CN118171191A - CPU fault diagnosis method based on EMD decomposition and RF classification - Google Patents
CPU fault diagnosis method based on EMD decomposition and RF classification Download PDFInfo
- Publication number
- CN118171191A CN118171191A CN202311802210.8A CN202311802210A CN118171191A CN 118171191 A CN118171191 A CN 118171191A CN 202311802210 A CN202311802210 A CN 202311802210A CN 118171191 A CN118171191 A CN 118171191A
- Authority
- CN
- China
- Prior art keywords
- cpu
- gas turbine
- imfs
- values
- fault diagnosis
- 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
- 238000000034 method Methods 0.000 title claims abstract description 47
- 238000003745 diagnosis Methods 0.000 title claims abstract description 31
- 238000000354 decomposition reaction Methods 0.000 title claims abstract description 21
- 238000004422 calculation algorithm Methods 0.000 claims abstract description 18
- 230000008569 process Effects 0.000 claims abstract description 18
- 238000007781 pre-processing Methods 0.000 claims abstract description 8
- 230000002159 abnormal effect Effects 0.000 claims description 9
- 238000007637 random forest analysis Methods 0.000 claims description 7
- 238000012545 processing Methods 0.000 abstract description 4
- 239000000284 extract Substances 0.000 abstract 1
- 239000007789 gas Substances 0.000 description 34
- 238000012423 maintenance Methods 0.000 description 10
- 230000006870 function Effects 0.000 description 9
- 238000010248 power generation Methods 0.000 description 3
- 230000008439 repair process Effects 0.000 description 3
- 238000011161 development Methods 0.000 description 2
- 230000018109 developmental process Effects 0.000 description 2
- 238000005516 engineering process Methods 0.000 description 2
- 239000000446 fuel Substances 0.000 description 2
- VNWKTOKETHGBQD-UHFFFAOYSA-N methane Chemical compound C VNWKTOKETHGBQD-UHFFFAOYSA-N 0.000 description 2
- 238000004088 simulation Methods 0.000 description 2
- UFHFLCQGNIYNRP-UHFFFAOYSA-N Hydrogen Chemical compound [H][H] UFHFLCQGNIYNRP-UHFFFAOYSA-N 0.000 description 1
- 230000005856 abnormality Effects 0.000 description 1
- 230000032683 aging Effects 0.000 description 1
- 230000009286 beneficial effect Effects 0.000 description 1
- 238000012512 characterization method Methods 0.000 description 1
- 238000002485 combustion reaction Methods 0.000 description 1
- 238000003066 decision tree Methods 0.000 description 1
- 238000013135 deep learning Methods 0.000 description 1
- 238000001514 detection method Methods 0.000 description 1
- 230000005611 electricity Effects 0.000 description 1
- 230000007613 environmental effect Effects 0.000 description 1
- 239000001257 hydrogen Substances 0.000 description 1
- 229910052739 hydrogen Inorganic materials 0.000 description 1
- 230000006698 induction Effects 0.000 description 1
- 238000007689 inspection Methods 0.000 description 1
- 230000007246 mechanism Effects 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 238000012544 monitoring process Methods 0.000 description 1
- 239000003345 natural gas Substances 0.000 description 1
- 238000005312 nonlinear dynamic Methods 0.000 description 1
- 230000000737 periodic effect Effects 0.000 description 1
- 238000011084 recovery Methods 0.000 description 1
- 238000012360 testing method Methods 0.000 description 1
- 239000002699 waste material Substances 0.000 description 1
Classifications
-
- Y—GENERAL TAGGING OF NEW TECHNOLOGICAL DEVELOPMENTS; GENERAL TAGGING OF CROSS-SECTIONAL TECHNOLOGIES SPANNING OVER SEVERAL SECTIONS OF THE IPC; TECHNICAL SUBJECTS COVERED BY FORMER USPC CROSS-REFERENCE ART COLLECTIONS [XRACs] AND DIGESTS
- Y02—TECHNOLOGIES OR APPLICATIONS FOR MITIGATION OR ADAPTATION AGAINST CLIMATE CHANGE
- Y02P—CLIMATE CHANGE MITIGATION TECHNOLOGIES IN THE PRODUCTION OR PROCESSING OF GOODS
- Y02P90/00—Enabling technologies with a potential contribution to greenhouse gas [GHG] emissions mitigation
- Y02P90/02—Total factory control, e.g. smart factories, flexible manufacturing systems [FMS] or integrated manufacturing systems [IMS]
Landscapes
- Testing And Monitoring For Control Systems (AREA)
Abstract
The invention provides a CPU fault diagnosis method based on EMD decomposition and RF classification, which comprises the following steps: acquiring an original process signal of the operation of a gas turbine sensor; preprocessing the signals to form an original data set; decomposing the original data set by using an EMD decomposition algorithm, and carrying out reconstruction processing on each component by using the similarity of each component fluctuation and the approximation degree of the sample entropy value to obtain a new sequence; and respectively extracting variances, average values, maximum values and minimum values of the reconstructed new sequences as characteristic values, combining the reconstructed IMFs component characteristic values as characteristic values of a CPU (central processing unit) of the gas turbine control system by adopting a weighted summation mode, and carrying out state identification and classification on the characteristic values by using an RF algorithm to identify the state of the CPU. The method uses an EMD algorithm to reconstruct the original process signals of the sensor, extracts characteristic values from reconstructed data, uses an RF algorithm to identify fault states, can rapidly identify CPU faults, and provides an engineering solution for the diagnosis of the gas turbine CPU faults.
Description
Technical Field
The invention relates to the technical field of fault diagnosis of gas turbine power plants, in particular to a CPU fault diagnosis method based on EMD decomposition and RF classification.
Background
With the increasing maturity of gas turbines and their combined cycle technologies, and the tremendous exploitation of natural gas resources worldwide and the increasing severity of global environmental pressures, gas turbine power generation is used not only as an emergency backup power source and peak load, but also as a clean energy source, distributed energy source and base load to deliver power to the grid. To date, gas turbines are continually growing and evolving in the power generation field; gas turbines using hydrogen as a fuel generate electricity, receiving unprecedented attention and great development.
The gas turbine control system is a complex nonlinear dynamic system formed by integrating a large number of parts according to a certain mode, function and requirement. The electronic controller (DCS system) in the gas turbine control system is the core and key of the whole gas turbine control system. How to effectively diagnose faults of an electronic controller of a fuel engine control system is always a difficult problem in the industry. In the traditional electronic controller fault diagnosis method based on BIT technology, a large number of redundant diagnosis circuits are needed to be added in the electronic component, so that the system cost is increased, and a new electronic fault point is also added. With the development of fault diagnosis methods based on data, particularly the rise of deep learning algorithms, fault diagnosis based on process signals is performed on a sensor with extremely high precision. This provides a basis for process signal based electronic controller fault simulation and fault diagnosis.
The manufacturers of different gas engine control systems have built in a plurality of professional system diagnosis functions in respective DCS systems, the functions can be seen only in engineer operation stations with corresponding authorities, the information acquisition process is passive, operators are required to find various alarm information from different positions, and then various drawing data of paper edition are combined, and even hardware indication states are required to be actually checked in a DCS cabinet, so that fault points of the system can be judged. Further judgment of the cause of the failure is more of the engineer's personal ability and experience. This severely restricts the quick judgment of the system failure in the gas turbine power plant, and as a result, unplanned shutdown caused by false alarms of the electronic controller system occurs frequently, which causes great economic loss to the power plant.
Currently, when industrial equipment of a factory is maintained, a mode of 'preventing mainly, planning maintenance mainly and temporarily repairing mainly' is adopted. Specifically, the factory is set to be in a period of three to five years according to the new and old conditions of industrial equipment, all industrial equipment in the factory is subjected to primary overhaul, namely, the operations of shutdown, disassembly, maintenance and refitting are carried out on all industrial equipment, whether certain industrial equipment needs to be maintained or not is not considered, the primary overhaul is generally carried out for more than two months, more than one hundred of professional maintenance personnel are needed, and the sum of all kinds of constructors is nearly thousands. During the two overhauls, the industrial equipment is overhauled once every one to two years, namely, most industrial equipment except auxiliary working equipment in a factory, such as a turbo generator set and other large core equipment in the power plant is disassembled, maintained and reinstalled, and the period is generally tens of days to two months. Meanwhile, the factory is subjected to 'scheduled maintenance' and is assisted with 'temporary rush repair', namely, the industrial equipment which fails to operate due to failure is subjected to temporary rush repair.
However, this maintenance of industrial equipment in a factory has two disadvantages: firstly, obviously, the maintenance mode of the industrial equipment can generate great resource waste, and as the scheduled maintenance completely ignores the specific running condition of each specific industrial equipment, all industrial equipment in a factory is treated by adopting a uniform maintenance mode, unnecessary huge investment is generated in personnel, time and economy; second, this type of maintenance of industrial equipment can cause accelerated failure and aging of some of the industrial equipment, which cannot guarantee reliable and smooth operation of each industrial equipment in the factory during two overhauls, and also requires assistance for temporary first-aid repair.
In the power generation industry of the gas turbine, the working mode of routine maintenance is not suitable for an electronic controller system represented by DCS. The DCS system has high-precision and high-density electronic components/chips, the operating principles and failure mechanisms of various components are completely different, and the failure modes are often 'abrupt', so that the problem of fault detection cannot be effectively solved only by means of daily inspection and test. In the daily operation process of the gas turbine, the temperature of the combustion chamber is the most important control monitoring parameter, and the stability and consistency of the group of parameters (the 9F unit is 31 measuring points which are circumferentially arranged) are directly related to whether the unit can continue to operate; once the measured value is distorted due to the CPU system failure, an immediate shutdown may be required, with serious consequences.
Disclosure of Invention
The invention aims to provide a CPU fault diagnosis method based on EMD decomposition and RF classification to solve the problem of CPU system fault diagnosis of an electronic controller in a gas turbine control system.
In order to achieve the above purpose, the technical scheme adopted by the invention is as follows:
the invention provides a CPU fault diagnosis method based on EMD decomposition and RF classification, which comprises the following steps:
step A, acquiring a gas turbine sensor process signal from a gas turbine electronic controller system, and performing data preprocessing to form an original normal data set;
Step B, decomposing the preprocessed original normal data set by using an EMD algorithm to obtain the IMFs component of the sensor;
step C, reconstructing each component by using the similarity of IMFs component fluctuation and the approximation degree of the sample entropy value to obtain a new sequence, and respectively extracting variances, average values, maximum values and minimum values of the reconstructed new sequence as characteristic values;
and D, combining the reconstructed IMFs component characteristic values as characteristic values of the electronic controller CPU of the gas turbine control system by adopting a weighted summation mode, carrying out state identification and classification on the characteristic values of the electronic controller CPU of the gas turbine control system by using an RF random forest, identifying the state of the CPU, and completing fault state identification.
Optionally, in said step a, real-time process signals of the sensors associated with the CPU are acquired from the gas turbine electronic controller DCS using OPC UA protocol.
Optionally, in the step a, the data preprocessing includes redundant sample removing operation, abnormal sample removing operation, and abnormal working condition removing data of the gas turbine.
Optionally, in the step B, decomposing the preprocessed raw normal data set by using an EMD algorithm to obtain sensor IMFs components, including:
Step B1, finding all local maximum and minimum points in a signal x (t), wherein t represents time;
step B2, connecting the maximum value and the minimum value by using spline interpolation or other interpolation methods to obtain an upper envelope u (t) and a lower envelope l (t);
Step B3, calculating an average value
Step B4, extracting an eigenmode function (IMF), comprising:
Step B41, calculating h (t) =x (t) -m (t);
Step B42, if h (t) is a one-dimensional extreme point sequence and the number of extreme points is less than 2 than the signal point, h (t) is the first eigenmode function (IMF 1) of the signal;
Step B42, if h (t) does not meet the above condition, taking h (t) as a new signal, repeating the above steps until the condition is met;
Step B5, extracting the remaining part, which includes:
Step B51, subtracting the first eigenmode function IMF 1 from the original signal to obtain a new signal x 1(t)＝x(t)-IMF1 (t);
Step B52, repeating steps B1 to B4 for a new signal x 1 (t) until the stop condition is satisfied, the number of IMFs reaching a predetermined threshold.
Optionally, in the step C, the components are reconstructed by using the similarity of the IMFs component fluctuation and the approximation degree of the sample entropy value, and the reconstructing is implemented by calculating the correlation of the IMFs components:
two IMFs components are provided: IMFi and IMFj;
respectively calculate the average value Standard deviation: sigma i,σj; covariance cov ij;
Correlation coefficient r:
The range of values of the correlation coefficient r is [ -1,1], and the closer to 1, the closer the fluctuations of the two IMFs components are.
Optionally, when reconstructing each component by calculating the correlation of IMFs components, the preset correlation coefficient range is [0.75,1].
The beneficial effects of the invention include:
the CPU fault diagnosis method based on EMD decomposition and RF classification provided by the invention comprises the following steps: acquiring an original process signal of the operation of a gas turbine sensor from a gas turbine electronic controller system; performing data preprocessing on the original process signals to remove redundant data, abnormal data and abnormal working condition data; decomposing the original data set into IMFs components by using an EMD decomposition algorithm, and carrying out reconstruction processing on each component by using the similarity of each component fluctuation and the approximation degree of a sample entropy value to obtain a new sequence; and respectively extracting variances, average values, maximum values and minimum values of the reconstructed new sequences as characteristic values, combining the reconstructed IMFs component characteristic values as characteristic values of a CPU (central processing unit) of the gas turbine control system by adopting a weighted summation mode, and carrying out state identification and classification on the characteristic values by using an RF algorithm to identify the state of the CPU. According to the method, an EMD algorithm is used for reconstructing an original process signal of the sensor, characteristic values of different IMFs components are calculated, an RF random forest algorithm is used for fault state identification, CPU faults can be identified rapidly, and an engineering solution is provided for gas turbine CPU fault diagnosis.
Drawings
In order to more clearly illustrate the technical solutions of the embodiments of the present invention, the drawings that are needed in the embodiments or the description of the prior art will be briefly introduced below, and it is obvious that the drawings in the following description are only some embodiments of the present invention, and that other drawings can be obtained according to these drawings without inventive effort for a person skilled in the art.
Fig. 1 shows a flow chart of a CPU fault diagnosis method based on EMD decomposition and RF classification according to an embodiment of the present invention.
Fig. 2 shows a flow chart of actual operation of the CPU fault diagnosis method based on EMD decomposition and RF classification provided by the embodiment example of the present invention.
Detailed Description
The following description of the embodiments of the present invention will be made clearly and completely with reference to the accompanying drawings, in which it is apparent that the embodiments described are only some embodiments of the present invention, but not all embodiments. All other embodiments, which can be made by those skilled in the art based on the embodiments of the invention without making any inventive effort, are intended to be within the scope of the invention.
Fig. 1 is a schematic flow chart of a CPU fault diagnosis method based on EMD decomposition and RF classification according to an embodiment of the present invention, and as shown in fig. 1, the present invention provides a CPU fault diagnosis method based on EMD decomposition and RF classification, which includes:
step A, acquiring a gas turbine sensor process signal from a gas turbine electronic controller system, and performing data preprocessing to form an original normal data set;
Step B, decomposing the preprocessed original normal data set by using an EMD algorithm to obtain the IMFs component of the sensor;
step C, reconstructing each component by using the similarity of IMFs component fluctuation and the approximation degree of the sample entropy value to obtain a new sequence, and respectively extracting variances, average values, maximum values and minimum values of the reconstructed new sequence as characteristic values;
and D, combining the reconstructed IMFs component characteristic values as characteristic values of the electronic controller CPU of the gas turbine control system by adopting a weighted summation mode, carrying out state identification and classification on the characteristic values of the electronic controller CPU of the gas turbine control system by using an RF random forest, identifying the state of the CPU, and completing fault state identification.
After a CPU of an electronic controller of a gas turbine control system fails, the state of sensor signals acquired by the CPU is abnormal synchronously, and the CPU fault diagnosis method based on EMD decomposition and RF classification provided by the invention uses an EMD algorithm to decompose the original process signals of the sensor, calculates the characteristic values of different IMFs components, reconstructs the components by using the similarity of the fluctuation of the components and the approximation degree of the entropy value of a sample, uses an RF random forest algorithm to perform fault state identification, can rapidly identify the CPU fault, and provides an engineering solution for the CPU fault diagnosis of the gas turbine.
Optionally, in step a, acquiring raw process signals of gas turbine sensors from the gas turbine electronic controller using OPC UA industrial protocol, and acquiring real-time process data of all sensors associated with the CPU system from the gas turbine electronic controller DCS system; the data preprocessing comprises redundant sample removing operation, abnormal sample removing operation and data of abnormal working conditions of the gas turbine.
Optionally, step B includes decomposing the preprocessed data with EMD to obtain IMFs components, including:
for signal x (t), where t represents time.
(1) All local maxima and minima points are found in the signal x (t).
(2) Connecting the maxima and minima using spline interpolation or other interpolation methods to obtain an upper envelope u (t) and a lower envelope l (t).
(3) Calculating the average value
(4) Extracting an eigenmode function (IMF):
1) H (t) =x (t) -m (t) is calculated.
2) If h (t) is a one-dimensional sequence of extreme points and the number of extreme points is less than 2 than the signal point, then h (t) is the first eigenmode function (IMF 1) of the signal.
3) If h (t) does not meet the above condition, repeating the above steps with h (t) as a new signal until the condition is met.
(5) Extracting the rest:
1) The first eigenmode function IMF1 is subtracted from the original signal to obtain a new signal x1 (t) =x (t) -IMF1 (t).
2) Steps 1 to 4 are repeated for a new signal x1 (t) until the stop condition is met and the number of IMFs reaches a predetermined threshold.
Optionally, step C includes: reconstructing each component by using the similarity of IMFs component fluctuation and the approximation degree of the sample entropy value, and calculating the correlation of the IMFs:
two IMFs components are provided: IMFi and IMFj. Respectively calculate the average value Standard deviation: sigma i,σj; covariance cov ij.
Correlation coefficient r:
The range of values of the correlation coefficient r is [ -1,1], and the closer to 1, the closer the fluctuations of the two IMFs components are.
Optionally, the preset correlation coefficient range is [0.75,1].
Alternatively, variance, average, maximum, minimum are extracted as eigenvalues using the reconstructed IMFs sequence.
CPU state identification is performed using a random forest RF. Random forests are an integrated learning method that improves overall performance by building multiple decision trees and integrating them.
For classification problems, information gain is used to measure node unreliability. Given node t and its child node t i, the information gain IG (t) is calculated as:
Where H (t) is the entropy of node t, ni is the number of samples N in child node t i, and the total number of samples in node t.
Fig. 2 shows a practical operation flowchart of a CPU fault diagnosis method based on EMD decomposition and RF classification according to an embodiment of the present invention.
In summary, the invention provides a convenient and feasible CPU system fault simulation method. Based on the fault of the CPU system of the electronic controller, the state of the sensor signal collected by the CPU is abnormal, such as abrupt jump (step fault) of the numerical value, normal recovery (pulse fault) after abrupt jump of the numerical value, continuous increase or decrease of the numerical value (time-varying/temperature-drifting fault), periodic interference in the numerical value (often because of strong electric induction signal serial in the board card), or poor signal stability (white noise interference, often caused by poor contact or grounding abnormality). According to the invention, an EMD algorithm is used for decomposing an original process signal of the sensor, each component is reconstructed by utilizing the similarity of each component fluctuation and the approximation degree of a sample entropy value, the characteristic value is extracted from the reconstructed data, the characteristic values are weighted and summed to obtain a CPU fault characterization characteristic value, and an RF algorithm is used for fault state identification, so that the CPU fault can be rapidly identified, and an engineering solution is provided for the gas turbine CPU fault diagnosis.
The above embodiments are only for illustrating the technical concept and features of the present invention, and are intended to enable those skilled in the art to understand the content of the present invention and implement the same, but not limit the scope of the present invention, and all equivalent changes or modifications made according to the spirit of the present invention should be included in the scope of the present invention.
Claims (6)
1. A CPU fault diagnosis method based on EMD decomposition and RF classification, the method comprising:
step A, acquiring a gas turbine sensor process signal from a gas turbine electronic controller system, and performing data preprocessing to form an original normal data set;
Step B, decomposing the preprocessed original normal data set by using an EMD algorithm to obtain the IMFs component of the sensor;
step C, reconstructing each component by using the similarity of IMFs component fluctuation and the approximation degree of the sample entropy value to obtain a new sequence, and respectively extracting variances, average values, maximum values and minimum values of the reconstructed new sequence as characteristic values;
and D, combining the reconstructed IMFs component characteristic values as characteristic values of the electronic controller CPU of the gas turbine control system by adopting a weighted summation mode, carrying out state identification and classification on the characteristic values of the electronic controller CPU of the gas turbine control system by using an RF random forest, identifying the state of the CPU, and completing fault state identification.
2. The CPU fault diagnosis method based on EMD decomposition and RF classification according to claim 1, wherein in said step a, the OPC UA protocol is used to obtain real-time process signals of sensors associated with the CPU from the gas turbine electronic controller DCS.
3. The CPU fault diagnosis method based on EMD decomposition and RF classification according to claim 2, wherein in said step a, the data preprocessing includes redundant sample removing operation, abnormal condition removing data of the gas turbine.
4. The CPU fault diagnosis method based on EMD decomposition and RF classification according to claim 1, wherein in the step B, decomposing the preprocessed raw normal data set using an EMD algorithm, obtaining sensor IMFs components, comprises:
Step B1, finding all local maximum and minimum points in a signal x (t), wherein t represents time;
step B2, connecting the maximum value and the minimum value by using spline interpolation or other interpolation methods to obtain an upper envelope u (t) and a lower envelope l (t);
Step B3, calculating an average value
Step B4, extracting an eigenmode function (IMF), comprising:
Step B41, calculating h (t) =x (t) -m (t);
Step B42, if h (t) is a one-dimensional extreme point sequence and the number of extreme points is less than 2 than the signal point, h (t) is the first eigenmode function (IMF 1) of the signal;
Step B42, if h (t) does not meet the above condition, taking h (t) as a new signal, repeating the above steps until the condition is met;
Step B5, extracting the remaining part, which includes:
Step B51, subtracting the first eigenmode function IMF 1 from the original signal to obtain a new signal x 1(t)＝x(t)-IMF1 (t);
Step B52, repeating steps B1 to B4 for a new signal x 1 (t) until the stop condition is satisfied, the number of IMFs reaching a predetermined threshold.
5. The CPU fault diagnosis method based on EMD decomposition and RF classification according to claim 1, wherein in said step C, the components are reconstructed using the similarity of IMFs component fluctuation and the approximation of the sample entropy value, by calculating the correlation of IMFs components:
two IMFs components are provided: IMFi and IMFj;
respectively calculate the average value Standard deviation: sigma i,σj; covariance cov ij;
Correlation coefficient r:
The range of values of the correlation coefficient r is [ -1,1], and the closer to 1, the closer the fluctuations of the two IMFs components are.
6. The CPU fault diagnosis method based on EMD decomposition and RF classification according to claim 5, wherein the predetermined correlation coefficient range is [0.75,1] when reconstructing each component by calculating the correlation of IMFs components.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202311802210.8A CN118171191A (en) | 2023-12-25 | 2023-12-25 | CPU fault diagnosis method based on EMD decomposition and RF classification |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202311802210.8A CN118171191A (en) | 2023-12-25 | 2023-12-25 | CPU fault diagnosis method based on EMD decomposition and RF classification |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN118171191A true CN118171191A (en) | 2024-06-11 |
Family
ID=91355405
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202311802210.8A Pending CN118171191A (en) | 2023-12-25 | 2023-12-25 | CPU fault diagnosis method based on EMD decomposition and RF classification |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN118171191A (en) |
-
2023
- 2023-12-25 CN CN202311802210.8A patent/CN118171191A/en active Pending
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| Yang et al. | Wind turbine gearbox failure detection based on SCADA data: A deep learning-based approach | |
| Liu et al. | A condition monitoring and fault isolation system for wind turbine based on SCADA data | |
| CN110259646B (en) | Wind generating set component state early warning method based on historical data | |
| CN111159844B (en) | Abnormity detection method for exhaust temperature of gas turbine of power station | |
| Hsu et al. | An adaptive forecast-based chart for non-Gaussian processes monitoring: with application to equipment malfunctions detection in a thermal power plant | |
| CN110208647B (en) | Power distribution network single-phase earth fault section positioning method based on random matrix | |
| CN109655266A (en) | A kind of Wind turbines Method for Bearing Fault Diagnosis based on AVMD and spectral coherence analysis | |
| Tao et al. | Abnormal detection of wind turbine based on SCADA data mining | |
| CN111999591B (en) | Method for identifying abnormal state of primary equipment of power distribution network | |
| CN114519382A (en) | Nuclear power plant key operation parameter extraction and abnormity monitoring method | |
| Cao et al. | Remaining useful life prediction of wind turbine generator bearing based on EMD with an indicator | |
| CN118171191A (en) | CPU fault diagnosis method based on EMD decomposition and RF classification | |
| CN115600695B (en) | Fault diagnosis method for metering equipment | |
| Li et al. | Condition monitoring of sensors in a NPP using optimized PCA | |
| CN115167366A (en) | Board card fault diagnosis method based on weighted square sum of sensor signal residual sequence | |
| Fumagalli et al. | Agile diagnostic tool based on electrical signature analysis | |
| Sivakumar et al. | Implementation of VLSI model as a tool in diagnostics of slowly varying process parameters which affect the performance of steam turbine | |
| CN117687390A (en) | Thermocouple bus network fault diagnosis method for electronic controller of gas turbine | |
| Han et al. | An online calculation method of theoretical power losses for high-voltage distribution system based on rapid modeling and data quality analysis | |
| CN117930818A (en) | Board fault diagnosis method based on wavelet band subdivision and multi-level clustering | |
| CN115167181A (en) | Sensor signal processing method for fault simulation of gas turbine electronic controller board card | |
| CN115167365A (en) | Board card fault diagnosis method based on sensor signal wavelet energy spectrum entropy calculation | |
| CN117908520A (en) | Board card fault diagnosis method based on sensor signal autoregressive moving average model | |
| Saucedo-Dorantes et al. | Novelty Detection Methodology Based on Self-Organizing Maps for Power Quality Monitoring | |
| CN117951633B (en) | Photovoltaic power generation equipment fault diagnosis method and system |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination |