A method and device for monitoring power equipment faults based on data fusion

Abstract

The present invention discloses a method and device for monitoring faults of electric power equipment based on data fusion, including collecting each electric power interference signal and each electric power monitoring signal of each electric power equipment, calculating the fault assessment factor of each electric power monitoring signal, obtaining each electric power monitoring key signal and each electric power interference key signal, constructing an electric power monitoring correlation matrix, an interference signal correlation matrix and an interference monitoring correlation matrix, and further constructing an interference monitoring strong correlation matrix, obtaining electric power strong interference signals and electric power main signals, obtaining non-interference abnormal mutation points of each electric power main signal, and weighted summing all fault assessment factors according to the proportion of non-interference abnormal mutation points in each electric power main signal to obtain the fault judgment factor of each electric power equipment. The method and device for monitoring faults of electric power equipment based on data fusion provided by the embodiment of the present invention improves the reliability of fault monitoring of electric power equipment.

Description

Technical Field

[0001] The present invention relates to the technical field of electric power equipment, and in particular to a method and device for monitoring electric power equipment faults based on data fusion. Background Art

[0002] Power equipment refers to equipment related to power generation, transmission and transformation, and power grids. Power equipment includes power generation equipment and power supply equipment, and plays a vital role in the power system. Their stable operation is the basis for ensuring the reliability of power supply. Since power equipment is usually exposed to the external environment, it may be affected by various factors, causing power equipment failure. By monitoring power equipment for faults, potential faults can be discovered in time, large-scale power outages caused by equipment failures can be avoided, and the continuity and stability of the power system can be ensured. Therefore, fault monitoring of power equipment is extremely important.

[0003] However, since the operation of power equipment is easily interfered by many factors, the reliability of fault monitoring of power equipment is low.

[0004] It can be seen that how to improve the reliability of fault monitoring of power equipment has become a technical problem that needs to be solved urgently by those skilled in the art. Summary of the invention

[0005] The present invention provides a method and device for monitoring faults of electric equipment based on data fusion, so as to solve the technical problem of low reliability of fault monitoring of electric equipment in the prior art.

[0006] In order to solve the above technical problems, an embodiment of the present invention provides a method for monitoring power equipment faults based on data fusion.

[0007] Collect each power interference signal and each power monitoring signal of each power equipment; Analyze the fluctuation stability characteristics and fluctuation persistence characteristics of each power monitoring signal to obtain the fault assessment factor of each power monitoring signal; compare and analyze all the fault assessment factors to obtain each power monitoring key signal; obtain each power interference key signal based on the fluctuation correlation characteristics between each power interference signal and all the power monitoring key signals; Constructing a power monitoring correlation matrix according to the correlation characteristics between all the power monitoring key signals, constructing an interference signal correlation matrix according to the correlation characteristics between all the power interference key signals, and constructing an interference monitoring correlation matrix according to the correlation characteristics between each of the power interference key signals and each of the power monitoring key signals; Based on the linear relationship between the power monitoring correlation matrix, the interference signal correlation matrix and the interference monitoring correlation matrix, an interference monitoring strong correlation matrix is ​​constructed; the interference monitoring strong correlation matrix is ​​compared and screened to obtain a power strong interference signal and a power main signal; Based on the overlapping characteristics of the mutation points in all the strong power interference signals and the mutation points in each of the main power signals, a non-interference abnormal mutation point of each of the main power signals is obtained; A weighted sum is taken of all the fault assessment factors according to the proportion of the non-interference abnormal mutation points in each of the main power signals to obtain a fault judgment factor for each of the power equipment.

[0008] As one of the preferred solutions, the fluctuation stability characteristics and fluctuation continuity characteristics of each of the power monitoring signals are analyzed to obtain the fault assessment factor of each of the power monitoring signals, including: Extract all first mutation points in each of the power monitoring signals using a mutation point detection algorithm; The fluctuation prediction deviation index of each power monitoring signal is calculated based on the difference characteristics between the actual values ​​of all the first mutation points in each power monitoring signal and the corresponding predicted values; the continuous average interval of mutations of each power monitoring signal is calculated based on the aggregation characteristics of all the first mutation points in each power monitoring signal; Based on the linear relationship between the fluctuation prediction deviation index of each of the power monitoring time signals and the corresponding continuous average interval of the mutations, the fault assessment factor of each of the power monitoring signals is calculated.

[0009] As one of the preferred solutions, the method of obtaining each power interference key signal based on the fluctuation correlation characteristics between each power interference signal and all the power monitoring key signals includes: Calculate the TLCC time lag mutual correlation coefficients of each of the power interference signals and each of the power monitoring key signals at all lag orders, and take the lag order corresponding to the maximum value of all the TLCC time lag mutual correlation coefficients as the nearest lag correlation interval between each of the power interference signals and each of the power monitoring key signals; The minimum value of the most recent lag correlation interval between each power interference signal and all the power monitoring key signals is used as the monitoring interference index of each power interference signal, and the monitoring interference indexes of all the power interference signals are sorted and analyzed to obtain each power interference key signal.

[0010] As one of the preferred solutions, the comparison and screening of the interference monitoring strong correlation matrix to obtain the power strong interference signal and the power main signal includes: The average value of all elements in the interference monitoring strong correlation matrix is ​​used as the interference monitoring average correlation coefficient; the elements in the interference monitoring strong correlation matrix that are greater than or equal to the interference monitoring average correlation coefficient are used as the interference monitoring strong correlation coefficient; The key power monitoring signals corresponding to all the interference monitoring strong correlation coefficients are used as the main power signals; and the key power interference signals corresponding to all the interference monitoring strong correlation coefficients are used as the strong power interference signals.

[0011] As one of the preferred solutions, the method of obtaining the non-interference abnormal mutation point of each of the main power signals based on the overlapping characteristics of the mutation points in all the strong power interference signals and the mutation points in each of the main power signals includes: Using a mutation point detection algorithm to extract all second mutation points in each of the power interference signals; Inputting the minimum value of the time interval between each of the first mutation points in each of the power main signals and the second mutation points in all of the power strong interference signals into the interference mutation evaluation expression, and calculating the interference cause index of each of the first mutation points in each of the power main signals; The interference mutation evaluation expression is:

[0012] in, is the interference cause index of the mth first mutation point in the ith power main signal, is an exponential function with a natural constant as base, is the minimum value of the time interval between the mth first mutation point in the ith power main signal and the second mutation points in all the power strong interference signals; An average value of all the interference cause indexes is used as an interference cause reference index, and the first mutation points in all the power main signals that are less than or equal to the interference cause reference index are used as non-interference abnormal mutation points.

[0013] As one of the preferred solutions, the fluctuation prediction deviation index of each power monitoring signal is calculated based on the difference characteristics between the actual values ​​of all the first mutation points in each power monitoring signal and the corresponding predicted values, including: taking the absolute value of the difference between the element value of each first mutation point in each power monitoring signal and the obtained predicted value of each first mutation point as the fluctuation prediction deviation of each first mutation point in each power monitoring signal; The sum of the fluctuation prediction deviations of all the first mutation points in each of the power monitoring signals is used as the fluctuation prediction deviation index of each of the power monitoring signals.

[0014] As one of the preferred solutions, the continuous average interval of mutations of each power monitoring signal is calculated based on the aggregation characteristics of all the first mutation points in each power monitoring signal, including: Taking the time interval between each of the first mutation points and the next adjacent first mutation point in each of the power monitoring signals as the mutation continuous interval of each of the first mutation points in each of the power monitoring signals; Performing cluster analysis on all the first mutation points in each of the power monitoring signals based on the mutation continuous interval to obtain a clustered mutation point cluster and a discrete mutation point cluster; The average value of the mutation continuous intervals of all the first mutation points in all the clustered mutation point clusters is used as the mutation continuous average interval of each of the power monitoring time signals.

[0015] As one of the preferred solutions, the comparative analysis of all the fault assessment factors to obtain each key power monitoring signal includes the following specific methods: The average value of all the fault assessment factors is used as the fault assessment mean value, and all the power monitoring signals that are greater than or equal to the fault assessment mean value are used as power monitoring key signals.

[0016] As one of the preferred solutions, the construction of an interference monitoring strong correlation matrix based on the linear relationship among the power monitoring correlation matrix, the interference signal correlation matrix and the interference monitoring correlation matrix includes: The product of the power monitoring correlation matrix and the interference monitoring correlation matrix is ​​used as a power interference strong correlation matrix; The product of the power interference strong correlation matrix and the interference signal correlation matrix is ​​used as the interference monitoring strong correlation matrix.

[0017] Another embodiment of the present invention provides an electric power equipment fault monitoring device based on data fusion, comprising a processor, a memory, and a computer program stored in the memory and configured to be executed by the processor, wherein when the processor executes the computer program, it implements the electric power equipment fault monitoring method based on data fusion as described above.

[0018] Compared with the prior art, the embodiments of the present invention have the following advantages: (1) Collect each power interference signal and each power monitoring signal of each power equipment, first analyze the fluctuation stability characteristics and fluctuation continuity characteristics of each power monitoring signal, obtain the fault assessment factor of each power monitoring signal, and use it to evaluate the possibility of abnormal fluctuation of each power monitoring signal, and compare and analyze all fault assessment factors to screen the power monitoring signals and obtain each power monitoring key signal; then analyze the fluctuation correlation characteristics between each power interference signal and all power monitoring key signals to screen the power interference signals, obtain each power interference key signal, and reduce the interference caused by redundant signals to the subsequent fault monitoring of power equipment; (2) A power monitoring correlation matrix is ​​constructed based on the correlation characteristics between all power monitoring key signals, an interference signal correlation matrix is ​​constructed based on the correlation characteristics between all power interference key signals, and an interference monitoring correlation matrix is ​​constructed based on the correlation characteristics between each power interference key signal and each power monitoring key signal. While considering the correlation between each power monitoring key signal and the power interference key signal, the correlation between the power monitoring key signals and the correlation between the power interference key signals are considered, thereby improving the accuracy of evaluating the degree of correlation between the power monitoring key signals and the power interference key signals. Based on the linear relationship between the power monitoring correlation matrix, the interference signal correlation matrix and the interference monitoring correlation matrix, an interference monitoring strong correlation matrix is ​​constructed. The interference monitoring strong correlation matrix is ​​compared and screened to obtain the power strong interference signal and the power main signal, and the key signals for fault monitoring of the power equipment are further screened out, thereby improving the reliability of subsequent fault monitoring of the power equipment. (3) Based on the overlapping characteristics of the mutation points in all strong power interference signals and the mutation points in each power main signal, the non-interference abnormal mutation points of each power main signal are obtained. According to the proportion of non-interference abnormal mutation points in each power main signal, all fault assessment factors are weighted and summed to obtain the fault judgment factor of each power equipment. According to the proportion of non-interference abnormal mutation points in each power main signal, a higher weight is set for the power main signal that is more likely to have abnormal fluctuations, thereby further improving the reliability of fault monitoring of power equipment. BRIEF DESCRIPTION OF THE DRAWINGS

[0019] Figure 1 It is a flow chart of a method for monitoring power equipment faults based on data fusion in one embodiment of the present invention; Figure 2 It is a schematic diagram of obtaining a strong correlation matrix for interference monitoring in one embodiment of the present invention. DETAILED DESCRIPTION

[0020] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. The purpose of providing these embodiments is to make the disclosure of the present invention more thorough and comprehensive. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0021] In the description of this application, the terms "first", "second", "third", etc. are used for descriptive purposes only and are not to be understood as indicating or implying relative importance or implicitly indicating the number of the indicated technical features. Thus, a feature defined as "first", "second", "third", etc. may explicitly or implicitly include one or more of the feature. In the description of this application, unless otherwise specified, "plurality" means two or more.

[0022] In the description of the present application, it should be noted that, unless otherwise clearly specified and limited, the terms "installed", "connected", and "connected" should be understood in a broad sense. For example, it can be a fixed connection, a detachable connection, or an integral connection; it can be a mechanical connection or an electrical connection; it can be a direct connection, or it can be indirectly connected through an intermediate medium, or it can be a connection between the two elements. The terms "vertical", "horizontal", "left", "right", "upper", "lower" and similar expressions used herein are only for illustrative purposes, and do not indicate or imply that the device or element referred to must have a specific orientation, be constructed and operated in a specific orientation, and therefore cannot be understood as a limitation on the present invention. The term "and / or" used herein includes any and all combinations of one or more related listed items. For those of ordinary skill in the art, the specific meanings of the above terms in this application can be understood according to specific circumstances.

[0023] In the description of this application, it should be noted that, unless otherwise defined, all technical and scientific terms used in the present invention have the same meaning as those commonly understood by those skilled in the art. The terms used in the specification of the present invention are only for the purpose of describing specific embodiments and are not intended to limit the present invention. For those of ordinary skill in the art, the specific meanings of the above terms in this application can be understood by specific circumstances.

[0024] An embodiment of the present invention provides a method for monitoring power equipment faults based on data fusion. For details, see Figure 1~Figure 2 , Figure 1 FIG. 1 is a flow chart of a method for monitoring power equipment faults based on data fusion in one embodiment of the present invention. Figure 2It is a schematic diagram showing the acquisition of the interference monitoring strong correlation matrix in one embodiment of the present invention.

[0025] A flowchart of a method for monitoring power equipment faults based on data fusion in one embodiment of the present invention includes the following steps S1 to S5, which are specifically as follows: Step S1: Collect each power interference signal and each power monitoring signal of each power device.

[0026] It should be noted that when fault monitoring is performed on power equipment, it will be interfered by various signals. These interference signals may come from multiple aspects such as the internal and external environment of the power system and the signal transmission process. Therefore, when fault monitoring is performed on power equipment, many types of signals need to be collected, including various power monitoring signals of the power equipment itself and various power interference signals that interfere with the fault monitoring of the power equipment.

[0027] Specifically, each power interference signal and each power monitoring signal of each power device are collected.

[0028] As an embodiment of the present application, the power monitoring signal includes a voltage signal, a current signal, a power signal, a vibration signal and a partial discharge signal, and the power interference signal includes an ambient temperature signal, an ambient humidity signal and an electromagnetic field signal.

[0029] Step S2: Analyze the fluctuation stability characteristics and fluctuation continuity characteristics of each power monitoring signal to obtain the fault assessment factor of each power monitoring signal, perform comparative analysis on all fault assessment factors to obtain each power monitoring key signal; obtain each power interference key signal based on the fluctuation correlation characteristics between each power interference signal and all power monitoring key signals.

[0030] It should be noted that since there are many types of signals involved in fault monitoring of power equipment, the existence of redundant signals will interfere with the fault monitoring of power equipment. Therefore, in order to improve the reliability of fault monitoring of power equipment, redundant signals are eliminated before fault monitoring of power equipment.

[0031] Specifically, a mutation point detection algorithm is used to extract all the first mutation points in each power monitoring signal, and the fluctuation prediction deviation index of each power monitoring signal is calculated according to the difference characteristics between the actual values ​​of all the first mutation points in each power monitoring signal and the corresponding predicted values; the continuous average interval of mutations of each power monitoring signal is calculated according to the aggregation characteristics of all the first mutation points in each power monitoring signal, and based on the linear relationship between the fluctuation prediction deviation index of each power monitoring time signal and the corresponding continuous average interval of mutations, the fault assessment factor of each power monitoring signal is calculated.

[0032] It should be noted that the mutation point detection algorithm aims to identify points in the data set that are significantly different from other data points. These points usually represent abnormal change events in the data. By detecting these mutation points, hidden patterns, trends or abnormal behaviors in the data can be revealed, providing key information for decision-making. The mutation point detection algorithm is a well-known technology and will not be described in detail in this embodiment.

[0033] As an embodiment of the present application, the Pettitt mutation point detection algorithm is used to extract all the first mutation points in each power monitoring signal. As for the selection of the mutation point detection algorithm, as other implementation methods, the implementer can choose it at his own discretion, and the present application does not impose any special restrictions on this.

[0034] In this step, the specific steps of calculating the fluctuation prediction deviation index of each power monitoring signal according to the difference characteristics between the actual values ​​of all first mutation points in each power monitoring signal and the corresponding predicted values ​​are as follows: A time series prediction algorithm is used to predict the element value of each first mutation point in each power monitoring signal. The absolute value of the difference between the element value of each first mutation point in each power monitoring signal and the predicted value of each first mutation point is taken as the fluctuation prediction deviation of each first mutation point in each power monitoring signal. The sum of the fluctuation prediction deviations of all first mutation points in each power monitoring signal is taken as the fluctuation prediction deviation index of each power monitoring signal.

[0035] It should be noted that the time series prediction algorithm analyzes the historical changes of time series data, reveals the patterns, trends and periodicities in the data, and then predicts future values. By comparing and analyzing the actual values ​​and predicted values, the degree to which the mutation point deviates from the overall trend is evaluated, and the possibility of abnormal fluctuations in each power monitoring signal is evaluated, thus obtaining the fluctuation prediction deviation index of each power monitoring signal.

[0036] As an embodiment of the present application, an autoregressive moving average model is used to predict the element value of each first mutation point in each power monitoring signal. As for the selection of the time series prediction algorithm, as other implementation methods, the implementer can choose it at his own discretion, and the present application does not impose any special restrictions on this.

[0037] The specific steps of calculating the continuous average interval of mutations of each power monitoring signal according to the aggregation characteristics of all the first mutation points in each power monitoring signal are as follows: The time interval between each first mutation point and the adjacent next first mutation point in each power monitoring signal is taken as the mutation continuity interval of each first mutation point in each power monitoring signal; based on the mutation continuity interval, all first mutation points in each power monitoring signal are clustered and analyzed to obtain clusters of aggregated mutation points and discrete mutation point clusters; the average value of the mutation continuity intervals of all first mutation points in all aggregated mutation point clusters is taken as the average mutation continuity interval of each power monitoring time signal.

[0038] Furthermore, the average value of all fault assessment factors is used as the fault assessment mean value, and all power monitoring signals greater than or equal to the fault assessment mean value are used as power monitoring key signals.

[0039] It should be noted that by analyzing the fluctuation correlation characteristics between the power interference signal and the power monitoring signal, the abnormal fluctuation of the power monitoring signal caused by the interference can be eliminated, thereby improving the reliability of power equipment fault monitoring. However, since the power monitoring signal is affected by many power interference signals, in order to reduce the interference of redundant signals to the subsequent power equipment fault monitoring, the power interference signal is screened.

[0040] Specifically, the TLCC time lag correlation coefficients of each power interference signal and each power monitoring key signal at all lag orders are calculated, and the lag order corresponding to the maximum value of all TLCC time lag correlation coefficients is used as the nearest lag correlation interval between each power interference signal and each power monitoring key signal; the minimum value of the nearest lag correlation interval between each power interference signal and all power monitoring key signals is used as the monitoring interference index of each power interference signal, and the monitoring interference indexes of all power interference signals are sorted and analyzed to obtain each power interference key signal.

[0041] It should be noted that the TLCC time-lagged mutual correlation coefficient is used to evaluate the degree of temporal alignment of two time series signals, and whether one signal is time-lagged or leading relative to the other signal. The calculation of the TLCC time-lagged mutual correlation coefficient is a well-known technology and will not be described in detail in this embodiment.

[0042] Step S3: construct a power monitoring correlation matrix based on the correlation characteristics between all power monitoring key signals, construct an interference signal correlation matrix based on the correlation characteristics between all power interference key signals, and construct an interference monitoring correlation matrix based on the correlation characteristics between each power interference key signal and each power monitoring key signal.

[0043] It should be noted that power monitoring signals and power interference signals are closely correlated in the power system. For example, changes in voltage and current will directly affect the power output. The more highly correlated the power monitoring signal is, the more likely it is to be a key signal for power equipment fault monitoring. By analyzing the correlation between these signals, the operating status of the power system and possible problems can be revealed.

[0044] Specifically, the first Pearson correlation coefficient between any two power monitoring key signals is calculated, and a power monitoring association matrix is ​​constructed based on all first Pearson correlation coefficients; the second Pearson correlation coefficient between any two power interference key signals is calculated, and an interference signal association matrix is ​​constructed based on all second Pearson correlation coefficients; the third Pearson correlation coefficient between each power monitoring key signal and each power interference key signal is calculated, and an interference monitoring association matrix is ​​constructed based on all third Pearson correlation coefficients.

[0045] It should be noted that the Pearson correlation coefficient is used to measure the correlation between two variables. The calculation of the Pearson correlation coefficient is a well-known technique and will not be described in detail in this embodiment.

[0046] Step S4: construct an interference monitoring strong correlation matrix based on the linear relationship between the power monitoring correlation matrix, the interference signal correlation matrix and the interference monitoring correlation matrix; perform comparative screening on the interference monitoring strong correlation matrix to obtain a power strong interference signal and a power main signal.

[0047] Specifically, the product of the power monitoring correlation matrix and the interference monitoring correlation matrix is ​​used as the power interference strong correlation matrix; the product of the power interference strong correlation matrix and the interference signal correlation matrix is ​​used as the interference monitoring strong correlation matrix.

[0048] It should be noted that by multiplying the power monitoring correlation matrix, the interference signal correlation matrix and the interference monitoring correlation matrix together through matrix multiplication, while considering the correlation between each power monitoring key signal and the power interference key signal, the correlation between the power monitoring key signals and the correlation between the power interference key signals are also considered, thereby improving the accuracy of evaluating the degree of correlation between the power monitoring key signals and the power interference key signals.

[0049] Furthermore, the average value of all elements in the interference monitoring strong correlation matrix is ​​taken as the interference monitoring average correlation coefficient, and the elements in the interference monitoring strong correlation matrix that are greater than or equal to the interference monitoring average correlation coefficient are taken as the interference monitoring strong correlation coefficient; the power monitoring key signals corresponding to all interference monitoring strong correlation coefficients are taken as the power main signals, and the power interference key signals corresponding to all interference monitoring strong correlation coefficients are taken as the power strong interference signals.

[0050] It should be noted that the reliability of subsequent fault monitoring of power equipment is improved by further screening the key power monitoring signals and the key power interference signals through the interference monitoring strong correlation matrix.

[0051] Step S5: Based on the overlapping characteristics of the mutation points in all strong power interference signals and the mutation points in each main power signal, the non-interference abnormal mutation points of each main power signal are obtained; according to the proportion of non-interference abnormal mutation points in each main power signal, all fault assessment factors are weighted and summed to obtain the fault judgment factor of each power equipment.

[0052] Specifically, a mutation point detection algorithm is used to extract all second mutation points in each power interference signal; the minimum time interval between each first mutation point in each power main signal and the second mutation points in all power strong interference signals is input into the interference mutation evaluation expression, and the interference cause index of each first mutation point in each power main signal is calculated.

[0053] The interference mutation evaluation expression is:

[0054] in, is the interference cause index of the mth first mutation point in the ith power main signal, is an exponential function with a natural constant as base, is the minimum value of the time interval between the mth first mutation point in the ith power main signal and the second mutation points in all the power strong interference signals.

[0055] It should be noted that, when the time interval between each first mutation point and the second mutation point in all strong power interference signals is smaller, it means that the mutation of the first mutation point is more likely to be caused by the mutation of the interference signal, and the interference cause index is larger.

[0056] The average value of all interference cause indexes is taken as the interference cause reference index, and the first mutation point in all power main signals that is less than or equal to the interference cause reference index is taken as the non-interference abnormal mutation point.

[0057] Furthermore, all fault assessment factors are weighted and summed according to the proportion of non-interference abnormal mutation points in each power main signal to obtain the fault judgment factor of each power equipment.

[0058] As an embodiment of the present application, the number of non-interference abnormal mutation points in all power main signals is taken as the total number of non-interference mutations, the ratio of the number of non-interference abnormal mutation points in each power main signal to the total number of non-interference mutations is taken as the fault assessment weight of each power main signal, the fault assessment factors of all power main signals of each power equipment are normalized to obtain the corresponding fault assessment normalization factor, and the sum of the products of the fault assessment weights of all power main signals of each power equipment and the corresponding fault assessment factors is taken as the fault judgment factor of each power equipment to perform fault monitoring on each power equipment.

[0059] Correspondingly, another embodiment of the present invention provides an electric power equipment fault monitoring device based on data fusion, comprising a processor, a memory, and a computer program stored in the memory and configured to be executed by the processor, and when the processor executes the computer program, it implements an electric power equipment fault monitoring method based on data fusion as described above.

[0060] The above-mentioned embodiments only express several implementation methods of the present invention, and the description thereof is relatively specific and detailed, but it cannot be understood as limiting the scope of the patent of the present invention. It should be pointed out that, for ordinary technicians in this field, several variations and improvements can be made without departing from the concept of the present invention, which all belong to the protection scope of the present invention. Therefore, the protection scope of the patent of the present invention shall be subject to the attached claims.

Claims

1. A method for monitoring power equipment faults based on data fusion, characterized in that: The method comprises: Collect each power interference signal and each power monitoring signal of each power equipment; Analyze the fluctuation stability characteristics and fluctuation persistence characteristics of each power monitoring signal to obtain the fault assessment factor of each power monitoring signal; compare and analyze all the fault assessment factors to obtain each power monitoring key signal; obtain each power interference key signal based on the fluctuation correlation characteristics between each power interference signal and all the power monitoring key signals; Constructing a power monitoring correlation matrix according to the correlation characteristics between all the power monitoring key signals, constructing an interference signal correlation matrix according to the correlation characteristics between all the power interference key signals, and constructing an interference monitoring correlation matrix according to the correlation characteristics between each of the power interference key signals and each of the power monitoring key signals; Based on the linear relationship between the power monitoring correlation matrix, the interference signal correlation matrix and the interference monitoring correlation matrix, an interference monitoring strong correlation matrix is ​​constructed; the interference monitoring strong correlation matrix is ​​compared and screened to obtain a power strong interference signal and a power main signal; Based on the overlapping characteristics of the mutation points in all the strong power interference signals and the mutation points in each of the main power signals, a non-interference abnormal mutation point of each of the main power signals is obtained; A weighted sum is taken of all the fault assessment factors according to the proportion of the non-interference abnormal mutation points in each of the main power signals to obtain a fault judgment factor for each of the power equipment.

2. A method for monitoring power equipment faults based on data fusion according to claim 1, characterized in that: The analyzing the fluctuation stability characteristics and the fluctuation continuity characteristics of each of the power monitoring signals to obtain the fault assessment factor of each of the power monitoring signals includes: Using a mutation point detection algorithm to extract all first mutation points in each of the power monitoring signals; The fluctuation prediction deviation index of each power monitoring signal is calculated based on the difference characteristics between the actual values ​​of all the first mutation points in each power monitoring signal and the corresponding predicted values; the continuous average interval of mutations of each power monitoring signal is calculated based on the aggregation characteristics of all the first mutation points in each power monitoring signal; Based on the linear relationship between the fluctuation prediction deviation index of each of the power monitoring time signals and the corresponding continuous average interval of the mutations, the fault assessment factor of each of the power monitoring signals is calculated.

3. The method for monitoring power equipment faults based on data fusion according to claim 1, characterized in that: The obtaining of each power interference key signal based on the fluctuation correlation characteristics between each power interference signal and all the power monitoring key signals includes: Calculate the TLCC time lag mutual correlation coefficients of each of the power interference signals and each of the power monitoring key signals at all lag orders, and take the lag order corresponding to the maximum value of all the TLCC time lag mutual correlation coefficients as the nearest lag correlation interval between each of the power interference signals and each of the power monitoring key signals; The minimum value of the most recent lag correlation interval between each power interference signal and all the power monitoring key signals is used as the monitoring interference index of each power interference signal, and the monitoring interference indexes of all the power interference signals are sorted and analyzed to obtain each power interference key signal.

4. The method for monitoring power equipment faults based on data fusion according to claim 1, characterized in that: The comparison and screening of the interference monitoring strong correlation matrix to obtain the power strong interference signal and the power main signal includes: The average value of all elements in the interference monitoring strong correlation matrix is ​​used as the interference monitoring average correlation coefficient; the elements in the interference monitoring strong correlation matrix that are greater than or equal to the interference monitoring average correlation coefficient are used as the interference monitoring strong correlation coefficient; The key power monitoring signals corresponding to all the interference monitoring strong correlation coefficients are used as the main power signals; and the key power interference signals corresponding to all the interference monitoring strong correlation coefficients are used as the strong power interference signals.

5. The method for monitoring power equipment faults based on data fusion according to claim 2, characterized in that: The method of obtaining the non-interference abnormal mutation point of each of the main power signals based on the overlapping characteristics of the mutation points in all the strong power interference signals and the mutation points in each of the main power signals comprises: Using a mutation point detection algorithm to extract all second mutation points in each of the power interference signals; Inputting the minimum value of the time interval between each of the first mutation points in each of the power main signals and the second mutation points in all of the power strong interference signals into the interference mutation evaluation expression, and calculating the interference cause index of each of the first mutation points in each of the power main signals; The interference mutation evaluation expression is: in, is the interference cause index of the mth first mutation point in the ith power main signal, is an exponential function with a natural constant as base, is the minimum time interval between the mth first mutation point in the ith power main signal and the second mutation point in all the strong power interference signals; the average value of all the interference cause indexes is taken as the interference cause reference index, and the first mutation points in all the power main signals that are less than or equal to the interference cause reference index are taken as non-interference abnormal mutation points.

6. A method for monitoring power equipment faults based on data fusion according to claim 2, characterized in that: The calculating, according to the difference characteristics between the actual values ​​of all the first mutation points in each of the power monitoring signals and the corresponding predicted values, of the fluctuation prediction deviation index of each of the power monitoring signals comprises: taking the absolute value of the difference between the element value of each first mutation point in each power monitoring signal and the obtained predicted value of each first mutation point as the fluctuation prediction deviation of each first mutation point in each power monitoring signal; The sum of the fluctuation prediction deviations of all the first mutation points in each of the power monitoring signals is used as the fluctuation prediction deviation index of each of the power monitoring signals.

7. The method for monitoring power equipment faults based on data fusion according to claim 2, characterized in that: The step of calculating the continuous average interval of mutations of each power monitoring signal according to the aggregated characteristics of all the first mutation points in each power monitoring signal comprises: Taking the time interval between each of the first mutation points and the next adjacent first mutation point in each of the power monitoring signals as the mutation continuous interval of each of the first mutation points in each of the power monitoring signals; Performing cluster analysis on all the first mutation points in each of the power monitoring signals based on the mutation continuous interval to obtain a clustered mutation point cluster and a discrete mutation point cluster; The average value of the mutation continuous intervals of all the first mutation points in all the clustered mutation point clusters is used as the mutation continuous average interval of each of the power monitoring time signals.

8. The method for monitoring power equipment faults based on data fusion according to claim 1, characterized in that: The specific method of comparing and analyzing all the fault assessment factors to obtain each key power monitoring signal includes: The average value of all the fault assessment factors is used as the fault assessment mean value, and all the power monitoring signals that are greater than or equal to the fault assessment mean value are used as power monitoring key signals.

9. The method for monitoring power equipment faults based on data fusion according to claim 1, characterized in that: The constructing of the interference monitoring strong correlation matrix based on the linear relationship among the power monitoring correlation matrix, the interference signal correlation matrix and the interference monitoring correlation matrix includes: The product of the power monitoring correlation matrix and the interference monitoring correlation matrix is ​​used as a power interference strong correlation matrix; The product of the power interference strong correlation matrix and the interference signal correlation matrix is ​​used as the interference monitoring strong correlation matrix.

10. A power equipment fault monitoring device based on data fusion, characterized in that: It comprises a processor, a memory, and a computer program stored in the memory and configured to be executed by the processor, and when the processor executes the computer program, it implements a method for monitoring faults of electric power equipment based on data fusion as described in any one of claims 1 to 9.

Images