CN116089897A - Power equipment operation state evaluation method and system based on multi-source information fusion - Google Patents
Power equipment operation state evaluation method and system based on multi-source information fusion Download PDFInfo
- Publication number
- CN116089897A CN116089897A CN202211488457.2A CN202211488457A CN116089897A CN 116089897 A CN116089897 A CN 116089897A CN 202211488457 A CN202211488457 A CN 202211488457A CN 116089897 A CN116089897 A CN 116089897A
- Authority
- CN
- China
- Prior art keywords
- weight
- matrix
- power equipment
- signal
- calculating
- 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
- 238000011156 evaluation Methods 0.000 title claims abstract description 61
- 230000004927 fusion Effects 0.000 title claims abstract description 19
- 239000011159 matrix material Substances 0.000 claims abstract description 70
- 238000000034 method Methods 0.000 claims abstract description 25
- 238000007781 pre-processing Methods 0.000 claims abstract description 8
- 238000000354 decomposition reaction Methods 0.000 claims description 14
- 238000000605 extraction Methods 0.000 claims description 12
- 238000004364 calculation method Methods 0.000 claims description 9
- 238000012545 processing Methods 0.000 claims description 9
- 230000005484 gravity Effects 0.000 claims description 3
- 238000004140 cleaning Methods 0.000 claims description 2
- 238000011160 research Methods 0.000 abstract description 4
- 238000012544 monitoring process Methods 0.000 abstract description 2
- 230000009466 transformation Effects 0.000 description 4
- 238000004458 analytical method Methods 0.000 description 3
- 230000008859 change Effects 0.000 description 3
- 230000008901 benefit Effects 0.000 description 2
- 230000005540 biological transmission Effects 0.000 description 2
- 238000010586 diagram Methods 0.000 description 2
- 230000000694 effects Effects 0.000 description 2
- 230000009467 reduction Effects 0.000 description 2
- 230000002159 abnormal effect Effects 0.000 description 1
- 230000003044 adaptive effect Effects 0.000 description 1
- 230000002411 adverse Effects 0.000 description 1
- 230000032683 aging Effects 0.000 description 1
- 230000015556 catabolic process Effects 0.000 description 1
- 238000006243 chemical reaction Methods 0.000 description 1
- 238000006731 degradation reaction Methods 0.000 description 1
- 230000003993 interaction Effects 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 230000008569 process Effects 0.000 description 1
- 230000005855 radiation Effects 0.000 description 1
- 238000007873 sieving Methods 0.000 description 1
- 230000000007 visual effect Effects 0.000 description 1
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
- G06Q50/00—Information and communication technology [ICT] specially adapted for implementation of business processes of specific business sectors, e.g. utilities or tourism
- G06Q50/06—Energy or water supply
-
- 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
- Y04—INFORMATION OR COMMUNICATION TECHNOLOGIES HAVING AN IMPACT ON OTHER TECHNOLOGY AREAS
- Y04S—SYSTEMS INTEGRATING TECHNOLOGIES RELATED TO POWER NETWORK OPERATION, COMMUNICATION OR INFORMATION TECHNOLOGIES FOR IMPROVING THE ELECTRICAL POWER GENERATION, TRANSMISSION, DISTRIBUTION, MANAGEMENT OR USAGE, i.e. SMART GRIDS
- Y04S10/00—Systems supporting electrical power generation, transmission or distribution
- Y04S10/50—Systems or methods supporting the power network operation or management, involving a certain degree of interaction with the load-side end user applications
Landscapes
- Business, Economics & Management (AREA)
- Health & Medical Sciences (AREA)
- Engineering & Computer Science (AREA)
- Economics (AREA)
- Public Health (AREA)
- Water Supply & Treatment (AREA)
- General Health & Medical Sciences (AREA)
- Human Resources & Organizations (AREA)
- Marketing (AREA)
- Primary Health Care (AREA)
- Strategic Management (AREA)
- Tourism & Hospitality (AREA)
- Physics & Mathematics (AREA)
- General Business, Economics & Management (AREA)
- General Physics & Mathematics (AREA)
- Theoretical Computer Science (AREA)
- Management, Administration, Business Operations System, And Electronic Commerce (AREA)
Abstract
The invention belongs to the technical field of power equipment operation evaluation, and particularly provides a power equipment operation state evaluation method based on multi-source information fusion. S1, preprocessing running state data of power equipment, and extracting features to form a feature set; s2, calculating a membership matrix of the feature set by using a differential algorithm as subjective weight; s3, calculating the combined assigned weight of the feature set by adopting an AHP algorithm and a gray correlation algorithm to serve as a constant weight coefficient, and then calculating a variable weight coefficient to serve as an objective weight; s4, carrying out linear weighted combination on the subjective weight and the objective weight to construct a fuzzy matrix; s5: and judging the fuzzy matrix by adopting a fuzzy comprehensive judging method, and outputting a judging result of the running state of the power equipment. The technical scheme is used for solving the key technical research of establishing the dynamic model of the running state of the power equipment, realizing accurate evaluation of the running state of the power equipment, being applicable to a power monitoring system and establishing demonstration application in the power industry and province by representative industrial users.
Description
Technical Field
The invention belongs to the technical field of power equipment operation evaluation, and particularly provides a power equipment operation state evaluation method and system based on multi-source information fusion.
Background
The state evaluation of the power transformation and distribution equipment is subjected to the randomness constraint of various natural environment factors and internal degradation or aging factors, so that the theoretical modeling is quite complex, the factors influencing the evaluation result are more, the influence degree is different, and a certain difficulty exists in realizing the accurate evaluation of the state of the power transformation and distribution equipment. Aiming at the current research situation at home and abroad, the project realizes the rationality of the state grading strategy of the power transmission and transformation equipment by adopting an advanced algorithm in combination with an expert database, and provides a guarantee for the accurate evaluation of the power transmission and transformation and distribution equipment.
Disclosure of Invention
In order to solve the above problems, the present invention provides a power equipment operation state evaluation method based on multi-source information fusion, comprising,
s1, preprocessing running state data of power equipment, and extracting features to form a feature set;
s2, calculating a membership matrix of the feature set by using a differential algorithm as subjective weight;
s3, calculating the combined assigned weight of the feature set by adopting an AHP algorithm and a gray correlation algorithm to serve as a constant weight coefficient, and then calculating a variable weight coefficient to serve as an objective weight;
s4, carrying out linear weighted combination on the subjective weight and the objective weight to construct a fuzzy matrix;
s5: and judging the fuzzy matrix by adopting a fuzzy comprehensive judging method, and outputting a judging result of the running state of the power equipment.
Further, in step S1, feature extraction of the running state data includes cleaning the collected running state data, identifying illegal values therein, and data items of the error source, legalizing the illegal values, and performing simple feature extraction on the cleaned data to add the illegal values to the feature set.
Further, the extracted features include time domain features and frequency domain features of the data.
Further, in step S1, feature extraction of the running state data further includes dividing the running state data into two parts of a high-frequency signal and a low-frequency signal by using wavelet packet decomposition, and then performing EMD decomposition on the two signals divided in the narrow band to obtain feature components of the IMF linear steady-state signal, and inputting the feature components into the feature set.
Further, the signal acquisition refers to the original data provided by the running state of the electric power equipment as an information source, the EMD decomposition comprises the following steps,
1) Analyzing an original signal X (t), connecting all maximum value points and minimum value points on the X (t) by adopting a cubic spline interpolation curve so as to form an upper envelope line and a lower envelope line, and taking the average value of the X (t) and the upper envelope line and the lower envelope line as m 1 The two are subtracted, and the difference is:
h 1 ＝X(t)-m 1
handle h 1 Repeating the above steps as a new X (t) signal until h i When two conditions of IMF are satisfied, it becomes a first order linear steady state signal IMF selected from X (t), denoted as C 1 In general C 1 The linear steady state signal IMF contains the highest frequency component of the signal, and two conditions are satisfied simultaneously: the number of extreme points of the obtained difference signal h is consistent with that of zero crossing points, or the difference absolute value is smaller than 1; and the second is: the local waveform of the obtained difference signal h is symmetrical;
2) C is C 1 Separating from the original signal X (t) to be measured to obtain an interpolated signal r from which high frequency components have been removed 1 The method comprises the following steps:
r 1 ＝X(t)-C 1
handle r 1 Regarded as a new X (t) signal, and the steps are repeatedStep 1), until the residual signal of the nth order accords with a monotonic function, and the IMF component is not screened, namely:
r n ＝r n-1 -C n
3) The original signal to be measured X (t) consists of n IMF components and a residual term, namely expressed as:
wherein r is n (t) -residuals, representing the average trend in the signal; each IMF component C j And (t) represents the components of different frequency bands from high to low of X (t), respectively.
Further, in steps S2 and S3, the feature set is normalized by an index to satisfy the subjective and objective weighting conditions.
Further, in the step S3, the final objective weight calculation includes the following steps,
1) Calculating subjective weight of the feature set by adopting an AHP algorithm;
2) Calculating an entropy weight matrix of the feature set by adopting a gray correlation algorithm, and taking the entropy weight matrix as an objective weight;
3) And (3) carrying out linear combination on the principal and objective weights in the step (1) and the step (2) to obtain a weighted weight serving as a constant weight coefficient, and introducing a variable weight formula based on the constant weight coefficient to calculate to obtain a variable weight coefficient serving as a final objective weight.
Further, the weight change formula is specifically that,
in the middle of
Variable weight coefficient for the ith characteristic parameter, < ->
Constant as the i-th characteristic parameterWeight coefficient, x i The i-th characteristic parameter is n, which is the characteristic number.
Further, the entropy weight matrix is obtained through the following steps,
1) Constructing a data matrix
If the electric equipment has m running state evaluation indexes, the comprehensive evaluation problem of the n electric equipment is formed, a data matrix X is constructed, and the evaluation object marks the value of the evaluation indexes as X ij (i＝1,2,L,n；j＝1,2,L,m)：
X in matrix ij The value of the j-th evaluation index representing the i-th evaluation object;
2) Data matrix standard forward processing
Normalizing the data matrix, and marking each element in the normalized matrix as z ij ，
3) Calculating index specific gravity of ith object under jth index in matrix
For the calculated non-negative standard matrix Z, the characteristic proportion of the ith evaluation object to the jth evaluation index is recorded as p ij ，
The probability matrix P can be obtained by calculating the feature weights of all the terms,
easy to verify
The probability sum corresponding to each index is guaranteed to be 1;
4) Calculating entropy value of jth index in matrix
For the j-th index in the matrix, the information entropy e ij The calculation formula of (2) is expressed as:
e j the larger the value of (a), i.e. the larger the information entropy of the jth index, the less information indicating the jth index;
5) Calculating the difference coefficient of the jth index in the matrix
The coefficient of difference may also be referred to as the information utility value, which is noted as d j
d j ＝1-e j
If the utility value of the information obtained by calculation is larger, the corresponding information quantity is represented more;
6) Calculating weights
The entropy weight W of each index can be obtained by normalizing the difference coefficient, namely the information utility value j
According to another aspect of the present invention, there is provided an electrical device operation state evaluation system based on multi-source information fusion, including,
the feature extraction unit is used for preprocessing the running state data of the power equipment and extracting features to form a feature set;
the subjective weight generating unit is used for calculating a membership matrix of the feature set by using a differential algorithm as subjective weight;
the objective weight generating unit calculates the combined weighting of the feature set by adopting an AHP algorithm and a gray correlation algorithm as a constant weight coefficient, and then calculates a variable weight coefficient as an objective weight;
the linear weighted combination unit is used for carrying out linear weighted combination on the subjective weight and the objective weight to construct a fuzzy matrix;
and the judging unit is used for judging the fuzzy matrix by adopting a fuzzy comprehensive judging method and outputting the judging result of the running state of the power equipment.
The invention mainly solves the problem of establishing a key technical research on a dynamic model of the running state of the power equipment, and the whole algorithm is the dynamic model from data acquisition to final prediction result. The operation state of the power equipment is accurately evaluated; the method can be applied to a power monitoring system, and establishes demonstration application in power industry and industrial users with representativeness in provinces, gradually forms an provincial demonstration base, and realizes national effects of driving and radiation in the provincial. The related technical achievements of the project are hopeful to promote the electric power safety industry to realize revolutionary technological progress.
Drawings
FIG. 1 is a schematic flow chart of a power equipment operation state evaluation method according to the present invention;
FIG. 2 is a schematic diagram of the EMD algorithm flow;
FIG. 3 is a method for extracting characteristic parameters of the operation state of the power equipment;
fig. 4 is a schematic diagram of the weight coefficient of each feature.
Detailed Description
The present invention will be described in further detail with reference to the drawings and examples, in order to make the objects, technical solutions and advantages of the present invention more apparent. It should be understood that the specific embodiments described herein are for purposes of illustration only and are not intended to limit the scope of the invention.
Referring to fig. 1, the present invention provides a power equipment operation state evaluation method based on multi-source information fusion, including,
step one: preprocessing the running state data of the power equipment, extracting features to form a feature set, and referring to fig. 3, signal acquisition refers to the original data provided by the running state of the power equipment as an information source. The collected data of the power equipment often has the characteristics of multidimensional, massive, high interference (high noise), nonlinearity and the like, and the data is collected, uploaded and stored and then is prepared for processing.
The data preprocessing firstly cleans the collected data, distinguishes data items which affect subsequent analysis such as illegal values, error sources and the like, and adopts a proper method to legalize the data items, such as: null value zero setting, positive and negative conversion, etc. And directly extracting the simple and easily distinguished features after data are cleaned, and adding the features into the feature set. Such as: time domain features, frequency domain features, etc. The wavelet packet decomposition is used for data denoising, and the time-frequency plane is divided more finely, so that the method has higher resolution for high-frequency signals, and therefore, the high-frequency signals and the low-frequency signals can be separated for fine noise treatment respectively.
The running state data is divided into a high-frequency signal and a low-frequency signal by wavelet packet decomposition, EMD decomposition is carried out on the two signals divided in the narrow band respectively to obtain IMF linear steady-state signal characteristic components, and the IMF linear steady-state signal characteristic components are input into a characteristic set.
Referring to fig. 2, the emd decomposition includes the steps of,
(1) Analyzing an original signal X (t), connecting all maximum value points and minimum value points on the X (t) by adopting a cubic spline interpolation curve so as to form an upper envelope line and a lower envelope line, and taking the average value of the X (t) and the upper envelope line and the lower envelope line as m 1 The two are subtracted, and the difference is:
h 1 ＝X(t)-m 1
handle h 1 Repeating the above steps as a new X (t) signal until h i When two conditions of IMF are satisfied, it becomes a first order IMF selected from X (t), denoted as C 1 In general C 1 The linear steady state signal IMF, which contains the highest frequency component of the signal, satisfies two conditions: the number of extreme points of the obtained difference signal h is consistent with that of zero crossing points, or the difference absolute value is smaller than 1; and the second is: the local waveform of the obtained difference signal h is symmetrical;
(2) C is C 1 Separating from the original signal X (t) to be measured to obtain an interpolated signal r from which high frequency components have been removed 1 The method comprises the following steps:
r 1 ＝X(t)＝C 1
handle r 1 And (3) taking the residual signal as a new X (t) signal, repeating the step (1) until the residual signal of the nth order accords with a monotonic function, and sieving the IMF component, namely:
r n ＝r n-1 -C n
(3) In the mathematical field, the original signal to be measured X (t) consists of n IMF components and one residual term, expressed as:
wherein: r is (r) n (t) -residuals, representing the average trend in the signal; each IMF component C j And (t) represents the components of different frequency bands from high to low of X (t), respectively.
The EMD decomposition is also a time-frequency domain analysis method, and compared with the wavelet packet decomposition, the method is more visual and adaptive. After the wavelet packet is decomposed, EMD performs noise processing on the high-frequency and low-frequency mixed population, and the method gradually decomposes noise and source data by multiple iteration and a large amount of calculation.
The frequency components are also different in each frequency segment, the instant frequencies are also different at different moments in the same IMF, and the local time distribution is changed with the signal itself in each frequency segment. When the EMD noise reduction method is used, most of the EMD noise reduction method is to decompose a signal to be detected, find out high-frequency components in the signal to be detected and directly remove the high-frequency components as noise, so that a plurality of useful signal components are removed erroneously.
The mode of combining wavelet packet decomposition and EMD decomposition is used for removing noise, so that the research has the serious difficulty, the anti-interference rate is extremely high, the processed source signal is hardly interfered, and a very good foundation is laid for subsequent feature extraction. The MF component is input into the feature set as the result of EMD processing, and feature fusion is performed. Thus, the task of signal acquisition from the power equipment and finally feature extraction of the signals is completed.
And obtaining a feature set from the result of the power equipment feature extraction technology, and giving weight to each feature to achieve accurate power equipment state evaluation.
Step two: firstly, index standardization is carried out on the feature set to enable the feature set to meet the subjective and objective weighting conditions. And calculating a membership matrix of the feature set by using a differential algorithm as subjective weight. The weight in this scheme is essentially a vector matrix.
Step three: calculating the combined assigned weight of the feature set by adopting an AHP algorithm and a gray correlation algorithm as a constant weight coefficient, and then calculating a variable weight coefficient as an objective weight;
1) Calculating subjective weight of the feature set by adopting an AHP algorithm;
2) Calculating an entropy weight matrix of the feature set by adopting an entropy weight algorithm, and taking the entropy weight matrix as objective weight;
specifically, the method comprises the following steps,
1) Constructing a data matrix
For the comprehensive evaluation problem of n electric equipment with m running state evaluation indexes, constructing a data matrix X, and marking the value of an evaluation object on the evaluation indexes as X ij (i＝1,2,L,n；j＝1,2,L,m)：
X in matrix ij The value of the j-th evaluation index representing the i-th evaluation object;
2) Standard forward processing of data matrix
The entropy weight method is used for processing data, and a method for evaluating the ratio of a certain index of an object to the sum of the same index is adopted, so that adverse effects can be caused on an evaluation result, in order to reduce analysis deviation, a data matrix is subjected to standardization processing, and each element in the standardization matrix is marked as z ij
3) Calculating index specific gravity of ith object under jth index in matrix
According to the information entropy theory, the magnitude of the entropy value can measure the magnitude of the information quantity, the information quantity is increased, the entropy is reduced, and conversely, the information quantity is reduced, and the entropy is increased. For the evaluation index j, z ij The higher the difference degree, the larger the influence degree of the evaluation index on the evaluation object is, and on the contrary, the lower the difference degree is, the smaller the influence degree is;
for the calculated non-negative standard matrix Z, the characteristic proportion of the ith evaluation object to the jth evaluation index is recorded as p ij
The probability matrix P can be obtained by calculating the characteristic weights of all the terms
Easy to verify
The probability sum corresponding to each index is guaranteed to be 1;
4) Calculating entropy value of jth index in matrix
For the j-th index in the matrix, the information entropy e ij The calculation formula of (2) is expressed as:
e j the larger the value of (a), i.e. the larger the information entropy of the jth index, the less information indicating the jth index;
5) Calculating the difference coefficient of the j-th index in the matrix
The coefficient of difference may also be referred to as the information utility value, which is noted as d j
d j ＝1-e j
If the utility value of the information obtained by calculation is larger, the corresponding information quantity is represented more;
6) Calculating weights
The entropy weight W of each index can be obtained by normalizing the difference coefficient, namely the information utility value j
Entropy weight W j The entropy weight matrix is obtained.
3) And (3) carrying out linear combination on the principal and objective weights in the step (1) and the step (2) to obtain a weighted weight as a constant weight coefficient, and introducing a variable weight formula to calculate a variable weight coefficient based on the constant weight coefficient to serve as a final objective weight.
The weight-changing formula is specifically that,
in the method, in the process of the invention,
variable weight coefficient for the ith characteristic parameter, < ->
Chang Quanchong coefficient, x, being the i-th characteristic parameter i The i-th characteristic parameter is n, which is the characteristic number. Under the condition of constant weight, when the state of the electrical equipment is abnormal just at first, most characteristic parameter indexes are probably not changed greatly, and an evaluation index with a certain weight which is particularly small can be seriously deviated from a normal value, and the evaluation result still shows a normal state because the index occupies a smaller weight, which is not consistent with the state of the actual electrical equipment. In view of the disadvantage of the constant weight coefficient, in order to enable the operation state of the electrical equipment to be more real and accurate, a variable weight theory is introduced to adjust the Chang Quanchong coefficient in real time, so that the judging result is more real and accurate, and refer to fig. 4.
The evaluation of the running state of the power equipment involves multiple factors and multiple layers and is a comprehensive result under the interaction of the multiple factors, so that the determination of the index weight is particularly important. The Chang Quanchong coefficient will not change due to the change of the state quantity, when the value of a certain state quantity of the power equipment deviates from the normal value seriously, i.e. the performance of a certain aspect of the power equipment is seriously reduced, if the operation state evaluation is performed only according to the Chang Quanchong coefficient, the whole evaluation result of the power equipment may be normal because the state quantity weight is not large in the whole proportion, thereby causing great potential safety hazard
Step four: linearly weighting and combining the subjective weight and the objective weight to construct a fuzzy matrix;
step five: and judging the fuzzy matrix by adopting a fuzzy comprehensive judging method, and outputting a judging result of the running state of the power equipment.
And judging the running state of the power equipment by adopting a fuzzy comprehensive judging method, wherein the weight is firstly weighted by subjective and objective weighting to obtain a Chang Quanchong coefficient, and then a variable weight formula is introduced to obtain a variable weight coefficient. The fuzzy comprehensive evaluation method is a method for comprehensively evaluating things with various indexes and various factors by using a fuzzy theory, and has the advantage of processing complex networks with multiple factors and dynamics. The membership of the device, that is, which class the current state of the device belongs to, such as good, general, attention, warning, etc., is obtained after the fuzzy relation matrix is evaluated, so that the state of the device is judged.
According to another aspect of the present invention, there is provided an electrical device operation state evaluation system based on multi-source information fusion, including,
the feature extraction unit is used for preprocessing the running state data of the power equipment and extracting features to form a feature set;
the subjective weight generating unit is used for calculating a membership matrix of the feature set by using a differential algorithm as subjective weight;
the objective weight generating unit calculates the combined weighting of the feature set by adopting an AHP algorithm and a gray correlation algorithm as a constant weight coefficient, and then calculates a variable weight coefficient as an objective weight;
the linear weighted combination unit is used for carrying out linear weighted combination on the subjective weight and the objective weight to construct a fuzzy matrix;
and the judging unit is used for judging the fuzzy matrix by adopting a fuzzy comprehensive judging method and outputting the judging result of the running state of the power equipment.
The foregoing description of the preferred embodiments of the invention is not intended to be limiting, but rather is intended to cover all modifications, equivalents, and alternatives falling within the spirit and principles of the invention.
Claims (10)
1. The utility model provides a power equipment running state evaluation method based on multisource information fusion which is characterized in that:
s1, preprocessing running state data of power equipment, and extracting features to form a feature set;
s2, calculating a membership matrix of the feature set by using a differential algorithm as subjective weight;
s3, calculating the combined assigned weight of the feature set by adopting an AHP algorithm and a gray correlation algorithm to serve as a constant weight coefficient, and then calculating a variable weight coefficient to serve as an objective weight;
s4, carrying out linear weighted combination on the subjective weight and the objective weight to construct a fuzzy matrix;
s5: and judging the fuzzy matrix by adopting a fuzzy comprehensive judging method, and outputting a judging result of the running state of the power equipment.
2. The power equipment operation state evaluation method based on multi-source information fusion as claimed in claim 1, wherein: in step S1, feature extraction of the running state data includes cleaning the collected running state data, distinguishing illegal values and data items of error sources, legalizing the illegal values and data items, and performing simple feature extraction on the cleaned data to add the data items into a feature set.
3. The power equipment operation state evaluation method based on multi-source information fusion as claimed in claim 2, wherein: the extracted features include time domain features and frequency domain features of the data.
4. The power equipment operation state evaluation method based on multi-source information fusion as claimed in claim 1, wherein: in step S1, the feature extraction of the running state data further includes dividing the running state data into two parts, namely a high-frequency signal and a low-frequency signal, by using wavelet packet decomposition, and then performing EMD decomposition on the two signals divided in the narrow band to obtain an IMF linear steady-state signal feature component, and inputting the feature component into the feature set.
5. The power equipment operation state evaluation method based on multi-source information fusion according to claim 4, wherein: the signal acquisition refers to the original data provided by the running state of the power equipment as an information source, and the EMD decomposition comprises the following steps,
1) Analyzing an original signal X (t), connecting all maximum value points and minimum value points on the X (t) by adopting a cubic spline interpolation curve so as to form an upper envelope line and a lower envelope line, and taking the average value of the X (t) and the upper envelope line and the lower envelope line as m 1 The two are subtracted, and the difference is:
h 1 ＝X(t)-m 1
handle h 1 Repeating the above steps as a new X (t) signal until h i When two conditions of IMF are satisfied, it becomes a first order linear steady state signal IMF selected from X (t), denoted as C 1 In general C 1 The linear steady state signal IMF contains the highest frequency component of the signal, and two conditions are satisfied simultaneously: the number of extreme points of the obtained difference signal h is consistent with that of zero crossing points, or the difference absolute value is smaller than 1; and the second is: the local waveform of the obtained difference signal h is symmetrical;
2) C is C 1 Separating from the original signal X (t) to be measured to obtain an interpolated signal r from which high frequency components have been removed 1 The method comprises the following steps:
r 1 ＝X(t)-C 1
handle r 1 Regarded as a new X (t) signal, repeatedStep 1), until the residual signal of the nth order accords with a monotonic function, and the IMF component is not screened, namely:
r n ＝r n-1 -C n
3) The original signal to be measured X (t) consists of n IMF components and a residual term, namely expressed as:
wherein r is n (t) -residuals, representing the average trend in the signal; each IMF component C j And (t) represents the components of different frequency bands from high to low of X (t), respectively.
6. The power equipment operation state evaluation method based on multi-source information fusion as claimed in claim 1, wherein: in the steps S2 and S3, index standardization is carried out on the feature set to enable the feature set to meet the subjective and objective weighting conditions.
7. The power equipment operation state evaluation method based on multi-source information fusion as claimed in claim 1, wherein: in the step S3, the final objective weight calculation includes the following steps,
1) Calculating subjective weight of the feature set by adopting an AHP algorithm;
2) Calculating an entropy weight matrix of the feature set by adopting a gray correlation algorithm, and taking the entropy weight matrix as an objective weight;
3) And (3) carrying out linear combination on the principal and objective weights in the step (1) and the step (2) to obtain a weighted weight serving as a constant weight coefficient, and introducing a variable weight formula based on the constant weight coefficient to calculate to obtain a variable weight coefficient serving as a final objective weight.
8. The power equipment operation state evaluation method based on multi-source information fusion according to claim 7, wherein: the weight-changing formula is specifically that,
in the middle of
The variable weight coefficient of the ith characteristic parameter, W i (0) Chang Quanchong coefficient, x, being the i-th characteristic parameter i The i-th characteristic parameter is n, which is the characteristic number.
9. The power equipment operation state evaluation method based on multi-source information fusion according to claim 7, wherein: the entropy weight matrix is obtained through the following steps,
1) Constructing a data matrix
If the electric equipment has m running state evaluation indexes, the comprehensive evaluation problem of the n electric equipment is formed, a data matrix X is constructed, and the evaluation object marks the value of the evaluation indexes as X ij (i＝1,2,L,n；j＝1,2,L,m)：
X in matrix ij The value of the j-th evaluation index representing the i-th evaluation object;
2) Data matrix standard forward processing
Normalizing the data matrix, and marking each element in the normalized matrix as z ij ，
3) Calculating index specific gravity of ith object under jth index in matrix
For the calculated non-negative standard matrix Z, the characteristic proportion of the ith evaluation object to the jth evaluation index is recorded as p ij ，
The probability matrix P can be obtained by calculating the feature weights of all the terms,
easy to verify
The probability sum corresponding to each index is guaranteed to be 1;
4) Calculating entropy value of jth index in matrix
For the j-th index in the matrix, the information entropy e ij The calculation formula of (2) is expressed as:
e j the larger the value of (a), i.e. the larger the information entropy of the jth index, the less information indicating the jth index;
5) Calculating the difference coefficient of the jth index in the matrix
The coefficient of difference may also be referred to as the information utility value, which is noted as d j
d j ＝1-e j
If the utility value of the information obtained by calculation is larger, the corresponding information quantity is represented more;
6) Calculating weights
The entropy weight W of each index can be obtained by normalizing the difference coefficient, namely the information utility value j
Entropy weight W j The entropy weight matrix is obtained.
10. An electric power equipment running state evaluation system based on multisource information fusion is characterized in that: comprising the steps of (a) a step of,
the feature extraction unit is used for preprocessing the running state data of the power equipment and extracting features to form a feature set;
the subjective weight generating unit is used for calculating a membership matrix of the feature set by using a differential algorithm as subjective weight;
the objective weight generating unit calculates the combined weighting of the feature set by adopting an AHP algorithm and a gray correlation algorithm as a constant weight coefficient, and then calculates a variable weight coefficient as an objective weight;
the linear weighted combination unit is used for carrying out linear weighted combination on the subjective weight and the objective weight to construct a fuzzy matrix;
and the judging unit is used for judging the fuzzy matrix by adopting a fuzzy comprehensive judging method and outputting the judging result of the running state of the power equipment.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202211488457.2A CN116089897A (en) | 2022-11-25 | 2022-11-25 | Power equipment operation state evaluation method and system based on multi-source information fusion |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202211488457.2A CN116089897A (en) | 2022-11-25 | 2022-11-25 | Power equipment operation state evaluation method and system based on multi-source information fusion |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN116089897A true CN116089897A (en) | 2023-05-09 |
Family
ID=86185774
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202211488457.2A Pending CN116089897A (en) | 2022-11-25 | 2022-11-25 | Power equipment operation state evaluation method and system based on multi-source information fusion |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN116089897A (en) |
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN118137812A (en) * | 2024-04-30 | 2024-06-04 | 深圳市联明电源股份有限公司 | Water-cooling power supply control method and system based on power factor correction |
-
2022
- 2022-11-25 CN CN202211488457.2A patent/CN116089897A/en active Pending
Cited By (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN118137812A (en) * | 2024-04-30 | 2024-06-04 | 深圳市联明电源股份有限公司 | Water-cooling power supply control method and system based on power factor correction |
| CN118137812B (en) * | 2024-04-30 | 2024-07-16 | 深圳市联明电源股份有限公司 | Water-cooling power supply control method and system based on power factor correction |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN108022014B (en) | Power system load prediction method and system | |
| CN110909928B (en) | Energy load short-term prediction method and device, computer equipment and storage medium | |
| CN108647807B (en) | River flow prediction method | |
| CN111397901A (en) | Rolling bearing fault diagnosis method based on wavelet and improved PSO-RBF neural network | |
| CN112668105B (en) | Helicopter transmission shaft abnormity judgment method based on SAE and Mahalanobis distance | |
| CN116089897A (en) | Power equipment operation state evaluation method and system based on multi-source information fusion | |
| CN111860624A (en) | Power grid fault information classification method based on decision tree | |
| CN116955936A (en) | Enterprise big data algorithm attribute data prediction method | |
| CN112215410A (en) | Power load prediction method based on improved deep learning | |
| CN114139820A (en) | Improved modal decomposition method for non-invasive electric energy load prediction | |
| CN110796303A (en) | Short-term power load prediction method based on EWT and ODBSCAN | |
| CN114626487A (en) | Line-variable relation checking method based on random forest classification algorithm | |
| CN112636328A (en) | Medium-voltage distribution network equipment load state identification method | |
| CN110750848A (en) | Method for estimating remaining life of software-hardware degradation system by considering software operation | |
| CN115389888B (en) | Partial discharge real-time monitoring system based on high-voltage cable | |
| CN111709116A (en) | Blind signal decomposition method based on similarity measurement | |
| CN116527346A (en) | Threat node perception method based on deep learning graph neural network theory | |
| CN114896228B (en) | Industrial data stream cleaning model and method based on filtering rule multistage combination optimization | |
| CN115018007A (en) | Sensitive data classification method based on improved ID3 decision tree | |
| CN110705373B (en) | Tank voltage signal classification method based on wavelet packet analysis and residual error network | |
| CN116933119A (en) | Signal data trend removal method based on convolutional neural network | |
| CN114218980A (en) | Method for predicting residual life of rolling bearing | |
| CN114462306A (en) | Non-invasive electric load decomposition method based on variable weight time domain convolution network | |
| Laosiritaworn et al. | Classification techniques for control chart pattern recognition: A case of metal frame for actuator production | |
| CN112016732A (en) | Power load frequency domain prediction method and system based on IRF and ODBSCAN |
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 |