specific steps of device feature preprocessing and vectorization comprise:
s21, after the real-time monitoring data are acquired, preprocessing operations including cleaning, interpolation, noise reduction, association, standardization and other related preprocessing operations are carried out on the real-time monitoring data, and each data acquisition monitoring point is ensured to have normal data at any data acquisition time point;
and S22, mapping the real-time monitoring data of the equipment to a vector space, setting the numerical value of the corresponding dimensionality of the vector according to the standardized result of the real-time monitoring data value, and acquiring a time sequence vector sequence aiming at the measuring point of the same equipment.
s3. As shown in FIG. 2, the specific steps of the power generation equipment fault detection model establishment include:
s31, dividing the real-time monitoring data into a plurality of groups of training data according to a time sequence, wherein the time length of each group of data is the same, the training feature set is characteristic data, and the characteristic data set is characteristic data with a time axis aligned with the characteristic data;
and S32, aiming at each group of real-time monitoring data, preferably, selecting a regression analysis model of the support vector machine as a machine learning method to establish a regression analysis fitting model.
Step s32 further includes:
s321, preferably, selecting a regression analysis model of the support vector machine for training;
s322, initializing parameters of a power generation equipment fault detection model;
s323, training a power generation equipment fault detection model by using a training set to obtain a fitting model of regression analysis;
s324, evaluating the effect of the detection model by using the test set, and recording the model result;
and s325, updating the model parameters according to the evaluation result, and repeating the steps s323-s324 until the preset condition is met.
Step two: determining whether the power generation equipment is in fault or not at present based on the current voltage fitting value and the voltage monitoring value;
the process of model training involved in s3 is illustrated in FIG. 2.
Acquiring real-time equipment monitoring data aiming at specific power generation equipment, and mapping the data to a vector space;
calculating the characteristic vector of the current moment by using the fault detection and situation prediction model trained in the step s3 to obtain a characteristic fitting value of the equipment at the current moment and a predicted value after a certain time;
and determining whether the equipment fails according to the difference between the fitting value of the characterization data and the actual monitoring value and other conditions.
Step three: substituting the prediction time into a pre-established power generation equipment operation situation prediction model to obtain a voltage prediction value of the power generation equipment at the prediction time;
s4. training of a prediction model of the operation situation of the power generation equipment: training and establishing an equipment operation situation prediction model based on a support vector machine regression analysis method;
s4. the concrete steps of the establishment of the prediction model of the power generation equipment running situation include:
s41, dividing the real-time monitoring data into a plurality of groups of training data according to a time sequence, wherein the time length of each group of data is the same, the training feature set is characteristic data, and the characteristic data set is characteristic data of which the time axis is delayed by a certain time compared with the characteristic data;
and s42, for each group of real-time monitoring data, preferably, selecting a regression analysis model of the support vector machine as a machine learning method to establish a regression analysis prediction model.
Step s42 further includes:
preferably, selecting a regression analysis model of the support vector machine for training;
s422, initializing parameters of a power generation equipment operation situation prediction model;
step 423, training a power generation equipment operation situation prediction model by using a training set to obtain a prediction model of regression analysis;
s424, evaluating the prediction effect of the detection model by using the test set, and recording the model result;
and s425, updating the model parameters according to the evaluation result, and repeating the steps s323-s324 until the preset condition is met.
Step four: determining whether the power generation equipment predicted time is out of order based on the voltage predicted value.
The process of model training involved in s4 is illustrated in FIG. 2.
Acquiring real-time equipment monitoring data aiming at specific power generation equipment, and mapping the data to a vector space;
the fault detection and situation prediction model trained in the step s4 calculates the feature vector at the current moment to obtain a characterization fitting value of the equipment at the current moment and a predicted value after a certain time;
and determining whether the equipment fails according to the difference between the fitting value of the characterization data and the actual monitoring value and other conditions.
Example 2
Based on the same concept, the invention also provides a power generation equipment fault detection system, which comprises:
a first drive-in module: the system comprises a power generation device, a fault detection fitting model, a voltage value fitting model and a voltage value fitting model, wherein the power generation device is used for generating power for the power generation device;
a first determination module: the current voltage fitting value and the voltage monitoring value are used for determining whether the power generation equipment is in fault currently;
a second bringing module: the device is used for substituting the prediction time into a pre-established power generation equipment operation situation prediction model to obtain a voltage prediction value of the power generation equipment at the prediction time;
a second determination module: for determining whether the power plant predicted time failed based on the voltage predicted value.
The first bring-in module further comprises: a power generation equipment fault detection fitting model module;
obtaining corresponding voltage based on the historical current and the generated power of the power generation equipment;
dividing the power generation equipment historical temperature, voltage and power generation data into a training set and a test set;
determining the high-dimensional space distance relation between the historical temperature, the historical voltage and the generated power by adopting a support vector machine regression analysis fitting model based on a training set;
and testing based on the test set, and adjusting the high-dimensional space distance relation parameters of the historical temperature, the historical voltage and the generated power to obtain a power generation equipment fault detection fitting model.
The second obtaining module further includes: establishing a power generation equipment operation situation prediction model module;
determining a high-dimensional space distance relation among the historical temperature, the historical voltage and the generated power by adopting a support vector machine regression analysis prediction model based on a training set;
and testing based on the test set, and adjusting the high-dimensional space distance relation parameters of the historical temperature, the historical voltage and the generated power to obtain a power generation equipment operation situation prediction model.
As will be appreciated by one skilled in the art, embodiments of the present application may be provided as a method, system, or computer program product. Accordingly, the present application may take the form of an entirely hardware embodiment, an entirely software embodiment or an embodiment combining software and hardware aspects. Furthermore, the present application may take the form of a computer program product embodied on one or more computer-usable storage media (including, but not limited to, disk storage, CD-ROM, optical storage, and the like) having computer-usable program code embodied therein.
The present application is described with reference to flowchart illustrations and block diagrams of methods, systems, and computer program products according to embodiments of the application. It will be understood that each flow and block of the flow diagrams and block diagrams, and combinations of flows and blocks in the flow diagrams and block diagrams, can be implemented by computer program instructions. These computer program instructions may be provided to a processor of a general purpose computer, special purpose computer, embedded processor, or other programmable data processing apparatus to produce a machine, such that the instructions, which execute via the processor of the computer or other programmable data processing apparatus, create means for implementing the functions specified in the flowchart flow or flows and block diagram block or blocks.
These computer program instructions may also be stored in a computer-readable memory that can direct a computer or other programmable data processing apparatus to function in a particular manner, such that the instructions stored in the computer-readable memory produce an article of manufacture including instruction means which implement the function specified in the flowchart flow or flows and block diagram block or blocks.
These computer program instructions may also be loaded onto a computer or other programmable data processing apparatus to cause a series of operational steps to be performed on the computer or other programmable apparatus to produce a computer implemented process such that the instructions which execute on the computer or other programmable apparatus provide steps for implementing the functions specified in the flowchart flow or flows and block diagram block or blocks.
The present invention is not limited to the above embodiments, and any modifications, equivalent replacements, improvements, etc. made within the spirit and principle of the present invention are included in the scope of the claims of the present invention which are filed as the application.
Claims (6)
1. A method of power generation equipment fault detection, the method comprising:
substituting the current generating power of the generating equipment into a pre-established generating equipment fault detection fitting model to obtain a current voltage fitting value; determining whether the power generation equipment is in fault or not at present based on the current voltage fitting value and the voltage monitoring value;
the establishment of the power generation equipment fault detection fitting model comprises the following steps:
obtaining corresponding voltage based on the historical current and the generated power of the power generation equipment;
dividing the power generation equipment historical temperature, voltage and power generation data into a training set and a test set;
determining the high-dimensional space distance relation between the historical temperature, the historical voltage and the generated power by adopting a support vector machine regression analysis fitting model based on a training set;
testing based on the test set, and adjusting the high-dimensional space distance relation parameters of the historical temperature, the historical voltage and the generated power to obtain a power generation equipment fault detection fitting model;
for each group of real-time monitoring data, selecting a support vector machine regression analysis model as a machine learning method to establish a regression analysis fitting model, which specifically comprises the following steps:
a. selecting a regression analysis model of a support vector machine for training;
b. initializing parameters of a power generation equipment fault detection model;
c. training a power generation equipment fault detection model by using a training set to obtain a fitting model of regression analysis;
d. evaluating the effect of the detection model by using the test set, and recording the model result;
e. updating the model parameters according to the evaluation result, and repeating the steps c-d until the preset condition is met;
substituting the prediction time into a pre-established power generation equipment operation situation prediction model to obtain a voltage prediction value of the power generation equipment at the prediction time; determining whether the power generation equipment predicted time is out of order based on a voltage predicted value;
the establishment of the power generation equipment operation situation prediction model comprises the following steps:
determining a high-dimensional space distance relation among the historical temperature, the historical voltage and the generated power by adopting a support vector machine regression analysis prediction model based on a training set;
testing based on the test set, and adjusting the high-dimensional space distance relation parameters of the historical temperature, the historical voltage and the generated power to obtain a power generation equipment operation situation prediction model;
training a power generation equipment operation situation prediction model: training and establishing an equipment operation situation prediction model based on a support vector machine regression analysis method;
the specific steps of establishing the power generation equipment operation situation prediction model comprise:
dividing the real-time monitoring data into a plurality of groups of training data according to a time sequence, wherein the time length of each group of data is the same, the training feature set is characteristic data, and the characteristic data set is characteristic data of which the time axis is delayed for a certain time compared with the characteristic data;
for each group of real-time monitoring data, selecting a support vector machine regression analysis model as a machine learning method to establish a regression analysis prediction model, which specifically comprises the following steps:
a. selecting a regression analysis model of a support vector machine for training;
b. initializing parameters of a power generation equipment operation situation prediction model;
c. training a power generation equipment operation situation prediction model by using a training set to obtain a prediction model of regression analysis;
d. evaluating the prediction effect of the detection model by using the test set, and recording the model result;
e. and updating the model parameters according to the evaluation result, and repeating the steps c-d until the preset condition is met.
2. The power generation equipment fault detection method of claim 1, wherein the partitioning into a training set and a test set based on power generation equipment historical temperature, voltage, and generated power data comprises:
acquiring historical temperature, current and power generation monitoring data samples of each monitoring point of the power generation equipment;
performing vector mapping based on the monitoring data samples to obtain a time sequence vector set corresponding to each monitoring point;
dividing the time sequence vector set according to a time sequence;
and dividing the divided time sequence vector set into a training set and a testing set according to a proportion.
3. The power generation equipment fault detection method of claim 1, wherein said determining whether the power generation equipment is currently faulty based on the current voltage fit value and a voltage monitor value comprises:
and comparing the current voltage fitting value with the current voltage monitoring value, and if the difference value between the current voltage fitting value and the current voltage monitoring value exceeds a preset range, judging that the power generation equipment has a current operation fault.
4. The power plant fault detection method of claim 1, wherein the determining whether the power plant predicted time is faulty based on the voltage predicted value comprises:
and comparing the predicted voltage value with a preset normal voltage range, and if the predicted voltage value is not in the normal voltage range, carrying out operation failure on the power generation equipment within the prediction time.
5. The power generation equipment fault detection method of claim 4, wherein the determining of the operating faults of the power generation equipment at the current and predicted times based on the current voltage fit value, the current voltage monitored value, the predicted voltage value and the voltage under normal operating conditions further comprises:
and when the operation fault of the power generation equipment is determined, warning the power generation equipment based on the current voltage fitting value and the voltage prediction value.
6. A power generation equipment fault detection system, the system comprising:
a first drive-in module: the system comprises a power generation device, a fault detection fitting model, a voltage value fitting model and a voltage value fitting model, wherein the power generation device is used for generating power for the power generation device;
a first determination module: the current voltage fitting value and the voltage monitoring value are used for determining whether the power generation equipment is in fault currently;
the first bring-in module further comprises: a power generation equipment fault detection fitting model module;
obtaining corresponding voltage based on the historical current and the generated power of the power generation equipment;
dividing the power generation equipment historical temperature, voltage and power generation data into a training set and a test set;
determining the high-dimensional space distance relation between the historical temperature, the historical voltage and the generated power by adopting a support vector machine regression analysis fitting model based on a training set;
testing based on the test set, and adjusting the high-dimensional space distance relation parameters of the historical temperature, the historical voltage and the generated power to obtain a power generation equipment fault detection fitting model;
for each group of real-time monitoring data, selecting a support vector machine regression analysis model as a machine learning method to establish a regression analysis fitting model, which specifically comprises the following steps:
a. selecting a regression analysis model of a support vector machine for training;
b. initializing parameters of a power generation equipment fault detection model;
c. training a power generation equipment fault detection model by using a training set to obtain a fitting model of regression analysis;
d. evaluating the effect of the detection model by using the test set, and recording the model result;
e. updating the model parameters according to the evaluation result, and repeating the steps c-d until the preset condition is met;
a second bringing module: the device is used for substituting the prediction time into a pre-established power generation equipment operation situation prediction model to obtain a voltage prediction value of the power generation equipment at the prediction time;
a second determination module: determining whether the power generation equipment predicted time is out of order based on a voltage predicted value;
the second drop-in module further comprises: establishing a power generation equipment operation situation prediction model module;
determining a high-dimensional space distance relation among the historical temperature, the historical voltage and the generated power by adopting a support vector machine regression analysis prediction model based on a training set;
testing based on the test set, and adjusting the high-dimensional space distance relation parameters of the historical temperature, the historical voltage and the generated power to obtain a power generation equipment operation situation prediction model;
training a power generation equipment operation situation prediction model: training and establishing an equipment operation situation prediction model based on a support vector machine regression analysis method;
the specific steps of establishing the power generation equipment operation situation prediction model comprise:
dividing the real-time monitoring data into a plurality of groups of training data according to a time sequence, wherein the time length of each group of data is the same, the training feature set is characteristic data, and the characteristic data set is characteristic data of which the time axis is delayed for a certain time compared with the characteristic data;
for each group of real-time monitoring data, selecting a support vector machine regression analysis model as a machine learning method to establish a regression analysis prediction model, which specifically comprises the following steps:
a. selecting a regression analysis model of a support vector machine for training;
b. initializing parameters of a power generation equipment operation situation prediction model;
c. training a power generation equipment operation situation prediction model by using a training set to obtain a prediction model of regression analysis;
d. evaluating the prediction effect of the detection model by using the test set, and recording the model result;
e. and updating the model parameters according to the evaluation result, and repeating the steps c-d until the preset condition is met.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201811088218.1A CN109470946B (en) | 2018-09-18 | 2018-09-18 | Power generation equipment fault detection method and system |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201811088218.1A CN109470946B (en) | 2018-09-18 | 2018-09-18 | Power generation equipment fault detection method and system |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN109470946A CN109470946A (en) | 2019-03-15 |
| CN109470946B true CN109470946B (en) | 2022-07-08 |
Family
ID=65663072
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201811088218.1A Active CN109470946B (en) | 2018-09-18 | 2018-09-18 | Power generation equipment fault detection method and system |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN109470946B (en) |
Families Citing this family (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN110488188B (en) * | 2019-08-09 | 2021-05-04 | 中国电建集团华东勘测设计研究院有限公司 | Unit three-dimensional health quantitative evaluation method based on dynamic threshold |
| CN111047082B (en) * | 2019-12-02 | 2024-03-29 | 广州智光电气股份有限公司 | Early warning method and device of equipment, storage medium and electronic device |
| CN111310946B (en) * | 2020-02-26 | 2022-11-29 | 清华大学 | Transformer operation situation sensing method and system based on intelligent operation and maintenance |
| CN113487062A (en) * | 2021-05-31 | 2021-10-08 | 国网上海市电力公司 | Power load prediction method based on periodic automatic encoder |
| CN113837289B (en) * | 2021-09-26 | 2024-03-19 | 创新奇智(重庆)科技有限公司 | Model training method, fault prediction device and electronic equipment |
| CN117574244B (en) * | 2024-01-15 | 2024-04-02 | 成都秦川物联网科技股份有限公司 | Ultrasonic water meter fault prediction method, device and equipment based on Internet of things |
Citations (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101109938A (en) * | 2007-08-23 | 2008-01-23 | 上海交通大学 | Monitoring system for large wind power station on the sea |
| CN102566435A (en) * | 2012-02-17 | 2012-07-11 | 冶金自动化研究设计院 | Performance prediction and fault alarm method for photovoltaic power station |
| CN102721924A (en) * | 2012-06-26 | 2012-10-10 | 新疆金风科技股份有限公司 | Fault early warning method of wind generating set |
| CN104124757A (en) * | 2014-06-30 | 2014-10-29 | 国网青海省电力公司电力科学研究院 | Background operation monitoring system for power grid fault simulation detection device of high-altitude photovoltaic power station |
| CN104655977A (en) * | 2014-12-19 | 2015-05-27 | 华北电力大学（保定） | Generator exciting winding short-circuit fault diagnosis method based on torque comparison principle |
| CN105004962A (en) * | 2015-03-24 | 2015-10-28 | 华北电力大学（保定） | Improved method for detecting short trouble of exciting winding of turbonator |
Family Cites Families (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US9019108B2 (en) * | 2010-08-05 | 2015-04-28 | General Electric Company | Thermal measurement system for fault detection within a power generation system |
| US9845012B2 (en) * | 2011-07-06 | 2017-12-19 | General Electric Company | System and method for predicting mechanical failure of a motor |
| US9461573B2 (en) * | 2012-04-24 | 2016-10-04 | Masdar Institute Of Science And Technology | Fault handling system for doubly fed induction generator |
| CN103558547B (en) * | 2013-11-01 | 2015-12-02 | 东南大学 | A kind of intelligent failure diagnosis method of permagnetic synchronous motor |
-
2018
- 2018-09-18 CN CN201811088218.1A patent/CN109470946B/en active Active
Patent Citations (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101109938A (en) * | 2007-08-23 | 2008-01-23 | 上海交通大学 | Monitoring system for large wind power station on the sea |
| CN102566435A (en) * | 2012-02-17 | 2012-07-11 | 冶金自动化研究设计院 | Performance prediction and fault alarm method for photovoltaic power station |
| CN102721924A (en) * | 2012-06-26 | 2012-10-10 | 新疆金风科技股份有限公司 | Fault early warning method of wind generating set |
| CN104124757A (en) * | 2014-06-30 | 2014-10-29 | 国网青海省电力公司电力科学研究院 | Background operation monitoring system for power grid fault simulation detection device of high-altitude photovoltaic power station |
| CN104655977A (en) * | 2014-12-19 | 2015-05-27 | 华北电力大学（保定） | Generator exciting winding short-circuit fault diagnosis method based on torque comparison principle |
| CN105004962A (en) * | 2015-03-24 | 2015-10-28 | 华北电力大学（保定） | Improved method for detecting short trouble of exciting winding of turbonator |
Non-Patent Citations (3)
| Title |
|---|
| Multi-class Support Vector Machine approach for fault classification in power transmission line;V. Malathi 等;《2008 IEEE International Conference on Sustainable Energy Technologies》;20090109;第67-71页 * |
| 光伏发电系统的故障诊断与健康监测技术研究;王凌霄;《中国优秀博硕士学位论文全文数据库(硕士)工程科技Ⅱ辑》;20180615(第06期);第C042-384页 * |
| 含分布式电源的智能配电网的故障自愈技术研究;李大鹏;《中国优秀博硕士学位论文全文数据库(硕士)工程科技Ⅱ辑》;20120815(第08期);第C042-141页 * |
Also Published As
| Publication number | Publication date |
|---|---|
| CN109470946A (en) | 2019-03-15 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN109470946B (en) | Power generation equipment fault detection method and system | |
| CN112202736B (en) | Communication network anomaly classification method based on statistical learning and deep learning | |
| CN108375715B (en) | Power distribution network line fault risk day prediction method and system | |
| US7107187B1 (en) | Method for modeling system performance | |
| JP2021064370A (en) | Method and system for semi-supervised deep abnormality detection for large-scale industrial monitoring system based on time-series data utilizing digital twin simulation data | |
| CN111459700A (en) | Method and apparatus for diagnosing device failure, diagnostic device, and storage medium | |
| CN109840157A (en) | Method, apparatus, electronic equipment and the storage medium of fault diagnosis | |
| KR20180108446A (en) | System and method for management of ict infra | |
| CN104317681A (en) | Behavioral abnormality automatic detection method and behavioral abnormality automatic detection system aiming at computer system | |
| CN111143097B (en) | GNSS positioning service-oriented fault management system and method | |
| CN111061293B (en) | Multi-parameter coupled aircraft fault positioning method, aircraft and storage medium | |
| CN111199346A (en) | Comprehensive management gallery supervision method and device, computer equipment and storage medium | |
| JP2016539425A (en) | Computer-implemented method and system for automatically monitoring and determining the status of all process segments in a process unit | |
| CN114066262A (en) | Method, system and device for estimating cause-tracing reasoning of abnormal indexes after power grid dispatching and storage medium | |
| CN115438726A (en) | Device life and fault type prediction method and system based on digital twin technology | |
| CN111125195B (en) | Data anomaly detection method and device | |
| CN110850837A (en) | System life analysis and fault diagnosis method based on long-time and short-time memory neural network | |
| CN111931323B (en) | Memory, hydrocracking equipment fault prediction method, device and equipment | |
| CN111934903A (en) | Docker container fault intelligent prediction method based on time sequence evolution genes | |
| KR102110319B1 (en) | System for generating learning data | |
| KR101929955B1 (en) | Analysis method of factory automation system using artificial intelligence-based reverse engineering | |
| JP7062505B2 (en) | Equipment management support system | |
| CN116383048A (en) | Software quality information processing method and device | |
| CN115600695A (en) | Fault diagnosis method of metering equipment | |
| CN104751059A (en) | Function template based software behavior analysis method |
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 |