A fault monitoring system for high voltage lines in electrical engineering

Abstract

The present application relates to the field of line monitoring technology, and specifically to a fault monitoring system for high-voltage lines in electrical engineering. The system includes: a data acquisition module: collecting current signals at each moment of monitoring nodes on mining high-voltage lines; an abnormality analysis module: constructing the harmonic abnormality characteristic value of each current signal according to the current signal changes and data mutation characteristics in the adjacent time period of each current signal; obtaining the abnormality score correction value of each current signal based on the probability density and harmonic abnormality characteristic value of each current signal; a power transmission control module: obtaining all abnormal current signals based on the abnormal score correction value using an abnormality detection algorithm; obtaining the determination result of abnormal faults based on all abnormal current signals. Thereby, the monitoring of mining high-voltage lines is realized, the accuracy of abnormal fault monitoring of mining high-voltage line power supply is improved, and the stability of mining high-voltage line power supply is effectively guaranteed.

Description

Technical Field

[0001] The present application relates to the technical field of line monitoring, and in particular to a high-voltage line fault monitoring system for electrical engineering. Background Art

[0002] In the context of the new era, mining safety has become an important management category in my country. For mining, safety issues are as important as health and life. Electricity is the main energy source for mining. The stability of the power supply of high-voltage lines for mining can be said to be the lifeline of mine workers. Only when the power supply of high-voltage lines is stable, effective and safe can the normal construction and operation of other facilities and equipment be guaranteed. At the same time, it also provides reliable power support for mining safety. Among them, mining high-voltage lines are a type of electrical engineering high-voltage lines.

[0003] However, mining operations often require large-scale production equipment, which is prone to generate harmonic components during actual operation. When the harmonic components pass through the high-voltage lines, they will affect the stability of the power supply of the mine high-voltage lines, and easily cause the mining equipment to fail to work properly or to have safety hazards such as short circuits and leakage.

[0004] With the continuous development of the field of machine learning, most of the existing technologies monitor the current signal of the high-voltage power supply line for mines and use anomaly detection algorithms to monitor abnormal faults in the power supply of the high-voltage power supply line for mines. For example, the common HBOS anomaly detection algorithm (Histogram-based Outlier Score, HBOS) monitors abnormal faults in the power supply of the high-voltage power supply line for mines by calculating anomaly scores. However, the anomaly score in the traditional HBOS anomaly detection algorithm is only calculated based on the probability density of power data, without specifically considering the abnormal characteristics of the harmonic current. Therefore, the use of the traditional HBOS anomaly detection algorithm to monitor abnormal faults in the power supply of the high-voltage power supply line for mines is likely to result in poor accuracy in monitoring abnormal faults in the power supply of the high-voltage power supply line for mines, and the stability of the power supply of the high-voltage power supply line for mines cannot be guaranteed. Summary of the invention

[0005] In order to solve the above technical problems, the purpose of this application is to provide an electrical engineering high-voltage line fault monitoring system, and the technical solutions adopted are as follows:

[0006] The present application proposes an electrical engineering high voltage line fault monitoring system, the system comprising:

[0007] Data acquisition module: collects the current signals of the monitoring nodes on the mining high-voltage line at each moment; records the sequence composed of all current signals as the current signal time series;

[0008] Abnormal analysis module: (1) Obtain the surge current significance value of each current signal according to the current signal changes in the adjacent time period of each current signal;

[0009] (2) Constructing the harmonic component interference degree of each current signal based on the change of the surge current significant value of the current signal in the current signal time series;

[0010] (3) Constructing the harmonic abnormal characteristic value of each current signal based on the interference degree of the harmonic component and the data mutation characteristics of the current signal in the adjacent time period of each current signal;

[0011] (4) Obtaining an abnormal score correction value of each current signal based on the probability density and harmonic abnormal characteristic value of each current signal;

[0012] Power transmission control module: using an abnormality detection algorithm to obtain all abnormal current signals based on the abnormality score correction value; obtaining abnormal fault determination results based on all abnormal current signals.

[0013] In one embodiment, the surge current significant value of each current signal is obtained by:

[0014] In the current signal time series, a set consisting of the first number of current signals having the closest Euclidean distance to any current signal is taken as a neighboring current signal set of the any current signal, and a sequence consisting of all elements in the neighboring current signal set is recorded as a current change characteristic sequence;

[0015] constructing a current peak characteristic value of each current signal based on the data prominence degree in a set of adjacent current signals of each current signal;

[0016] The set consisting of the absolute values ​​of the change trends of all data points on the fitting curve of the current change characteristic sequence of each current signal is used as the sharp change set of each current signal; the sharp characteristic value of the current change of each current signal is obtained according to the data change in the sharp change set;

[0017] The forward fusion result of the current peak characteristic value and the current rapid change characteristic value is used as the surge current significant value.

[0018] In one embodiment, the process of acquiring the current peak characteristic value of each current signal is as follows:

[0019] Obtaining the discreteness and minimum value of all elements in the set of adjacent current signals of each current signal, which are recorded as the first discreteness and the minimum current respectively;

[0020] Calculate the difference between each current signal and the minimum current in the set of adjacent current signals, and record it as the first difference;

[0021] The current peak characteristic value of each current signal is determined based on the first difference of each current signal and the first discreteness of the corresponding adjacent current signal set, and the current peak characteristic value of each current signal is positively correlated with the first difference of each current signal and the first discreteness of the corresponding adjacent current signal set.

[0022] In one embodiment, the process of acquiring the current change sharp characteristic value of each current signal is as follows:

[0023] Calculate the disorder degree and mean value of all elements in the set of rapid changes of each current signal, and record them as the first disorder degree and the first mean value respectively;

[0024] A current change sharp characteristic value of each current signal is determined based on the first chaos degree and the first mean value, and the current change sharp characteristic value of each current signal is positively correlated with the first chaos degree and the first mean value respectively.

[0025] In one embodiment, the process of obtaining the interference degree of the harmonic components of each current signal is as follows:

[0026] Taking a sequence consisting of surge current significant values ​​of all current signals in a set of adjacent current signals of any current signal in the current signal time series as a surge change characteristic sequence of the any current signal; taking a first-order difference sequence of the surge change characteristic sequence as a surge change difference sequence;

[0027] The interference degree of the harmonic components of each current signal is a forward fusion result of the absolute values ​​of all elements in the surge change difference sequence of each current signal.

[0028] In one embodiment, the process of obtaining the harmonic abnormality characteristic value of each current signal is as follows:

[0029] Constructing a mutation signal sequence of each current signal based on a mutation point in a set of neighboring current signals of each current signal;

[0030] The mean of all current signals in the current change characteristic sequence of each current signal is calculated and recorded as the second mean; the absolute value of the difference between the value of each mutation point in the mutation signal sequence of each current signal and the second mean is calculated and recorded as the first difference absolute value; the harmonic abnormal characteristic value of each current signal is positively correlated with the interference degree of the harmonic component of each current signal and the first difference absolute value.

[0031] In one embodiment, the process of acquiring the mutation signal sequence of each current signal is as follows:

[0032] The current change characteristic sequence of each current signal is used as the input of the mutation point detection algorithm, and the output is each mutation point; the sequence composed of all mutation points in the current change characteristic sequence of each current signal is used as the mutation signal sequence of each current signal.

[0033] In one embodiment, the process of obtaining the abnormal score correction value of each current signal is as follows:

[0034] Obtain the probability density of each current signal in the current signal time series; obtain a logarithmic function with a natural constant as the base and the probability density of each current signal as the true number, and use the inverse of the product of the calculation result of the logarithmic function and the harmonic abnormal characteristic value of each current signal as the abnormal score correction value of each current signal.

[0035] In one embodiment, the abnormal current signal is obtained by:

[0036] All current signals in the current signal time series are used as inputs of an anomaly detection algorithm, a calculation method of anomaly scores in the anomaly detection algorithm is replaced with a calculation method of the anomaly score correction value, and all abnormal current signals are output.

[0037] In one embodiment, the abnormal fault determination result is obtained based on all abnormal current signals, specifically:

[0038] The average of the abnormal score correction values ​​of all abnormal current signals is used as the abnormal fault assessment value;

[0039] When the abnormal fault assessment value is greater than or equal to the preset assessment threshold, it is determined that an abnormal fault has occurred in the monitoring node on the mine high-voltage line; when the abnormal fault assessment value is less than the preset assessment threshold, it is determined that no abnormal fault has occurred in the monitoring node on the mine high-voltage line.

[0040] This application has the following beneficial effects:

[0041] The present application utilizes the HBOS anomaly detection algorithm to implement an abnormal fault monitoring system for monitoring nodes on a mining high-voltage line. Compared with utilizing the traditional HBOS anomaly detection algorithm to identify all abnormal current signals in the abnormal analysis layer, since the anomaly score in the traditional HBOS anomaly detection algorithm is calculated only based on the probability density of the power data, without specifically considering the abnormal characteristics of the harmonic current, it is easy to cause low accuracy in the current signal anomaly detection. The present application considers that large-scale production equipment is prone to generate harmonic components during actual operation, resulting in the occurrence of harmonic anomalies in the monitoring nodes on the mining high-voltage line, and utilizes the harmonic anomaly characteristic value to improve the anomaly score in the HBOS anomaly detection algorithm. By improving the accuracy of anomaly detection by the HBOS anomaly detection algorithm, the accuracy of abnormal fault monitoring of the mining high-voltage line power supply is improved, thereby effectively ensuring the stability of the mining high-voltage line power supply. BRIEF DESCRIPTION OF THE DRAWINGS

[0042] In order to more clearly illustrate the technical solutions and advantages in the embodiments of the present application or the prior art, the drawings required for use in the embodiments or the prior art descriptions are briefly introduced below. Obviously, the drawings described below are only some embodiments of the present application. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying any creative work.

[0043] Figure 1 A block diagram of an electrical engineering high voltage line fault monitoring system provided by one embodiment of the present application;

[0044] Figure 2 This is a schematic diagram of the output results of the HBOS anomaly detection algorithm;

[0045] Figure 3 It is a schematic diagram of the steps of the abnormal analysis module;

[0046] Figure 4 Schematic diagram of the steps of the power transmission control module. DETAILED DESCRIPTION

[0047] In order to further explain the technical means and effects adopted by the present application to achieve the predetermined invention purpose, the following is a detailed description of the specific implementation, structure, features and effects of an electrical engineering high-voltage line fault monitoring system proposed in the present application in combination with the accompanying drawings and preferred embodiments. In the following description, different "one embodiment" or "another embodiment" does not necessarily refer to the same embodiment. In addition, specific features, structures or characteristics in one or more embodiments may be combined in any suitable form.

[0048] Unless defined otherwise, all technical and scientific terms used herein have the same meaning as commonly understood by one of ordinary skill in the art to which this application belongs.

[0049] The specific scheme of an electrical engineering high-voltage line fault monitoring system provided by the present application is described in detail below with reference to the accompanying drawings.

[0050] See also Figure 1 , which shows a block diagram of an electrical engineering high-voltage line fault monitoring system provided by an embodiment of the present application, the system comprising:

[0051] The data acquisition module 101 acquires the current signal of the monitoring node through the data acquisition layer of the high-voltage transmission line monitoring system, and pre-processes the current signal in the abnormal analysis layer to acquire the current signal time series.

[0052] The purpose of this application is to improve the accuracy of abnormal fault monitoring of mine high-voltage power supply lines by improving the existing abnormality detection technology, thereby effectively ensuring the stability of mine high-voltage power supply lines.

[0053] The high-voltage transmission line monitoring system is divided into three layers, namely, the data acquisition layer, the abnormal analysis layer and the transmission control layer. The data acquisition layer includes a single-phase AC / DC high-voltage electric measurement and metering module, which uses the single-phase AC / DC high-voltage electric measurement and metering module to collect the current signal of the monitoring node on the mining high-voltage line in real time, and transmits the collected current signal of the monitoring node on the mining high-voltage line to the abnormal analysis layer;

[0054] The acquisition frequency of the current signal in this application is 50Hz, and the acquisition time is 20min. The implementer can determine the acquisition frequency and acquisition time according to the actual situation, and this application does not make specific restrictions;

[0055] In order to facilitate the subsequent analysis of the current signal of the monitoring node, the collected current signal is preprocessed in the abnormal analysis layer, and the current signal of the monitoring node is normalized using the Z-score normalization algorithm to obtain the normalized value of the current signal. The sequence of the normalized values ​​of the current signals of the monitoring nodes on the mining high-voltage line in ascending time order is used as the current signal time series of the monitoring nodes on the mining high-voltage line, wherein Z-score normalization is a well-known technology and the specific process will not be repeated here.

[0056] It should be noted that this embodiment only provides a normalization method for normalizing the current signal. There are many existing normalization methods. The implementer may also choose other normalization algorithms to normalize the current signal, and this embodiment does not make any specific restrictions.

[0057] The abnormality analysis module 102 obtains the current peak characteristic value and the current rapid change characteristic value based on the current signal time series, obtains the surge current significance value by using the current peak characteristic value and the current rapid change characteristic value, obtains the harmonic component interference degree based on the surge current significance value, obtains the harmonic abnormality characteristic value based on the harmonic component interference degree, and uses the harmonic abnormality characteristic value to improve the abnormality score in the HBOS abnormality detection algorithm to obtain the abnormality score correction value.

[0058] Normally, the current signal of the monitoring node on the mine high-voltage line will remain stable. However, in actual mining scenarios, there are a large number of large-scale production equipment, and large-scale production equipment is prone to generate high-frequency harmonics during actual operation, which causes the current at the monitoring node on the mine high-voltage line to show surge characteristic changes, thereby affecting the stability of the mine high-voltage line power supply.

[0059] (1) The surge current significance value of each current signal is obtained according to the current signal change in the adjacent time period of each current signal.

[0060] As an example, this step may include the following steps:

[0061] (1.1) Obtain the adjacent current signal set of each current signal and construct the current change characteristic sequence of each current signal:

[0062] In order to analyze the surge characteristic changes of the current of the monitoring node on the mining high-voltage line, in the current signal time series of the monitoring node, the set consisting of the first number of current signals with the closest Euclidean distance to any current signal is used as the neighboring current signal set of any current signal, and the sequence consisting of all elements in the neighboring current signal set of any current signal arranged in ascending order according to the corresponding acquisition time is recorded as the current change characteristic sequence. Among them, the first number is a preset value, which can be set by the implementer. In this embodiment, the first number is set to 50.

[0063] (1.2) Construct the current peak characteristic value of each current signal based on the data prominence degree in the set of adjacent current signals of each current signal:

[0064] When the surge characteristics of the current at the monitoring node on the mining high-voltage line change, the change of the current signal is prone to local current spike characteristics, which presents a large abnormal characteristic. Therefore, in order to improve the accuracy of abnormal fault monitoring of the mining high-voltage line power supply, it is necessary to analyze and calculate the local current spike characteristics of the current signal change at the monitoring node on the mining high-voltage line.

[0065] Obtaining the discreteness and minimum value of all elements in the set of adjacent current signals of each current signal, which are recorded as the first discreteness and the minimum current respectively;

[0066] Calculate the difference between each current signal and the minimum current in the set of adjacent current signals, and record it as the first difference;

[0067] The current peak characteristic value of each current signal is positively correlated with the first difference of each current signal and the first discreteness of a set of adjacent current signals.

[0068] It can be understood that the positive correlation in this application refers to the relationship between the independent variable and the dependent variable. The positive correlation means that the independent variable increases (decreases) as the dependent variable increases (decreases), which can be an additive relationship, a multiplicative relationship, etc.

[0069] Preferably, as an embodiment of the present application, the first discreteness may be the variance of all elements in a set of adjacent current signals of each current signal; the first difference may be the difference between each current signal and the minimum current in a set of adjacent current signals; the current peak characteristic value of each current signal may be the product of the first difference of each current signal and the first discreteness of a set of adjacent current signals;

[0070] In other embodiments of the present application, the first discreteness may also be the standard deviation of all elements in a set of adjacent current signals of each current signal; the first difference may be the ratio between each current signal and the minimum current in its set of adjacent current signals; the current peak characteristic value of each current signal may be the normalized value of the first difference of each current signal and the sum of the normalized values ​​of the first discreteness of its set of adjacent current signals.

[0071] The larger the first discreteness of the set of adjacent current signals, the greater the amplitude of the current signal change, and the easier it is for the current spike feature to appear. At the same time, the greater the difference between each current signal and the minimum value within its neighborhood, the more likely it is that the current spike feature will appear at the location representing the current signal, and the larger the current spike characteristic value.

[0072] (1.3) Construct a set of sharp changes of each current signal based on the change trend of the data points on the fitting curve of the current change characteristic sequence of each current signal; obtain the sharp characteristic value of the current change of each current signal according to the data change in the set of sharp changes:

[0073] Generally, when a surge current changes in the monitoring nodes of the mining high-voltage line, the current signal changes in a chaotic manner, which reflects the rapid change of the current when the surge occurs. Therefore, in order to more accurately measure the surge characteristic changes of the current at the monitoring nodes on the high-voltage line, it is necessary to analyze the characteristics of the rapid change of the current.

[0074] MATLAB software is used to draw a fitting curve of the current change characteristic sequence of each current signal, which is recorded as the current change characteristic curve, and the slope of each data point in the current change characteristic curve of each current signal is calculated. The set consisting of the absolute values ​​of the slopes of all data points in the current change characteristic curve of each current signal is used as the sharp change set of each current signal; wherein the calculation of the slope is a well-known technology, and the specific process is not repeated here; it should be noted that, for the fitting curve drawing of the current change characteristic sequence of each current signal, this embodiment only provides a curve drawing method, and there are many existing curve drawing methods. In other embodiments of the present application, the implementer may also use other curve drawing methods to draw a fitting curve for the current change characteristic sequence of each current signal;

[0075] The disorder degree and mean value of all elements in the sharp change set of each current signal are calculated and recorded as the first disorder degree and the first mean value respectively; the sharp current change characteristic value of each current signal is positively correlated with the first disorder degree and the first mean value respectively.

[0076] Preferably, as an embodiment of the present application, the first degree of disorder may be the information entropy of all elements in the set of sharp changes of each current signal; the sharp characteristic value of current change of each current signal may be the product of the first degree of disorder and the first mean value;

[0077] In other embodiments of the present application, the first degree of disorder may be the discrete coefficient of all elements in the sharp change set of each current signal; the sharp current change characteristic value of each current signal may be the sum of the first degree of disorder and the first mean value.

[0078] The larger the information entropy of the sharp change set, the higher the degree of chaos of the current signal change. At the same time, the larger the mean of the sharp change set, the higher the slope level of the current signal change. At this time, the greater the sharp change when the current surge occurs, the larger the sharp characteristic value of the current change.

[0079] (1.4) Based on the current peak characteristic value and the current change sharp characteristic value of each current signal, the surge current significant value of each current signal is obtained:

[0080] When the current in the monitoring node on the mining high-voltage line shows current spike characteristics and current rapidity characteristics, and the clearer the current spike characteristics and current rapidity characteristics are, that is, the larger the current spike characteristic value and the larger the current change rapidity characteristic value, the greater the possibility of surge current appearing on the monitoring node.

[0081] The forward fusion result of the current peak characteristic value and the current rapid change characteristic value of each current signal is taken as the surge current significant value of each current signal.

[0082] Preferably, as an embodiment of the present application, the surge current significant value of each current signal may be a normalized value of the product of the current peak characteristic value and the current change sharp characteristic value of each current signal.

[0083] The larger the current peak characteristic value and the larger the current change rapid characteristic value, the clearer the current peak characteristic and the current rapidity characteristic are. When a surge current appears at the monitoring node, a current peak characteristic and a rapid change characteristic will appear. That is, the greater the significance of the surge current at this time, the greater the surge current significance value.

[0084] (2) The harmonic component interference degree of each current signal is constructed based on the change of the surge current significance value of the current signal in the current signal time series.

[0085] As an example, this step may include the following steps:

[0086] Normally, high-frequency harmonics will be generated when surge current occurs. This is because large-scale mining equipment is prone to generate harmonic components during the production process during mining. When the harmonic components are passed through the high-voltage power supply system, they will cause large harmonic currents. At this time, the surge current significance value will change significantly. That is, the greater the significant change in the surge current, the more it can indicate that the monitoring node has interference from harmonic components.

[0087] A sequence of surge current significant values ​​of all current signals in a set of adjacent current signals of any current signal in the current signal time series in ascending order of time is used as a surge change characteristic sequence of any current signal; in order to analyze the difference in the change of current surges at the monitoring nodes on the mining high-voltage line, a first-order difference sequence of the surge change characteristic sequence of each current signal is used as a surge change difference sequence of each current signal; wherein the calculation of the first-order difference sequence is a well-known technology, and the specific process is not repeated here;

[0088] The harmonic component interference degree of each current signal is a forward fusion result of the absolute values ​​of all elements in the surge change difference sequence of each current signal.

[0089] Preferably, as an embodiment of the present application, the harmonic component interference degree of each current signal may be the average value of the absolute values ​​of all elements in the surge change difference sequence of each current signal;

[0090] In other embodiments of the present application, the harmonic component interference degree of each current signal may be the cumulative sum of the absolute values ​​of all elements in the surge change difference sequence of each current signal.

[0091] The larger the absolute value of all elements in the surge change difference sequence, the greater the difference in the significant change of the surge current. To a certain extent, it reflects that the monitoring node on the mining high-voltage line is more likely to be interfered by the harmonic component, which leads to a larger change difference in the surge current significant value characteristics. At this time, the corresponding harmonic component interference is greater.

[0092] (3) Construct the harmonic abnormal characteristic value of each current signal based on the interference degree of the harmonic component and the data mutation characteristics in the current change characteristic sequence:

[0093] Normally, when harmonic components appear in the high-voltage power supply system of a mine, the stability of the high-voltage power supply system is affected, resulting in abnormal current stability at the monitoring nodes on the mine high-voltage line. In order to more accurately extract the abnormal characteristics of harmonic components appearing at the monitoring nodes on the mine high-voltage line, it is necessary to analyze the abnormal phenomenon of current stability.

[0094] (3.1) Construct a mutation signal sequence of each current signal based on the mutation points in the adjacent current signal set of each current signal:

[0095] In order to analyze the abnormal characteristics of current signal fluctuations, the current change characteristic sequence of each current signal is used as the input of the Pettitt mutation point detection algorithm, and the output is all the mutation points in the current change characteristic sequence of each current signal; the sequence composed of all the mutation points in the current change characteristic sequence of each current signal is used as the mutation signal sequence of each current signal; the Pettitt mutation point detection algorithm is a well-known technology, and the specific process is not repeated here; it should be noted that, for the detection of mutation points in the current change characteristic sequence, this embodiment only provides a mutation point detection algorithm, there are many existing mutation point detection algorithms, and the implementer can also use other mutation point detection algorithms to perform mutation point detection on the current change characteristic sequence, and this embodiment does not make specific restrictions.

[0096] (3.2) Based on the interference degree of the harmonic components and the data difference between the current change characteristic sequence of each current signal and the mutation signal sequence, the harmonic abnormal characteristic value of each current signal is constructed:

[0097] The mean of all current signals in the current change characteristic sequence of each current signal is calculated and recorded as the second mean; the absolute value of the difference between the value of each mutation point in the mutation signal sequence of each current signal and the second mean is calculated and recorded as the first difference absolute value; the harmonic abnormal characteristic value of each current signal is positively correlated with the interference degree of the harmonic component of each current signal and the first difference absolute value.

[0098] Preferably, as an embodiment of the present application, the mean of all the first difference absolute values ​​of the mutation signal sequence of each current signal is calculated; the harmonic abnormality characteristic value of each current signal can be the product of the mean of the first difference absolute values ​​of each current signal and the interference degree of the harmonic component;

[0099] In other embodiments of the present application, the harmonic abnormality characteristic value of each current signal may be the sum of the mean of the absolute values ​​of the first differences of each current signal and the interference degree of the harmonic components.

[0100] The greater the difference between the elements in the mutation signal sequence and the mean value of the current signal, the less stable the local current signal is when it fluctuates, which reflects the greater the harmonic anomaly characteristics of its high-voltage power supply stability. At the same time, the greater the interference of the harmonic component, the greater the impact of the harmonic component on the monitoring node on the mine high-voltage line, the more likely it is that harmonic anomaly characteristics will appear, and the greater the harmonic anomaly characteristic value.

[0101] (4) Obtain the abnormal score correction value of each current signal based on the probability density and harmonic abnormal characteristic value of each current signal:

[0102] The anomaly score in the traditional HBOS anomaly detection algorithm is calculated only based on the probability density of power data, without specifically considering the abnormal characteristics of harmonic current, which easily leads to poor accuracy of abnormal fault monitoring. This application improves the calculation method of the anomaly score in the traditional HBOS anomaly detection algorithm and improves the accuracy of abnormal fault monitoring.

[0103] Obtaining the probability density of each current signal in the current signal time series; wherein the calculation of the probability density is a well-known technique, and the specific process will not be repeated here;

[0104] A logarithmic function with a natural constant as the base and the probability density of each current signal as the true number is obtained, and the inverse of the product of the calculation result of the logarithmic function and the harmonic abnormal characteristic value of each current signal is used as the abnormal score correction value of each current signal.

[0105] Compared with the traditional HBOS anomaly detection algorithm which only calculates anomaly scores based on probability density and lacks consideration of the abnormal characteristics of harmonic current, which easily leads to inaccurate anomaly scores, this application considers the abnormal characteristics of harmonics to improve the accuracy of the algorithm in calculating anomaly scores, thereby improving the accuracy of abnormal fault monitoring.

[0106] The power transmission control module 103 uses the HBOS abnormality detection algorithm to obtain all abnormal current signals of the abnormal analysis layer based on the abnormal score correction value, obtains the determination result of the abnormal fault based on all abnormal current signals of the abnormal analysis layer, and completes an electrical engineering high-voltage line fault monitoring system.

[0107] In order to improve the accuracy of abnormal fault monitoring of mine high-voltage power supply lines and effectively ensure the stability of mine high-voltage power supply lines, all current signals in the current signal time series are used as inputs of the HBOS anomaly detection algorithm, and the calculation method of the anomaly score in the HBOS anomaly detection algorithm is replaced by the calculation method of the anomaly score correction value. In this embodiment, the regularization term parameter in the HBOS anomaly detection algorithm is taken as 0.1, the tolerance parameter is taken as 0.5, and the abnormal proportion parameter is taken as 0.1. The output of the HBOS anomaly detection algorithm is used as all abnormal current signals of the abnormal analysis layer, wherein the HBOS anomaly detection algorithm is a well-known technology and the specific process will not be repeated; it should be noted that the values ​​of the regularization term parameter, the tolerance parameter and the abnormal proportion parameter can be set by the implementer, and this embodiment does not impose specific restrictions.

[0108] The output result diagram of HBOS anomaly detection algorithm is as follows: Figure 2 As shown, inliers represent normal current signals and outliers represent abnormal current signals.

[0109] The average of the abnormal score correction values ​​of all abnormal current signals of the abnormal analysis layer is taken as the abnormal fault assessment value;

[0110] When the abnormal fault evaluation value is greater than or equal to the evaluation threshold, it is determined that an abnormal fault has occurred in the monitoring node on the mine high-voltage line; when the abnormal fault evaluation value is less than the evaluation threshold, it is determined that no abnormal fault has occurred in the monitoring node on the mine high-voltage line; it should be noted that the evaluation threshold is an artificial preset value, which can be set by the implementer. In this embodiment, the evaluation threshold is set to 10.

[0111] Furthermore, the determination result of the abnormal fault in the abnormal analysis layer is transmitted to the power transmission control layer. When the determination result of the abnormal fault shows that an abnormal fault has occurred, the mine high-voltage power supply system is shut down. After the safety management personnel eliminate the abnormal fault, the mine high-voltage power supply system is restarted.

[0112] The steps of the abnormal analysis module are shown in the figure below: Figure 3 As shown; the schematic diagram of the steps of the power transmission control module is as follows Figure 4 shown.

[0113] In summary, the embodiment of the present application realizes an abnormal fault monitoring system for monitoring nodes on a mining high-voltage line by utilizing the HBOS anomaly detection algorithm. Compared with utilizing the traditional HBOS anomaly detection algorithm to identify all abnormal current signals of the abnormal analysis layer, since the anomaly score in the traditional HBOS anomaly detection algorithm is calculated only based on the probability density of the power data, without specifically considering the abnormal characteristics of the harmonic current, it is easy to cause low accuracy in the detection of current signal anomalies. The present application considers that large-scale production equipment is prone to generate harmonic components during actual operation, resulting in the occurrence of harmonic anomalies in the monitoring nodes on the mining high-voltage line, and utilizes the harmonic anomaly characteristic value to improve the anomaly score in the HBOS anomaly detection algorithm. By improving the accuracy of anomaly detection by the HBOS anomaly detection algorithm, the accuracy of abnormal fault monitoring of the mining high-voltage line power supply is improved, thereby effectively ensuring the stability of the mining high-voltage line power supply.

[0114] It should be noted that the sequence of the above embodiments of the present application is only for description and does not represent the advantages and disadvantages of the embodiments. The processes depicted in the accompanying drawings do not necessarily require the specific order or continuous order shown to achieve the desired results. In some embodiments, multitasking and parallel processing are also possible or may be advantageous.

[0115] The various embodiments in the present application are described in a progressive manner, and the same or similar parts between the various embodiments can be referenced to each other, and each embodiment focuses on the differences from other embodiments.

[0116] The embodiments described above are only used to illustrate the technical solutions of the present application, rather than to limit them. Modifications to the technical solutions recorded in the aforementioned embodiments, or equivalent replacement of some of the technical features therein, do not cause the essence of the corresponding technical solutions to deviate from the scope of the technical solutions of the embodiments of the present application, and should all be included in the protection scope of the present application.

Claims

1. An electrical engineering high voltage line fault monitoring system, characterized in that: The system comprises: Data acquisition module: collects the current signals of the monitoring nodes on the mining high-voltage line at each moment; records the sequence composed of all current signals as the current signal time series; Abnormal analysis module: (1) Obtain the surge current significance value of each current signal according to the current signal changes in the adjacent time period of each current signal; (2) Constructing the harmonic component interference degree of each current signal based on the change of the surge current significant value of the current signal in the current signal time series; (3) Constructing the harmonic abnormal characteristic value of each current signal based on the interference degree of the harmonic component and the data mutation characteristics of the current signal in the adjacent time period of each current signal; (4) Obtaining an abnormal score correction value of each current signal based on the probability density and harmonic abnormal characteristic value of each current signal; Power transmission control module: using an abnormality detection algorithm to obtain all abnormal current signals based on the abnormality score correction value; obtaining abnormal fault determination results based on all abnormal current signals; The method for obtaining the surge current significant value of each current signal is as follows: In the current signal time series, a set consisting of the first number of current signals having the closest Euclidean distance to any current signal is taken as a neighboring current signal set of the any current signal, and a sequence consisting of all elements in the neighboring current signal set is recorded as a current change characteristic sequence; constructing a current peak characteristic value of each current signal based on the data prominence degree in a set of adjacent current signals of each current signal; A set consisting of the absolute values ​​of the change trends of all data points on the fitting curve of the current change characteristic sequence of each current signal is used as the sharp change set of each current signal; and a sharp current change characteristic value of each current signal is obtained according to the data change in the sharp change set; Taking the forward fusion result of the current peak characteristic value and the current rapid change characteristic value as the surge current significant value; The process of obtaining the current peak characteristic value of each current signal is as follows: Obtaining the discreteness and minimum value of all elements in the set of adjacent current signals of each current signal, which are recorded as the first discreteness and the minimum current respectively; Calculate the difference between each current signal and the minimum current in the set of adjacent current signals, and record it as the first difference; Determine a current peak characteristic value of each current signal based on the first difference of each current signal and a first discreteness of a corresponding set of adjacent current signals, wherein the current peak characteristic value of each current signal is positively correlated with the first difference of each current signal and a first discreteness of a corresponding set of adjacent current signals; The process of obtaining the current change sharp characteristic value of each current signal is as follows: Calculate the disorder degree and mean value of all elements in the set of rapid changes of each current signal, and record them as the first disorder degree and the first mean value respectively; Determine a current change sharp characteristic value of each current signal based on the first chaos degree and the first mean value, wherein the current change sharp characteristic value of each current signal is positively correlated with the first chaos degree and the first mean value respectively; The process of obtaining the interference degree of the harmonic components of each current signal is as follows: Taking a sequence consisting of surge current significant values ​​of all current signals in a set of adjacent current signals of any current signal in the current signal time series as a surge change characteristic sequence of the any current signal; taking a first-order difference sequence of the surge change characteristic sequence as a surge change difference sequence; The harmonic component interference degree of each current signal is a forward fusion result of the absolute values ​​of all elements in the surge change difference sequence of each current signal.

2. The electrical engineering high voltage line fault monitoring system according to claim 1, characterized in that: The process of obtaining the harmonic abnormal characteristic value of each current signal is as follows: Constructing a mutation signal sequence of each current signal based on mutation points in a set of neighboring current signals of each current signal; The mean of all current signals in the current change characteristic sequence of each current signal is calculated and recorded as the second mean; the absolute value of the difference between the value of each mutation point in the mutation signal sequence of each current signal and the second mean is calculated and recorded as the first difference absolute value; the harmonic abnormal characteristic value of each current signal is positively correlated with the interference degree of the harmonic component of each current signal and the first difference absolute value.

3. The electrical engineering high voltage line fault monitoring system according to claim 2, characterized in that: The process of acquiring the mutation signal sequence of each current signal is as follows: The current change characteristic sequence of each current signal is used as the input of the mutation point detection algorithm, and the output is each mutation point; the sequence composed of all mutation points in the current change characteristic sequence of each current signal is used as the mutation signal sequence of each current signal.

4. The electrical engineering high voltage line fault monitoring system according to claim 1, characterized in that: The process of obtaining the abnormal score correction value of each current signal is as follows: Obtain the probability density of each current signal in the current signal time series; obtain a logarithmic function with a natural constant as the base and the probability density of each current signal as the true number, and use the inverse of the product of the calculation result of the logarithmic function and the harmonic abnormal characteristic value of each current signal as the abnormal score correction value of each current signal.

5. The electrical engineering high voltage line fault monitoring system according to claim 1, characterized in that: The acquisition process of the abnormal current signal is as follows: All current signals in the current signal time series are used as inputs of an anomaly detection algorithm, a calculation method of anomaly scores in the anomaly detection algorithm is replaced with a calculation method of the anomaly score correction value, and all abnormal current signals are output.

6. The electrical engineering high voltage line fault monitoring system according to claim 1, characterized in that: The determination result of abnormal faults obtained based on all abnormal current signals is specifically as follows: The average of the abnormal score correction values ​​of all abnormal current signals is used as the abnormal fault assessment value; When the abnormal fault assessment value is greater than or equal to the preset assessment threshold, it is determined that an abnormal fault has occurred in the monitoring node on the mine high-voltage line; when the abnormal fault assessment value is less than the preset assessment threshold, it is determined that no abnormal fault has occurred in the monitoring node on the mine high-voltage line.

Images