Switchgear cable joint performance degradation assessment method and system based on IoT perception

Abstract

The present invention discloses a method and system for evaluating the performance degradation of a switch cabinet cable joint based on Internet of Things perception, which belongs to the technical field of performance degradation evaluation of electrical components, and includes: S1, obtaining operation data and monitoring data of the switch cabinet cable joint; S2, extracting characteristic indicators for characterizing the degradation trend of the cable joint based on the operation data and the monitoring data; S3, using the FPCA method to perform dimensionality reduction processing on the characteristic indicators; S4, based on the characteristic indicators after dimensionality reduction, obtaining evaluation indicators for characterizing the degradation damage of the cable joint by setting constraints, and performing degradation evaluation on the switch cabinet cable joint according to the positive indicators and negative indicators of the evaluation indicators. The present invention can monitor the current performance status and degradation trend of the cable joint in the switch cabinet, and evaluate them, thereby reducing productivity loss and downtime, avoiding catastrophic failures, and improving the reliability of the switch cabinet operation.

Description

Technical Field

[0001] The present invention relates to the technical field of electrical component performance degradation assessment, and in particular to a switch cabinet cable connector performance degradation assessment method and system based on Internet of Things perception. Background Art

[0002] With the rapid development of the economy and society and the continuous increase in electricity demand, users are no longer satisfied with basic electricity demand, and their requirements for power supply services such as power supply reliability, power quality and emergency response continue to increase. In recent years, the power load of the power grid has grown rapidly, and the number of distribution network equipment has continued to increase. According to statistics, 70-80% of the faults occur on the distribution side. The switchgear is an important part of the distribution network. The intelligent switchgear needs to accurately monitor and understand the operating status of the power equipment in real time. Cable joints are one of the weak links in switchgears, but the current real-time status analysis and degradation trend prediction of cable joints are very insufficient.

[0003] At present, the power grid adopts a regular inspection management method for distribution equipment. In the future, the number of distribution equipment of various models and grades will continue to increase. If we only rely on inspection data and manual experience, it will be difficult to accurately evaluate the current degradation status of the equipment, including future fault predictions and remaining life and other effective information. It will not be able to meet the future needs of power equipment to deeply participate in demand response and power grid operation.

[0004] In the existing technical solutions, the basic methods are divided into two categories. One is the cable joint degradation status monitoring based on temperature characteristics: mainly divided into electrical signal temperature measurement method and optical signal temperature measurement method. The electrical signal temperature measurement method can use thermocouple temperature measurement or digital sensor temperature measurement. When the insulation failure of the cable joint causes the temperature to be abnormal, the degradation status of the cable joint can be characterized by monitoring the temperature of the cable joint.

[0005] The other type is cable joint degradation status monitoring based on partial discharge characteristics: when an insulation fault occurs in a cable joint, the point field distribution will be distorted and partial discharge will occur. Extracting relevant feature quantities from the partial discharge signal can characterize the degradation state of the cable joint. There are two main methods for extracting feature quantities: based on time domain waveforms and based on phase distribution.

[0006] The above two methods both monitor a single characteristic quantity of the cable joint, which leads to problems such as susceptibility to external influences and low accuracy. Summary of the invention

[0007] The present invention provides a switch cabinet cable joint performance degradation assessment method and system based on Internet of Things perception, which aims to monitor the current performance status and degradation trend of cable joints in the switch cabinet, reduce productivity loss and downtime to avoid catastrophic failures, improve the reliability of switch cabinet operation, ensure the safe and stable operation of the switch cabinet, and thus improve the power supply reliability of the power grid.

[0008] A switch cabinet cable joint performance degradation assessment method based on IoT perception in this solution includes the following steps:

[0009] S1, obtain operation data and IoT sensing multi-parameter monitoring data;

[0010] S2, extract characteristic indicators that characterize the degradation trend of cable joints;

[0011] S3, using FPCA method to reduce the dimension of cable joint degradation trend index;

[0012] S4, construct the comprehensive degradation damage assessment index of cable joints.

[0013] Furthermore, step S1 of acquiring data specifically comprises: acquiring IoT sensing multi-parameter monitoring data of voltage, current, temperature and partial discharge according to sensors, performing signal processing in time domain, frequency domain and time-frequency domain on the collected data, and extracting original features.

[0014] Further, the method for extracting features in step S2 is specifically: using information gain to select features in the decision tree algorithm of random forest, the greater the information gain, the better the selectivity of this feature, and the information entropy represents the degree of classification confusion, the more chaotic the classification, the greater the information entropy;

[0015] Among them, the calculation formulas of information gain and information entropy are:

[0016]

[0017]

[0018] Among them, p i It means that there are k categories in each data set, and the proportion of each category in the data set;

[0019] The random forest method is improved and the information gain rate calculation formula is defined as:

[0020]

[0021] First, the original features input into the random forest are sorted according to the information gain rate, and the features with large information gain rate are selected as the subset. Then, the accuracy of the random forest model of the subset is determined and compared with the model accuracy with the total features. The smallest subset with the maximum accuracy is taken as the optimal feature subset.

[0022] Further, the FPCA method in step S3 is specifically: fitting the curve χ to the extracted data representing the cable joint degradation trend index i (t), where t∈T, i=1, K, n;

[0023]

[0024] st||μ|| 2 =∫ T μ(t)μ(t)dt=1

[0025] ∫ T μ(t)μ r (t)dt=0,r=1,K,k

[0026] Among them, μ(t) is the weight function, ξ i is the principal component score, ξ i =∫ T μ(t)χ i (t)dt.

[0027] Then the characteristic equation is:

[0028] ∫ T u(t,v)μ(v)dv=λμ(t)

[0029] Among them, u(t, v) is the covariance function and λ is the eigenvalue.

[0030] The contribution rate of the eigenvalue is used to measure the proportion of the principal component. The calculation formula is:

[0031]

[0032] Furthermore, in step S4, the construction of the cable joint comprehensive degradation damage assessment index is specifically as follows: according to the cable joint characteristics, a degradation index that can reflect the comprehensive performance of the cable joint should meet the following conditions:

[0033] (1) The degradation process is an irreversible process and is monotonic, i.e., the monotonic parameter;

[0034] (2) The characteristics change with the degradation of cable joint performance and are predictive, i.e., predictive parameters;

[0035] (3) It should be able to resist external noise interference, that is, the signal-to-noise ratio parameter;

[0036] Therefore, the monotonicity parameter, predictability parameter and signal-to-noise ratio parameter are comprehensively selected to construct a suitable comprehensive degradation damage assessment index for cable joints.

[0037] Furthermore, the monotonicity parameter evaluates the increase or decrease trend information of the feature, and the monotonicity parameter calculation formula of the feature population is:

[0038]

[0039] Where n represents the number of observations in the sample.

[0040] Furthermore, the predictive parameter measures the difference in critical failure values ​​in the system population, quantifies the variance score of the final failure value of each feature and the mean of the trend range, and the calculation formula of the predictive parameter is:

[0041]

[0042] Furthermore, the signal-to-noise ratio parameter is the ratio of the power of the desired signal to the power of the undesired noise, and is an indicator for measuring signal quality;

[0043] The calculation formula of the signal-to-noise ratio parameter is:

[0044]

[0045] Among them, P s is the signal power, P n is the noise power.

[0046] Furthermore, for the positive indicators among the three parameters of monotonicity parameter, predictability parameter and signal-to-noise ratio parameter:

[0047]

[0048] For the negative indicators among the three parameters:

[0049]

[0050] Where i represents the three characteristics of monotonicity, predictability and signal-to-noise ratio, i = 1, 2, 3, and j represents the jth indicator in the i-th characteristic;

[0051] The weight of the jth indicator in the i-th characteristic is:

[0052]

[0053] The entropy value of the jth indicator is:

[0054]

[0055] Define the coefficient of variation l of the jth indicator j The larger the coefficient of difference, the greater the effect of the indicator.

[0056] l j =1-γ j

[0057] The comprehensive degradation index of cable joints is:

[0058]

[0059] In this solution, a switch cabinet cable joint performance degradation assessment system based on IoT perception includes:

[0060] Data processing unit: used to obtain the operating data of the switch cabinet cable connector and the IoT sensing multi-parameter monitoring data, and perform signal processing on the collected data to extract the original features;

[0061] Feature selection unit: select features from the original features by using information gain in the decision tree algorithm of random forest, sort the features by gain rate, select features with large information gain rate as a subset, then determine the random forest model accuracy of the subset, and compare it with the model accuracy with the total features, and take the smallest subset with the maximum accuracy as the optimal feature subset;

[0062] Data fusion unit: Use the FPCA method to fuse the characteristic indicators that can reflect the degradation trend, including fitting a curve for the extracted data characterizing the cable joint degradation trend indicators, using the contribution rate of the characteristic value to measure the proportion of the main component, and selecting three parameters, namely, monotonicity parameter, predictive parameter, and signal-to-noise ratio parameter, to construct a suitable cable joint comprehensive degradation damage assessment index. The present invention provides a switch cabinet cable joint performance degradation assessment method and system based on Internet of Things perception, which aims to monitor the current performance status and degradation trend of the cable joint in the switch cabinet, reduce productivity loss and downtime, avoid catastrophic failures, improve the reliability of the switch cabinet operation, ensure the safe and stable operation of the switch cabinet, and thus improve the power supply reliability of the power grid.

[0063] The present invention discloses the following technical effects:

[0064] The present invention makes full use of the operating data in the switch cabinet and the IoT sensing multi-parameter status monitoring data to extract characteristic indicators that can characterize the degradation trend of the cable joint, and uses the data fusion model to construct a degradation evaluation model that can reflect the comprehensive performance of the cable joint. It has the following advantages:

[0065] (1) The method is more comprehensive and can extract data that can reflect the degradation trend from the operation data and IoT sensing monitoring data, and eliminate data that cannot reflect the degradation trend;

[0066] (2) High accuracy: data fusion is used to fuse characteristic indicators that can reflect the degradation trend, which can more accurately reflect the degradation trend of cable joints;

[0067] (3) It is highly practical and can provide operation and maintenance personnel with a more accurate degradation trend model. BRIEF DESCRIPTION OF THE DRAWINGS

[0068] In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings required for use in the embodiments will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative work.

[0069] Figure 1 is a flow chart of the switch cabinet cable joint performance degradation evaluation method of the present invention;

[0070] Figure 2 The present invention is a flow chart of the random forest. DETAILED DESCRIPTION

[0071] To make the purpose, technical scheme and advantages of the embodiments of the present application clearer, the technical scheme in the embodiments of the present application will be clearly and completely described below in conjunction with the drawings in the embodiments of the present application. Obviously, the described embodiments are only part of the embodiments of the present application, rather than all of the embodiments. The components of the embodiments of the present application generally described and shown in the drawings here can be arranged and designed in various different configurations. Therefore, the following detailed description of the embodiments of the present application provided in the drawings is not intended to limit the scope of the application claimed for protection, but merely represents the selected embodiments of the present application. Based on the embodiments of the present application, all other embodiments obtained by those skilled in the art without making creative work belong to the scope of protection of the present application.

[0072] like Figure 1-2 As shown, the switch cabinet cable joint performance degradation assessment method based on IoT perception mentioned in this solution includes the following steps:

[0073] S1, obtain operation data and IoT sensing multi-parameter monitoring data;

[0074] S2, extract characteristic indicators that characterize the degradation trend of cable joints;

[0075] S3, using FPCA method to reduce the dimension of cable joint degradation trend index;

[0076] S4, construct the comprehensive degradation damage assessment index of cable joints.

[0077] In this embodiment, step S1 of acquiring data specifically includes: acquiring IoT sensing multi-parameter monitoring data of voltage, current, temperature and partial discharge according to sensors, performing signal processing in time domain, frequency domain and time-frequency domain on the collected data, and extracting original features.

[0078] In this embodiment, the method for extracting features in step S2 is specifically as follows:

[0079] The extracted high-dimensional feature set may contain redundant features, which may result in long calculation time and excessive complexity. Therefore, the key is to select the best subset of low-dimensional features from the original feature set, that is, the optimal feature index that can characterize the degradation trend of the cable joint.

[0080] according to Figure 2 As shown in the figure, random forest is an embedded feature selection method, which is integrated by decision trees. Each decision tree is a classifier. For an input sample, N trees will have N classification results. Random forest integrates all classification voting results and designates the category with the most votes as the final output.

[0081] Information gain is used to select features in the decision tree algorithm of random forest. The larger the information gain, the better the selectivity of this feature. Information entropy indicates the degree of classification confusion. The more chaotic the classification, the greater the information entropy.

[0082] The calculation formulas for information gain and information entropy are:

[0083]

[0084]

[0085] Among them, p i It means that there are k categories in each data set, and the proportion of each category in the data set;

[0086] The random forest method is improved and the information gain rate calculation formula is defined as:

[0087]

[0088] First, the original features input into the random forest are sorted according to the information gain rate, and the features with large information gain rate are selected as the subset. Then, the accuracy of the random forest model of the subset is determined and compared with the model accuracy with the total features. The smallest subset with the maximum accuracy is taken as the optimal feature subset.

[0089] The random forest method is improved and used to extract characteristic indicators that can reflect the degradation trend of cable joints from switchgear operation data and IoT perception multi-parameter monitoring data.

[0090] In this embodiment, the FPCA method in step S3 is specifically as follows:

[0091] Data in high-dimensional space are usually sparsely distributed. Data dimension reduction can reduce high-dimensional data to low-dimensional data, which can make the data set more convenient to use. At the same time, it can ensure that the variables are independent of each other and reduce the computational cost of the algorithm. Although feature selection was performed to extract indicators that can characterize the deterioration trend of cable joints, in order to fit the curve χ to the extracted data characterizing the deterioration trend of cable joints, i (t), where t∈T, i=1, K, n;

[0092]

[0093] st||μ|| 2 =∫ T μ(t)μ(t)dt=1

[0094] ∫ T μ(t)μ r (t)dt=0,r=1,K,k

[0095] Among them, μ(t) is the weight function, ξ i is the principal component score, ξ i =∫ T μ(t)χ i (t)dt.

[0096] Then the characteristic equation is:

[0097] ∫ T u(t, v)μ(v)dv=λμ(t)

[0098] Where u(t, v) is the covariance function and λ is the eigenvalue.

[0099] The contribution rate of the eigenvalue is used to measure the proportion of the principal component. The calculation formula is:

[0100]

[0101] The characteristic indicators that can reflect the degradation trend are integrated through the FPCA method.

[0102] In this embodiment, in step S4, the construction of the cable joint comprehensive degradation damage assessment index is specifically as follows: according to the cable joint characteristics, a degradation index that can reflect the comprehensive performance of the cable joint should meet the following conditions:

[0103] (1) The degradation process is an irreversible process and is monotonic, i.e., the monotonic parameter;

[0104] (2) The characteristics change with the degradation of cable joint performance and are predictive, i.e., predictive parameters;

[0105] (3) It should be able to resist external noise interference, that is, the signal-to-noise ratio parameter;

[0106] Therefore, the monotonicity parameter, predictability parameter and signal-to-noise ratio parameter are comprehensively selected to construct a suitable comprehensive degradation damage index of cable joints.

[0107] In this embodiment, the monotonicity parameter evaluates the increase or decrease trend information of the feature. Monotonicity ignores the possibility of self-repair of the system. Even if the internal part of the system may self-repair, the monotonicity is still applicable to the entire system.

[0108] The formula for calculating the monotonicity parameter of the feature population is:

[0109]

[0110] Where n represents the number of observations in the sample.

[0111] In this embodiment, the predictive parameter measures the difference in critical failure values ​​in the system population, quantifies the variance score of the final failure value of each feature and the mean of the trend range, and the calculation formula of the predictive parameter is:

[0112]

[0113] In this embodiment, the signal-to-noise ratio parameter is the ratio of the power of the desired signal to the power of the undesired noise, and is an indicator for measuring signal quality;

[0114] The calculation formula of the signal-to-noise ratio parameter is:

[0115]

[0116] Where P s is the signal power, P n is the noise power.

[0117] In this embodiment, for the positive indicators among the three parameters of the monotonicity parameter, the predictability parameter, and the signal-to-noise ratio parameter:

[0118]

[0119] For the negative indicators among the three parameters:

[0120]

[0121] Where i represents the three characteristics of monotonicity, predictability and signal-to-noise ratio, i = 1, 2, 3, and j represents the jth indicator in the i-th characteristic;

[0122] The weight of the jth indicator in the i-th characteristic is:

[0123]

[0124] The entropy value of the jth indicator is

[0125]

[0126] Define the coefficient of variation l of the jth indicator j The larger the coefficient of difference, the greater the effect of the indicator.

[0127] l j =1-γ j

[0128] The comprehensive degradation index of cable joints is:

[0129]

[0130] The improved random forest method proposed in the present invention is used to extract multiple characteristic indicators that can characterize the degradation trend, and then the FPC method is used to reduce the dimension of the above degradation trend indicators, and the data fusion method is used to fuse them to construct a cable joint comprehensive degradation damage assessment index, which has strong practical value.

[0131] This embodiment also provides a switch cabinet cable joint performance degradation assessment system based on IoT perception, including:

[0132] Data processing unit: used to obtain the operating data of the switch cabinet cable connector and the IoT sensing multi-parameter monitoring data, and perform signal processing on the collected data to extract the original features;

[0133] Feature selection unit: select features from the original features by using information gain in the decision tree algorithm of random forest, sort the features by gain rate, select features with large information gain rate as a subset, then determine the random forest model accuracy of the subset, and compare it with the model accuracy with the total features, and take the smallest subset with the maximum accuracy as the optimal feature subset;

[0134] Data fusion unit: The FPCA method is used to fuse the characteristic indicators that can reflect the degradation trend, including fitting curves for the extracted data characterizing the cable joint degradation trend indicators, using the contribution rate of the characteristic value to measure the proportion of the main component, and selecting three parameters, namely monotonicity parameter, predictive parameter, and signal-to-noise ratio parameter, to construct a suitable cable joint comprehensive degradation damage assessment index.

[0135] The present invention is described with reference to flowcharts and / or block diagrams of methods, devices (systems), and computer program products according to embodiments of the present invention. It should be understood that each process and / or block in the flowchart and / or block diagram, as well as the combination of processes and / or blocks in the flowchart and / or block diagram, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, a special-purpose computer, an embedded processor, or other programmable data processing device to produce a machine, so that the instructions executed by the processor of the computer or other programmable data processing device generate instructions for implementing the processes in the flowchart and / or block diagram. Figure 1 A process or multiple processes and / or boxes Figure 1 A device that provides the functions specified in a block or multiple blocks.

[0136] In the description of the present invention, it should be understood that the terms "first" and "second" are used for descriptive purposes only and should not be understood as indicating or implying relative importance or implicitly indicating the number of technical features indicated. Thus, a feature defined as "first" or "second" may explicitly or implicitly include one or more of the features. In the description of the present invention, the meaning of "plurality" is two or more, unless otherwise clearly and specifically defined.

[0137] Obviously, those skilled in the art can make various changes and modifications to the present invention without departing from the spirit and scope of the present invention. Thus, if these modifications and variations of the present invention fall within the scope of the claims of the present invention and their equivalents, the present invention is also intended to include these modifications and variations.

Claims

1. A switch cabinet cable joint performance degradation assessment method based on IoT perception, characterized in that: The following steps are involved: S1, obtaining the operation data and monitoring data of the switch cabinet cable connector; S2, extracting characteristic indicators for characterizing the degradation trend of the cable joint based on the operation data and the monitoring data; S3, using the FPCA method to perform dimension reduction processing on the characteristic index; S4, based on the feature index after dimensionality reduction, by setting constraint conditions, obtaining an evaluation index for characterizing the degradation damage of the cable joint, and performing degradation evaluation on the switch cabinet cable joint according to the positive index and the negative index of the evaluation index; In step S2, based on the improved random forest decision tree algorithm, characteristic indicators are extracted according to information gain and information entropy, wherein the calculation formulas of the information gain and information entropy are respectively: Among them, p k It means that there are k categories in each data set; The improvement process of the random forest decision tree algorithm includes the following steps: The original features input into the random forest are sorted according to the information gain rate, and the features with large information gain rate are selected as the subset, where the information gain rate is expressed as: Determine the accuracy of the random forest model of the subset, and compare it with the accuracy of the model with the total features, and take the smallest subset with the maximum accuracy as the optimal feature subset; The FPCA method in step S3 is specifically as follows: fitting the curve x to the extracted data representing the cable joint degradation trend index i (t), where t∈Τ, i=1,...,n; s.t.||μ|| 2 =∫ Τ μ(t)μ(t)dt=1 ∫ Τ μ(t)μ r (t)dt=0,r=1,...,k Among them, μ(t) is the weight function, ξ i is the principal component score, ξ i =∫ Τ μ(t)χ i (t)dt; Then the characteristic equation is: Where u(t,v) is the covariance function and λ is the eigenvalue; The contribution rate of the eigenvalue is used to measure the proportion of the main component. The calculation formula is: In step S4, the construction of the cable joint comprehensive degradation damage assessment index is specifically as follows: according to the cable joint characteristics, the conditions satisfied by the degradation index obtained include: The monotonicity parameter condition is used to indicate that the degradation process is an irreversible process; Predictive parameter conditions to represent the change in characteristics as cable joint performance deteriorates; The signal-to-noise ratio parameter condition is used to indicate the ability to resist external noise interference; An index satisfying the monotonicity parameter condition, the predictive parameter condition and the signal-to-noise ratio parameter condition is obtained as a comprehensive degradation damage assessment index for a cable joint.

2. The switch cabinet cable joint performance degradation assessment method based on IoT perception according to claim 1 is characterized in that: In step S1, the monitoring data includes: voltage, current, temperature and partial discharge data.

3. The switch cabinet cable joint performance degradation assessment method based on IoT perception according to claim 1 is characterized in that: The monotonicity parameter evaluates the increase or decrease trend information of the feature, and the monotonicity parameter calculation formula of the feature population is: Where n represents the number of observations in the sample.

4. The switch cabinet cable joint performance degradation assessment method based on IoT perception according to claim 3 is characterized in that: The predictive parameter measures the difference in critical failure values ​​in the system population, quantifies the variance fraction of the final failure value of each feature and the mean of the trend range, and the calculation formula of the predictive parameter is:

5. The switch cabinet cable joint performance degradation assessment method based on IoT perception according to claim 3 is characterized in that: The signal-to-noise ratio parameter is the ratio of the power of the desired signal to the power of the undesired noise, and is an indicator for measuring signal quality; The calculation formula of the signal-to-noise ratio parameter is: Where P s is the signal power, P n is the noise power.

6. The switch cabinet cable joint performance degradation assessment method based on IoT perception according to claim 3 is characterized in that: For the positive indicators in the monotonicity parameter condition, predictive parameter condition, and signal-to-noise ratio parameter condition: For the medium negative indicators of three parameter conditions: Where i represents the three characteristics of monotonicity, predictability and signal-to-noise ratio, i = 1, 2, 3, and j represents the jth indicator in the i-th characteristic; The weight of the jth indicator in the i-th characteristic is: The entropy value of the jth indicator is: Define the coefficient of variation l of the jth indicator j , the larger the coefficient of difference, the greater the effect of the indicator; l j =1-c j The comprehensive degradation index of cable joints is:

7. A switch cabinet cable joint performance degradation assessment system based on IoT perception, characterized in that: include: Data processing unit: used to obtain the operating data of the switch cabinet cable connector and the IoT sensing multi-parameter monitoring data, and perform signal processing on the collected data to extract the original features; Feature selection unit: select features from the original features by using information gain in the decision tree algorithm of random forest, sort the features by gain rate, select features with large information gain rate as a subset, then determine the accuracy of the random forest model of the subset, and compare it with the accuracy of the model with total features, and take the smallest subset with the maximum accuracy as the optimal feature subset; Data fusion unit: Use the FPCA method to fuse the characteristic indicators that can reflect the degradation trend, fit the curve to the extracted data characterizing the cable joint degradation trend indicators, use the contribution rate of the characteristic value to measure the proportion of the main component, and select the degradation damage assessment index of the cable joint based on the monotonicity parameter conditions, predictive parameter conditions, and signal-to-noise ratio parameter conditions to evaluate the performance degradation of the switch cabinet cable joint.

Images