Method and system for detecting abnormity of wiring terminal of metering junction box

Abstract

The invention relates to the technical field of junction box anomaly detection, in particular to a metering junction box wiring terminal anomaly detection method and system. The method comprises the following steps: firstly, based on voltage drop deviation between different output voltages of voltage terminals, screening out suspected normal terminals from all the voltage terminals; then acquiring a temperature anomaly coefficient and an input anomaly coefficient of each group of suspected normal terminals at each moment, and further combining change correlation of the input voltage and the temperature anomaly coefficient to determine a to-be-analyzed terminal; and further evaluating the output abnormal coefficient of each group of terminals to be analyzed, and determining a normal terminal by combining the time sequence change correlation between the output voltage of each output end and the terminal temperature and the input abnormal coefficient. According to the invention, on the basis of the circuit characteristics of the voltage terminals and in combination with the change relevance between the temperature and the voltage, the voltage terminals are discussed according to conditions so as to accurately analyze and evaluate the abnormity possibility of each group of voltage terminals in real time, and the abnormity detection effect of the wiring terminals of the metering junction box is improved.

Description

Technical Field

[0001] The present invention relates to the technical field of abnormal detection of junction boxes, and particularly to a method and system for detecting abnormal connection terminals of a metering junction box. Background Art

[0002] A metering junction box is an important device in a power metering system, and its main function is to connect and transmit signals such as current and voltage in the power system for power metering, monitoring, and control. A connection terminal is a key mechanical component in the junction box responsible for connecting wires. During use, problems such as poor contact, looseness, or aging may occur, which may cause serious safety accidents. Therefore, it is crucial to detect abnormalities and perform regular maintenance on the connection terminals.

[0003] Currently, it is possible to analyze whether there is an abnormal fault in the connection terminals of the junction box by monitoring the real-time electrical data of the connection terminals. However, the electrical data changes little before the connection terminal fails, making it difficult to distinguish from the electrical data fluctuations during normal operation. Potential fault signs such as small electrical data fluctuations may also be masked by environmental factors or system noise, resulting in the inability to detect potential fault hazards in a timely manner, and thus the poor detection effect of the abnormal connection terminals of the metering junction box. Summary of the Invention

[0004] In order to solve the technical problem of the poor detection effect of the abnormal connection terminals of the metering junction box, the purpose of the present invention is to provide a method and system for detecting abnormal connection terminals of a metering junction box, and the specific technical solutions adopted are as follows: A method for detecting abnormal connection terminals of a metering junction box, the method comprising: In a metering junction box, obtain the terminal temperature, input voltage, and output voltage of each output terminal at each moment for each group of voltage terminals, as well as the ambient temperature; Determine suspected normal terminals from all voltage terminals according to the deviation of the output voltage of different output terminals of each group of voltage terminals relative to the input voltage at each moment; For each group of suspected normal terminals, at each moment, obtain a temperature anomaly coefficient according to the terminal temperature and its deviation relative to the ambient temperature; obtain an input anomaly coefficient for each group of suspected normal terminals according to the change in the input voltage and the temperature anomaly coefficient, and combine the sequential change correlation of the input voltage and the temperature anomaly coefficient to determine normal terminals and terminals to be analyzed from all suspected normal terminals; For each group of terminals to be analyzed, based on the output voltage changes of each output terminal in terms of time sequence, and their associations with the input voltage and terminal temperature changes respectively, combined with the temperature anomaly coefficient, an output anomaly coefficient is obtained; based on the time sequence associations between the output voltage of each output terminal and the terminal temperature of each group of terminals to be analyzed, and the time sequence associations between the input voltage and the temperature anomaly coefficient, combined with the input anomaly coefficient and the output anomaly coefficient, normal terminals are determined from all terminals to be analyzed.

[0005] Further, the method for determining the suspected normal terminals includes: At each moment, based on the difference between the output voltage and the input voltage of each output terminal of each group of voltage terminals, the voltage drop value of each output terminal is obtained, and based on the voltage drop value of each output terminal and the difference between the voltage drop values of different output terminals, the anomaly parameter of each group of voltage terminals is obtained; For each group of voltage terminals, the moment when the anomaly parameter is greater than the preset parameter threshold is taken as the anomaly moment, and when the total number of anomaly moments within the preset historical period is less than the preset number threshold, the voltage terminals are determined as suspected normal terminals.

[0006] Further, the method for obtaining the anomaly parameter includes: For each group of voltage terminals, the mean value of the voltage drop values of all output terminals is multiplied by the range of the voltage drop values of all output terminals, and the normalized value of the product is used as the anomaly parameter.

[0007] Further, the method for obtaining the temperature anomaly coefficient includes: At each moment, the difference between the terminal temperature and the ambient temperature of each group of suspected normal terminals is taken as the temperature deviation, and the sum of the temperature deviation and the terminal temperature is used as the temperature anomaly coefficient of the suspected normal terminals.

[0008] Further, the method for obtaining the input anomaly coefficient includes: The maximum change voltage between adjacent moments of the input voltage of each group of suspected normal terminals is taken as the first input anomaly parameter; the maximum temperature anomaly coefficient is taken as the second input anomaly parameter; the first input anomaly parameter and the second input anomaly parameter are fused to obtain the input anomaly coefficient of each group of suspected normal terminals.

[0009] Further, the method for determining the normal terminals and the terminals to be analyzed includes: For each group of suspected normal terminals, based on the input voltage and the temperature anomaly coefficient at each moment, an input voltage sequence and a temperature anomaly coefficient sequence are respectively constructed, and the Pearson correlation coefficient between the two sequences is used as the first confidence parameter; When the first confidence parameter is greater than a preset confidence threshold and the input anomaly coefficient is less than a preset anomaly threshold, determine the suspected normal terminal as a normal terminal, and use the remaining suspected normal terminals other than the normal terminal as terminals to be analyzed.

[0010] Further, the method for obtaining the output anomaly coefficient includes: According to the temporal variation correlation between the output voltage and the terminal temperature of each output terminal of each group of terminals to be analyzed, and the similarity of the fluctuation characteristics of the output voltage and the input voltage of each output terminal in terms of time sequence, obtain the anomaly reference weight of each output terminal; Take the maximum change voltage between the output voltages of each output terminal of each group of terminals to be analyzed at all adjacent moments as the first sub-anomaly parameter of each output terminal; take the maximum temperature anomaly coefficient as the second sub-anomaly parameter; fuse the first sub-anomaly parameter and the second sub-anomaly parameter to obtain the sub-anomaly coefficient of each output terminal; Use the anomaly reference weight of each output terminal to weight the corresponding sub-anomaly coefficient, and take the normalized result of the weighted sum value of all output terminals as the output anomaly coefficient of the corresponding terminal to be analyzed.

[0011] Further, the method for obtaining the anomaly reference weight includes: For each group of terminals to be analyzed, between adjacent moments, according to the difference between the change voltage of the output voltage and the change voltage of the input voltage of each output terminal, obtain the voltage difference and sort the voltage differences in chronological order to construct the voltage difference sequence of each output terminal, and construct the terminal temperature difference sequence according to the difference between the terminal temperatures at adjacent moments; take the Pearson correlation coefficient between the voltage difference sequence and the terminal temperature difference sequence of each output terminal as the first normal parameter of each output terminal; For each group of terminals to be analyzed, take the mean value of the change voltages of the output voltage of each output terminal at all adjacent moments and the standard deviation of all output voltages as vector elements to construct the fluctuation characteristic vector of each output terminal; take the mean value of the change voltages of the input voltage at all adjacent moments and the standard deviation of all input voltages as vector elements to construct the input fluctuation characteristic vector; take the cosine similarity between the input fluctuation characteristic vector and the fluctuation characteristic vector of each output terminal as the second normal parameter of each output terminal; Fuse the first normal parameter and the second normal parameter, and take the negative correlation mapping result of the fusion result as the anomaly reference weight of each output terminal.

[0012] Further, the method for determining normal terminals from all terminals to be analyzed includes: When the first confidence parameter is greater than a preset confidence threshold and the input anomaly coefficient is greater than a preset anomaly threshold, it is determined that the terminal to be analyzed is an abnormal terminal, and the remaining terminals to be analyzed except the abnormal terminal are used as target terminals; For each group of target terminals, the mean value of the first normal parameters of all output terminals is used as the second confidence parameter. The second confidence parameter and the first confidence parameter are fused, and the negative correlation mapping result of the fusion result is used as the abnormal confidence weight; The output anomaly coefficient and the input anomaly coefficient are fused, and the fusion result is weighted by the abnormal confidence weight. The normalized value of the weighted result is used as the anomaly index of the target terminal; The target terminals with an anomaly index less than the preset index threshold are used as normal terminals.

[0013] The present invention also provides an abnormal detection system for the wiring terminals of a metering junction box, including a memory, a processor, and a computer program stored in the memory and executable on the processor. When the processor executes the computer program, the steps of the abnormal detection method for the wiring terminals of the metering junction box are implemented.

[0014] The present invention has the following beneficial effects: The present invention first preliminarily screens out suspected normal terminals from all voltage terminals based on the voltage drop deviation between the input voltage and different output voltages of each group of voltage terminals in the metering junction box; then for each group of suspected normal terminals, the temperature anomaly coefficient of each group of suspected normal terminals at each moment is analyzed and evaluated, and based on the characteristics that abnormal conditions such as poor contact at the input end of the voltage terminal may lead to similar voltage drops at different output ends, transient changes in the input voltage and terminal temperature, the input anomaly coefficient of each group of suspected normal terminals is obtained, and the time-series change correlation between the input voltage and the temperature anomaly coefficient is combined to provide a confidence reference, taking into account the large fluctuations in the input voltage itself, so as to accurately determine the normal terminals and the terminals to be analyzed from all suspected normal terminals; further for each group of terminals to be analyzed, based on the theoretical basis that equal pressure between input and output and temperature affect resistance and thus affect voltage, the change correlation between the output voltage change of each output terminal in time series and the input voltage and terminal temperature changes is analyzed, and at the same time, combined with the characteristics that abnormal conditions such as poor contact at the output end of the voltage terminal may lead to transient changes in the output voltage and terminal temperature, the output anomaly coefficient of each group of terminals to be analyzed is comprehensively evaluated; then the time-series change correlation between the output voltage and terminal temperature of each output terminal of each group of terminals to be analyzed, and the time-series change correlation between the input voltage and temperature anomaly coefficient are analyzed, and combined with the input anomaly coefficient and output anomaly coefficient, normal terminals are determined from all terminals to be analyzed, and then all abnormal terminals in the metering junction box are determined. The present invention is based on the circuit characteristics of the voltage terminals, and combines the change correlation between temperature and voltage, and discusses different situations of the voltage terminals to analyze and evaluate the anomaly possibility of each group of voltage terminals in real time and accurately, improving the effect of abnormal detection of the wiring terminals of the metering junction box. Description of the Drawings

[0015] To more clearly illustrate the technical solutions and advantages in the embodiments of the present invention or the prior art, the drawings required for the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings in the following description are only some embodiments of the present invention. For those of ordinary skill in the art, without creative efforts, other drawings can be obtained based on these drawings.

[0016] Figure 1 Flowchart of a method for detecting abnormalities in the terminal of a metering junction box provided by an embodiment of the present invention; Figure 2 Schematic diagram of a metering junction box provided by an embodiment of the present invention; Figure 3 Flowchart of a method for obtaining an output abnormality coefficient provided by an embodiment of the present invention. Detailed Embodiments

[0017] To further elaborate on the technical means and effects adopted by the present invention to achieve the intended invention purpose, the following, in conjunction with the drawings and preferred embodiments, details the specific embodiments, structures, features, and effects of a method and system for detecting abnormalities in the terminals of a metering junction box proposed according to the present invention. In the following description, different "one embodiment" or "another embodiment" do not necessarily refer to the same embodiment. In addition, the specific features, structures, or characteristics in one or more embodiments can be combined in any suitable form.

[0018] Unless otherwise defined, all technical and scientific terms used herein have the same meaning as commonly understood by those skilled in the technical field to which the present invention belongs.

[0019] The following specifically describes the specific solutions of a method and system for detecting abnormalities in the terminals of a metering junction box provided by the present invention with reference to the drawings.

[0020] Please refer to Figure 1 , which shows a flowchart of a method for detecting abnormalities in the terminals of a metering junction box provided by an embodiment of the present invention, specifically including: Step S1, in the metering junction box, obtain the terminal temperature, input voltage, and output voltage of each output terminal at each moment for each group of voltage terminals, as well as the ambient temperature.

[0021] It should be noted that the abnormality detection in the embodiments of the present invention is only for the voltage terminals with a single-input and multiple-output structure in the metering junction box. The abnormality detection methods for voltage terminals with different structures are the same. Here, only the voltage terminals with a one-in-three-out structure are taken as an example for analysis and description; Please refer to Figure 2, which shows a schematic diagram of a metering junction box provided by an embodiment of the present invention. Figure 2 It contains a total of 7 groups of terminals, including 4 groups of voltage terminals and 3 groups of current terminals. Each small rectangle corresponds to a group of terminals, and each group of terminals contains terminal paddles. Figure 2 The terminals with vertical terminal paddles in are voltage terminals. The voltage terminals are one-in-three-out, that is, one input terminal and three output terminals.

[0022] For real-time anomaly detection of the metering junction box, in an embodiment of the present invention, temperature sensors are first set at the terminal paddles of each group of voltage terminals in the metering junction box, voltage sensors are respectively set on the power transmission lines of the input terminal and the output terminals, and environmental sensors are installed in the local area of the metering junction box, such as within a radius of 5 meters; the acquisition frequency of all sensors is set to 50Hz, and the implementer can also customize it; then during the operation of the metering junction box, the temperature sensors are used to collect the terminal temperatures of each group of voltage terminals in real time, the voltage sensors are used to collect the input voltage of each group of voltage terminals at the input terminal and the output voltage of each output terminal in real time, and the environmental sensors are used to collect the environmental temperature in real time.

[0023] It should be noted that the implementer can also deploy corresponding sensors according to the type of voltage terminals and actual applications, and can also use an infrared thermal imager to collect the terminal temperature; the selection, deployment and application of various sensors are all prior arts and will not be elaborated here.

[0024] It should be noted that in an embodiment of the present invention, the acquisition data within a preset historical period, such as the historical 3 minutes, at the current detection moment is used as the analysis basis, that is, the terminal temperature, input voltage and output voltage of each output terminal of each group of voltage terminals at each moment within the historical 3 minutes are obtained to analyze and evaluate the anomaly possibility of each group of voltage terminals at the current moment; each moment mentioned in the subsequent description is a moment within the preset historical period and will not be elaborated again; in other embodiments, the implementer can also customize the preset historical period, but it should not be too large to avoid being difficult to capture the data fluctuation characteristics.

[0025] Step S2, determine the suspected normal terminals from all the voltage terminals according to the deviation of the output voltage of different output terminals of each group of voltage terminals relative to the input voltage at each moment.

[0026] Considering that in voltage terminals, the voltage signal at the input terminal is usually distributed to multiple output terminals in a parallel manner, theoretically the output voltage of all output terminals should be the same as the input voltage at the input terminal; however, in actual applications, factors such as the wire impedance in the voltage terminals and the contact resistance of the connection screws may cause a small voltage drop at the output terminals, but since the electrical connection paths of different output terminals are similar, the differences in wire impedance and contact resistance are small, and the voltage drop levels should also be relatively similar. Based on this, in the embodiments of the present invention, within a preset historical period, according to the deviation of the output voltage of different output ends of each group of voltage terminals relative to the input voltage at each moment, the voltage drop situation of different output ends will be analyzed, so as to first screen out abnormal terminals from all voltage terminals, and then use the remaining voltage terminals as suspected normal terminals for further abnormal analysis.

[0027] Preferably, in an embodiment of the present invention, considering that at each moment, if the voltage drop difference between different output ends in each group of voltage terminals is greater, it indicates a greater possibility of abnormality, and the evaluation result at a single moment may have a certain degree of contingency. If the number of abnormal moments within the preset historical period is more, it indicates that the abnormality is persistent and the possibility of abnormality of the voltage terminal is greater. Therefore, the method for determining suspected normal terminals includes: At each moment, according to the difference between the output voltage and the input voltage of each output end of each group of voltage terminals, obtain the voltage drop value of each output end, and according to the voltage drop value of each output end and the difference between the voltage drop values of different output ends, obtain the abnormal parameter of each group of voltage terminals. For each group of voltage terminals, take the moment when the abnormal parameter is greater than the preset parameter threshold as the abnormal moment. When the total number of abnormal moments within the preset historical period is less than the preset number threshold, determine that the voltage terminal is a suspected normal terminal.

[0028] Among them, in a preferred embodiment of the present invention, considering that the average value of the voltage drop values of all output ends in the voltage terminal can reflect the degree of voltage drop, and the range of the voltage drop values can reflect the difference in the degree of voltage drop. When the degree of voltage drop is greater and the difference in the degree of voltage drop is greater, it indicates a greater possibility of abnormality of this group of voltage terminals. Therefore, the method for obtaining the abnormal parameter includes: For each group of voltage terminals, multiply the average value of the voltage drop values of all output ends by the range of the voltage drop values of all output ends, and take the normalized value of the product as the abnormal parameter.

[0029] As an example, for each output end of each group of voltage terminals, subtract the output voltage from the input voltage and then divide by the input voltage to obtain the voltage drop value of each output end, and further obtain the abnormal parameter of each group of voltage terminals; set the preset parameter threshold to 0.5 to determine all abnormal moments within the preset historical period; then set the preset number threshold to 20% of the total number of moments within the preset historical period to determine whether the voltage terminal is a suspected normal terminal; the voltage terminals other than the suspected normal terminals will be directly used as abnormal terminals in the junction box.

[0030] It should be noted that in other examples, the implementer can also define the preset parameter threshold and the preset number threshold by himself; in other embodiments, the implementer can also use variance instead of range to obtain the abnormal parameter.

[0031] Step S3: For each group of suspected normal terminals, at each moment, obtain the temperature anomaly coefficient according to the terminal temperature and its deviation from the ambient temperature; obtain the input anomaly coefficient of each group of suspected normal terminals according to the input voltage and the change of the temperature anomaly coefficient, and combine the sequential change relationship between the input voltage and the temperature anomaly coefficient to determine the normal terminals and the terminals to be analyzed from all the suspected normal terminals.

[0032] Considering that when there are abnormal conditions such as poor contact at the input end of the voltage terminal, the voltage drop degrees at different output ends may still remain highly similar. Based on the abnormal evaluation method in Step S2, there may be misjudgments, that is, there may still be abnormal terminals among the suspected normal terminals, and further analysis and evaluation are required.

[0033] Since poor contact at the input end of the voltage terminal will cause a significant increase in the terminal temperature, but the change range of the ambient temperature is relatively low in the short term, it can provide a certain reference for the analysis of the change range of the terminal temperature, and then help accurately evaluate whether the terminal temperature has an abnormal increase; Therefore, in the embodiment of the present invention, first, at each moment, obtain the temperature anomaly coefficient of each group of suspected normal terminals according to the terminal temperature of each group of suspected normal terminals and its deviation from the ambient temperature; the temperature anomaly coefficient evaluates whether each group of voltage terminals is abnormal from the perspective of thermal effect change, and prepares for the subsequent analysis of the possibility of abnormal terminals among the suspected normal terminals.

[0034] Preferably, in an embodiment of the present invention, the method for obtaining the temperature anomaly coefficient includes: At each moment, take the difference between the terminal temperature and the ambient temperature of each group of suspected normal terminals as the temperature deviation, and take the sum of the temperature deviation and the terminal temperature as the temperature anomaly coefficient of the suspected normal terminals.

[0035] It should be noted that the difference is specifically measured in the form of the absolute value of the difference. During the operation of the voltage terminal, the terminal temperature is usually higher than the ambient temperature, and the temperature deviation can also be directly measured in the form of the difference.

[0036] Considering that poor contact at the input end of the voltage terminal may cause a significant increase in the contact resistance of the terminal, which in turn leads to a large instantaneous fluctuation in the input voltage. At the same time, the joule heat will continue to accumulate after the poor contact, and the terminal temperature will increase significantly within a certain period of time; also considering that the temperature anomaly coefficient not only reflects the abnormal possibility of the suspected normal terminals, but also the change situation of the temperature anomaly coefficient in time sequence can reflect the change situation of the terminal temperature; Therefore, the embodiment of the present invention will obtain the input abnormality coefficient of each group of suspected normal terminals according to the changes in the input voltage and temperature abnormality coefficient; the input abnormality coefficient comprehensively evaluates the abnormal possibility of poor contact at the input end of the suspected normal terminal from two angles: the instantaneous fluctuation of the input voltage and the increase in the terminal temperature, so as to prepare for further screening and evaluation.

[0037] Preferably, in one embodiment of the present invention, considering the input voltage difference between adjacent moments of each group of suspected normal terminals in a preset historical period, the instantaneous change degree can be evaluated. The greater the instantaneous change of the input voltage, the greater the abnormal possibility of poor contact of the input terminal; considering that the higher the temperature anomaly coefficient, the higher the corresponding terminal temperature, the greater the possibility of poor contact of the input terminal; therefore, the method for obtaining the input anomaly coefficient includes: The maximum change voltage of the input voltage of each group of suspected normal terminals between all adjacent moments is used as the first input abnormality parameter; the maximum temperature abnormality coefficient is used as the second input abnormality parameter; the first input abnormality parameter and the second input abnormality parameter are combined to obtain the input abnormality coefficient of each group of suspected normal terminals.

[0038] It should be noted that the change voltage is the difference between the input voltage at each moment in the preset historical period and the input voltage at the previous adjacent moment, and the maximum change voltage reflects the maximum instantaneous change of the input voltage; the maximum temperature anomaly coefficient is the maximum value of the temperature anomaly coefficients at all moments in the preset historical period.

[0039] As an example, the first input abnormality parameter and the second input abnormality parameter are multiplied and combined, and the product is linearly normalized to obtain the input abnormality coefficient of each group of suspected normal terminals; in other examples, the implementer may also use basic mathematical operations such as addition or weighted summation to fuse the two.

[0040] Considering that the input voltage may have large fluctuations before being input to the voltage terminal, such fluctuations not caused by terminal abnormalities may also cause the input abnormality coefficient of the suspected normal terminal to be large. Therefore, the accuracy of evaluating the abnormal possibility of the suspected normal terminal based solely on the input abnormality coefficient is not high; Considering that there is a certain linear approximate relationship between temperature and resistance, changes in terminal temperature may cause changes in contact resistance in the terminal, thereby causing changes in input voltage. At the same time, changes in the temperature anomaly coefficient reflect changes in terminal temperature. Therefore, there is a certain change correlation between the temperature anomaly coefficient and the input voltage. The violent fluctuation of the input voltage itself may further aggravate the increase in terminal temperature, thereby causing a relative deviation in the change correlation between the terminal temperature and the temperature anomaly coefficient and the input voltage. Therefore, after obtaining the input anomaly coefficient of each group of suspected normal terminals in the embodiments of the present invention, the temporal variation correlation between the input voltage and the temperature anomaly coefficient will be further combined to determine the normal terminals and the terminals to be analyzed from all the suspected normal terminals; the temporal variation correlation between the input voltage and the temperature anomaly coefficient can provide a certain confidence reference, taking into account the violent fluctuations of the input voltage itself, so as to accurately analyze the anomaly possibility of the suspected normal terminals.

[0041] Preferably, in an embodiment of the present invention, considering that the Pearson correlation coefficient can reflect the variation correlation between time series data, the first confidence parameter of the input anomaly coefficient of each group of suspected normal terminals can be obtained, and then the anomaly possibility of each suspected normal terminal can be evaluated in combination with the input anomaly coefficient for screening; therefore, the method for determining the normal terminals and the terminals to be analyzed includes: For each group of suspected normal terminals, according to the input voltage and the temperature anomaly coefficient at each moment, an input voltage sequence and a temperature anomaly coefficient sequence are respectively constructed, and the Pearson correlation coefficient between the two sequences is used as the first confidence parameter; When the first confidence parameter is greater than the preset confidence threshold and the input anomaly coefficient is less than the preset anomaly threshold, it is determined that the suspected normal terminal is a normal terminal, and the remaining suspected normal terminals except the normal terminals are used as the terminals to be analyzed.

[0042] As an example, for each group of abnormal normal terminals, the input voltages at each moment within the preset historical period are sorted in chronological order to construct an input voltage sequence, and similarly, a temperature anomaly coefficient sequence is constructed, and then the first confidence parameter is obtained; then the preset confidence threshold is set to 0.8, and the preset anomaly threshold is set to 0.8, so as to determine all the normal terminals and the terminals to be analyzed among the suspected normal terminals.

[0043] It should be noted that the implementer can also use the DTW similarity or other correlation metric parameters to replace the Pearson correlation coefficient, which are both existing technologies and will not be elaborated here.

[0044] It should be noted that when the preset confidence threshold is greater than 0.8, the possibility of violent fluctuations of the input voltage itself is excluded. On this basis, the normal terminals and abnormal terminals among the suspected normal terminals can be accurately screened according to the input anomaly coefficient. Among them, the suspected normal terminals with a preset anomaly threshold less than 0.8 are normal terminals, and those greater than or equal to 0.8 are abnormal terminals; when the preset confidence threshold is less than or equal to 0.8, the fluctuations of the input voltage itself will interfere with the accuracy of the anomaly analysis and evaluation, and further analysis and evaluation are required, that is, the terminals to be analyzed include voltage terminals whose anomaly possibility is difficult to accurately analyze and evaluate due to the violent fluctuations of the input voltage itself, namely the target terminals described later, and the embodiments of the present invention will further analyze and evaluate in step S4.

[0045] Step S4. For each group of terminals to be analyzed, based on the output voltage changes of each output terminal in terms of time sequence, and their associations with the input voltage and terminal temperature changes respectively, combined with the temperature anomaly coefficient, obtain the output anomaly coefficient; based on the time-sequence associations between the output voltage of each output terminal and the terminal temperature of each group of terminals to be analyzed, and the time-sequence associations between the input voltage and the temperature anomaly coefficient, combined with the input anomaly coefficient and the output anomaly coefficient, determine the normal terminals from all the terminals to be analyzed.

[0046] Considering that among the terminals to be analyzed, even if the input voltage fluctuates violently, if there are similar fluctuation changes between the input voltage and the output voltage of each output terminal, it indicates a lower possibility of abnormality; also considering that the change in terminal temperature will also cause changes in the output voltage of different output terminals, when there is a stronger change correlation between the terminal temperature and the output voltage, it indicates a lower possibility of abnormality for the terminals to be analyzed; further considering that in each group of terminals to be analyzed, if there are abnormalities such as poor contact in each output terminal, there will also be large instantaneous changes in the output voltage, and its temperature anomaly coefficient will also be larger, and the possibility of abnormality of the output terminal will also be greater; Therefore, in the embodiment of the present invention, for each group of terminals to be analyzed, based on the output voltage changes of each output terminal in terms of time sequence, and their associations with the input voltage and terminal temperature changes respectively, combined with the temperature anomaly coefficient, obtain the output anomaly coefficient; the output anomaly coefficient reflects the possibility of abnormality of the output terminals of the terminals to be analyzed, and prepares for further screening of the normal terminals among the terminals to be analyzed.

[0047] Preferably, in an embodiment of the present invention, the method for obtaining the output anomaly coefficient includes: Please refer to Figure 3 , which shows a flowchart of a method for obtaining an output anomaly coefficient provided by an embodiment of the present invention, specifically including: Step S401. Based on the time-sequence associations between the output voltage of each output terminal and the terminal temperature of each group of terminals to be analyzed, and the similarity of the fluctuation characteristics between the output voltage of each output terminal and the input voltage in terms of time sequence, obtain the abnormal reference weight of each output terminal.

[0048] Considering that the change values of the voltage and terminal temperature of each output terminal of the terminals to be analyzed at adjacent moments in terms of time sequence can effectively reveal the dynamic interaction between the voltage change and the temperature change, provide a more accurate dynamic evaluation for the time-sequence association between the output voltage and the terminal temperature, and thus the evaluation of the abnormality possibility of the terminals to be analyzed is also more accurate; and the Pearson correlation coefficient can reflect the change association situation between time-sequence data, so the first normal parameter of each output terminal of the terminals to be analyzed can be obtained; Also considering that the mean value of the voltage change values of the input voltage and the output voltage and the standard deviation of the voltage at adjacent moments can, to a certain extent, reflect their fluctuation characteristics. After constructing a fluctuation feature vector with the mean value and variance as vector elements, by evaluating the cosine similarity between vectors, the similarity of the fluctuation characteristics of the output voltage and the input voltage at each output terminal in terms of time series can be reflected, and then the second normal parameter can be obtained; Also considering that both the first normal parameter and the second normal parameter reflect the normal possibility of the terminal to be analyzed, so after fusion, a negative correlation mapping adjustment logic is required to reflect the abnormal possibility of each output terminal, that is, to obtain the abnormal reference weight, and the abnormal reference weight provides a reference basis for analyzing the output abnormal coefficient of the terminal to be analyzed subsequently; Based on this, in a preferred embodiment of the present invention, the method for obtaining the abnormal reference weight includes: For each group of terminals to be analyzed, between adjacent moments, according to the difference between the change voltage of the output voltage of each output terminal and the change voltage of the input voltage, obtain the voltage difference and sort the voltage differences in chronological order to construct a voltage difference sequence for each output terminal, and construct a terminal temperature difference sequence according to the difference between the terminal temperatures at adjacent moments; the normalized value of the Pearson correlation coefficient between the voltage difference sequence of each output terminal and the terminal temperature difference sequence is used as the first normal parameter of each output terminal; For each group of terminals to be analyzed, use the mean value of the change voltages of the output voltage of each output terminal between all adjacent moments and the standard deviation of all output voltages as vector elements to construct a fluctuation feature vector for each output terminal; use the mean value of the change voltages of the input voltage between all adjacent moments and the standard deviation of all input voltages as vector elements to construct an input fluctuation feature vector; the normalized value of the cosine similarity between the input fluctuation feature vector and the fluctuation feature vector of each output terminal is used as the second normal parameter of each output terminal; Fuse the first normal parameter and the second normal parameter, and use the negative correlation mapping result of the fusion result as the abnormal reference weight of each output terminal.

[0049] As an example, for each output terminal of each group of terminals to be analyzed, first subtract the difference between the output voltage at each moment in the preset historical period and the output voltage at the previous adjacent moment as the output change voltage at each moment; similarly, obtain the input change voltage and the terminal temperature difference at each moment; take the difference between the output change voltage and the input change voltage at the same moment as the voltage difference, sort the voltage differences in chronological order to construct a voltage difference sequence; sort the terminal temperature differences in chronological order to construct a terminal temperature difference sequence; then obtain the Pearson correlation coefficient, and use the Pearson correlation coefficient as x in is normalized so that the value is 0 - 1. The closer x is to 1, the larger the first normal parameter of each output terminal; Similarly, the cosine similarity between the input fluctuation feature vector and the fluctuation feature vector of each output terminal is used as the y in is normalized so that the value ranges from 0 to 1. The closer y is to 1, the larger the second normal parameter of each output terminal. Then, after multiplying and combining the first normal parameter and the second normal parameter, a reciprocal operation is performed after adding the preset non-zero normal constant 0.001 to avoid a denominator of 0. The logic is adjusted and normalized to obtain the abnormal reference weight corresponding to the output terminal.

[0050] In other examples, the implementer can also adopt other normalization means and negative correlation mapping means, and can also fuse the two by using basic mathematical operations such as addition or weighted summation.

[0051] Step S402: Take the maximum change voltage between the output voltages of each output terminal of each group of terminals to be analyzed at all adjacent moments as the first sub-abnormal parameter of each output terminal; take the maximum temperature abnormal coefficient as the second sub-abnormal parameter; fuse the first sub-abnormal parameter and the second sub-abnormal parameter to obtain the sub-abnormal coefficient of each output terminal.

[0052] Based on the same acquisition logic and method as the input abnormal parameter in step S3, obtain the sub-abnormal coefficient of each output terminal in each group of terminals to be analyzed. The sub-abnormal coefficient reflects the possibility of abnormalities such as poor contact at each output terminal, and prepares for the subsequent comprehensive evaluation of the output abnormal coefficient of the terminals to be analyzed.

[0053] As an example, taking any output terminal as an example, take the difference between the output voltage at each moment in the preset historical period and the output voltage at the previous adjacent moment as the change voltage, and then obtain the maximum change voltage to get the first sub-abnormal parameter; take the maximum temperature abnormal coefficient of the terminal to be analyzed in the preset historical period as the second sub-abnormal parameter; multiply and combine the first sub-abnormal parameter and the second sub-abnormal parameter, and perform linear normalization on the product to obtain the sub-abnormal coefficient of the corresponding output terminal of the terminal to be analyzed; in other examples, the implementer can also fuse the two by using basic mathematical operations such as addition or weighted summation.

[0054] Step S403: Weight the corresponding sub-abnormal coefficient by the abnormal reference weight of each output terminal, and take the normalized result of the weighted sum value of all output terminals as the output abnormal coefficient of the corresponding terminal to be analyzed.

[0055] It should be noted that weighted summation is already an existing technology and will not be elaborated here; specifically, linear normalization is adopted, and the implementer can also adopt other normalization means.

[0056] Considering that the temporal variation correlation between the output voltage of each output terminal of the terminal to be analyzed and the terminal temperature, as well as the temporal variation correlation between the input voltage and the temperature anomaly coefficient, can both provide a confidence reference for the normal voltage fluctuation in the terminal to be analyzed and also provide a reference for the abnormality of the terminal to be analyzed; and both the input anomaly coefficient and the output anomaly coefficient reflect the fault possibility of the voltage terminal; Therefore, in the embodiment of the present invention, based on the temporal variation correlation between the output voltage of each output terminal of each group of terminals to be analyzed and the terminal temperature, as well as the temporal variation correlation between the input voltage and the temperature anomaly coefficient, and combining the input anomaly coefficient and the output anomaly coefficient, normal terminals are determined from all the terminals to be analyzed.

[0057] Preferably, in an embodiment of the present invention, considering that the larger the first confidence parameter is, the more certain the fluctuation correlation between the input voltage and the temperature anomaly parameter is, and on this basis, the confidence level of the input anomaly coefficient analyzed and evaluated is also higher. Therefore, all the abnormal terminals in the terminals to be analyzed can be preliminarily screened out based on the input anomaly coefficient, and the remaining terminals are used as target terminals for further analysis; Also considering that the first normal parameter reflects the temporal variation correlation between the output voltage of each output terminal of the terminal to be analyzed and the terminal temperature, and the first confidence coefficient reflects the temporal variation correlation between the input voltage and the temperature anomaly coefficient, both of which reflect the normal possibility of the terminal to be analyzed. Therefore, after fusing them and performing a negative correlation mapping adjustment logic, it can be used as a reference for evaluating abnormalities; furthermore, the abnormality possibility of the target terminals can be comprehensively evaluated by combining the output anomaly coefficient and the input anomaly coefficient, and normal terminals can be screened out; Based on this, the method for determining normal terminals from all the terminals to be analyzed includes: When the first confidence parameter is greater than the preset confidence threshold and the input anomaly coefficient is greater than the preset anomaly threshold, it is determined that the terminal to be analyzed is an abnormal terminal, and the remaining terminals to be analyzed except the abnormal terminals are used as target terminals; For each group of target terminals, the mean value of the first normal parameters of all output terminals is used as the second confidence parameter, the second confidence parameter and the first confidence parameter are fused, and the negative correlation mapping result of the fusion result is used as the abnormality confidence weight; The output anomaly coefficient and the input anomaly coefficient are fused, the fusion result is weighted by the abnormality confidence weight, and the normalized value of the weighted result is used as the abnormality index of the target terminal; the target terminals with the abnormality index less than the preset index threshold are used as normal terminals.

[0058] As an example, first set the preset confidence threshold to 0.8 and the preset anomaly threshold to 0.8 as well, and obtain the abnormal terminals and target terminals in the terminals to be analyzed; then calculate the average of the second confidence parameter and the first confidence parameter of the target terminals, and use the average as the x in the exponential function exp(-x) with the natural constant e as the base, and obtain the abnormal confidence weight through negative correlation mapping; then add the output anomaly coefficient and the input anomaly coefficient, multiply the sum by the abnormal confidence weight and perform linear normalization to obtain the anomaly index of the target terminals; further set the preset index threshold to 0.8 to screen the normal terminals in the target terminals, and the remaining terminals in the target terminals except the normal terminals are the abnormal terminals.

[0059] In other examples, the implementer can also adopt other fusion means such as multiplication or weighted summation, can also customize various thresholds, and can also adopt other negative correlation mapping means such as taking the reciprocal operation, which will not be elaborated here.

[0060] In an embodiment of the present invention, based on the continuous screening in steps S2 - S4, the abnormal terminals in all voltage terminals can be determined; when there is any abnormal terminal, the metering junction box is abnormal, and the numbers of the voltage terminals with abnormal faults can be returned. For example, V1, V2, V3, and V4 represent voltage terminals. When a voltage terminal is detected to be abnormal, the corresponding warning light is lit according to the severity of the abnormality. Through the communication system of the power grid such as power line carrier communication or wireless communication, the abnormal number information is sent to the terminal device of the maintenance personnel.

[0061] The present invention also proposes a metering junction box terminal abnormal detection system, including a memory, a processor, and a computer program stored in the memory and executable on the processor. When the processor executes the computer program, the steps of the above-mentioned metering junction box terminal abnormal detection method are implemented.

[0062] In summary, the present invention first preliminarily screens out suspected normal terminals from all voltage terminals based on the voltage drop deviation between the input voltage and different output voltages of each group of voltage terminals in the metering junction box; then for each group of suspected normal terminals, analyzes and evaluates the temperature anomaly coefficient of each group of suspected normal terminals at each moment, and obtains the input anomaly coefficient of each group of suspected normal terminals. Further, in combination with the temporal variation correlation between the input voltage and the temperature anomaly coefficient, accurately determines the normal terminals and the terminals to be analyzed from all suspected normal terminals; for each group of terminals to be analyzed, first evaluates the output anomaly coefficient of each group of terminals to be analyzed, and then analyzes the temporal variation correlation between the output voltage and the terminal temperature of each output terminal of each group of terminals to be analyzed, and the temporal variation correlation between the input voltage and the temperature anomaly coefficient. Combining the input anomaly coefficient and the output anomaly coefficient, determines the normal terminals from all terminals to be analyzed. The present invention is based on the circuit characteristics of the voltage terminals, and combines the variation correlation between temperature and voltage, and discusses each group of voltage terminals in different cases to analyze and evaluate the anomaly possibility of each group of voltage terminals in real time and accurately, improving the anomaly detection effect of the wiring terminals of the metering junction box.

[0063] It should be noted that the above sequence of embodiments of the present invention is only for description and does not represent the superiority or inferiority 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 result. In some embodiments, multitasking and parallel processing are also possible or may be advantageous.

[0064] Each embodiment in this specification is described in a progressive manner, and the same or similar parts between the embodiments can be referred to each other. The key point of each embodiment is to illustrate the differences from other embodiments.

Claims

1. A method for detecting abnormality of a terminal block of a metering junction box, characterized in that: The method comprises: In the metering junction box, the terminal temperature of each group of voltage terminals, the input voltage and the output voltage of each output terminal at each moment, and the ambient temperature are obtained; According to the deviation of the output voltage of different output ends of each group of voltage terminals relative to the input voltage at each moment, the suspected normal terminal is determined from all the voltage terminals; For each group of suspected normal terminals, at each moment, the temperature anomaly coefficient is obtained according to the terminal temperature and its deviation from the ambient temperature; the input anomaly coefficient of each group of suspected normal terminals is obtained according to the change of the input voltage and the temperature anomaly coefficient, and the normal terminals and the terminals to be analyzed are determined from all the suspected normal terminals in combination with the temporal change correlation of the input voltage and the temperature anomaly coefficient; For each group of terminals to be analyzed, the output abnormality coefficient is obtained according to the output voltage change of each output terminal in the timing, and the association with the changes of the input voltage and the terminal temperature respectively, combined with the temperature abnormality coefficient; according to the association between the output voltage of each output terminal of each group of terminals to be analyzed and the timing change of the terminal temperature, and the timing change association between the input voltage and the temperature abnormality coefficient, combined with the input abnormality coefficient and the output abnormality coefficient, the normal terminals are determined from all the terminals to be analyzed.

2. A method for detecting abnormality of a connection terminal of a metering junction box according to claim 1, characterized in that: The method for determining the suspected normal terminal includes: At each moment, according to the difference between the output voltage and the input voltage of each output end of each group of voltage terminals, a voltage drop value of each output end is obtained, and according to the voltage drop value of each output end and the difference between the voltage drop values ​​of different output ends, an abnormal parameter of each group of voltage terminals is obtained; For each group of voltage terminals, the moment when the abnormal parameter is greater than a preset parameter threshold is taken as the abnormal moment, and when the total number of the abnormal moments in a preset historical period is less than a preset number threshold, the voltage terminal is determined to be a suspected normal terminal.

3. A method for detecting abnormality of a connection terminal of a metering junction box according to claim 2, characterized in that: The method for obtaining the abnormal parameters includes: For each group of voltage terminals, the mean of the voltage drop values ​​of all output terminals is multiplied by the range of the voltage drop values ​​of all output terminals, and a normalized value of the product is used as an abnormal parameter.

4. A method for detecting abnormality of a connection terminal of a metering junction box according to claim 1, characterized in that: The method for obtaining the temperature anomaly coefficient includes: At each moment, the difference between the terminal temperature of each group of suspected normal terminals and the ambient temperature is taken as the temperature deviation, and the sum of the temperature deviation and the terminal temperature is taken as the temperature anomaly coefficient of the suspected normal terminals.

5. A method for detecting abnormality of a connection terminal of a metering junction box according to claim 1, characterized in that: The method for obtaining the input abnormal coefficient includes: The maximum change voltage of the input voltage of each group of suspected normal terminals between all adjacent moments is used as the first input abnormality parameter; the maximum temperature abnormality coefficient is used as the second input abnormality parameter; the first input abnormality parameter and the second input abnormality parameter are combined to obtain the input abnormality coefficient of each group of suspected normal terminals.

6. A method for detecting abnormality of a connection terminal of a metering junction box according to claim 1, characterized in that: The method for determining the normal terminals and the terminals to be analyzed includes: For each group of suspected normal terminals, the input voltage sequence and the temperature anomaly coefficient sequence are constructed according to the input voltage and the temperature anomaly coefficient at each moment, and the Pearson correlation coefficient between the two sequences is used as the first confidence parameter; When the first confidence parameter is greater than a preset confidence threshold and the input abnormality coefficient is less than a preset abnormality threshold, the suspected normal terminal is determined to be a normal terminal, and the remaining suspected normal terminals except the normal terminal are taken as terminals to be analyzed.

7. A method for detecting abnormality of a connection terminal of a metering junction box according to claim 6, characterized in that: The method for obtaining the output abnormal coefficient includes: According to the correlation between the output voltage of each output terminal of each group of terminals to be analyzed and the change of the terminal temperature in time series, and the similarity of the fluctuation characteristics of the output voltage of each output terminal and the input voltage in time series, the abnormal reference weight of each output terminal is obtained; The maximum change voltage of the output voltage of each output terminal of each group of terminals to be analyzed between all adjacent moments is used as the first sub-abnormal parameter of each output terminal; the maximum temperature abnormal coefficient is used as the second sub-abnormal parameter; the first sub-abnormal parameter and the second sub-abnormal parameter are combined to obtain the sub-abnormal coefficient of each output terminal; The corresponding sub-abnormality coefficient is weighted by using the abnormality reference weight of each output terminal, and the normalized result of the weighted sum value of all output terminals is used as the output abnormality coefficient of the corresponding terminal to be analyzed.

8. A method for detecting abnormality of a connection terminal of a metering junction box according to claim 7, characterized in that: The method for obtaining the abnormal reference weight includes: For each group of terminals to be analyzed, between adjacent moments, according to the difference between the change voltage of the output voltage of each output terminal and the change voltage of the input voltage, the voltage difference is obtained and the voltage difference is sorted in time sequence to construct a voltage difference sequence of each output terminal, and according to the difference between the terminal temperatures at adjacent moments, a terminal temperature difference sequence is constructed; the Pearson correlation coefficient between the voltage difference sequence of each output terminal and the terminal temperature difference sequence is used as the first normal parameter of each output terminal; For each group of terminals to be analyzed, the mean of the change voltage of the output voltage of each output terminal between all adjacent moments and the standard deviation of all output voltages are used as vector elements to construct the fluctuation characteristic vector of each output terminal; the mean of the change voltage of the input voltage between all adjacent moments and the standard deviation of all input voltages are used as vector elements to construct the input fluctuation characteristic vector; the cosine similarity between the input fluctuation characteristic vector and the fluctuation characteristic vector of each output terminal is used as the second normal parameter of each output terminal; The first normal parameter and the second normal parameter are fused, and a negative correlation mapping result of the fusion result is used as an abnormal reference weight of each output end.

9. A method for detecting abnormality of a connection terminal of a metering junction box according to claim 8, characterized in that: The method for determining a normal terminal from all terminals to be analyzed comprises: When the first confidence parameter is greater than a preset confidence threshold and the input abnormality coefficient is greater than a preset abnormality threshold, the terminal to be analyzed is determined to be an abnormal terminal, and the remaining terminals to be analyzed except the abnormal terminal are taken as target terminals; For each group of target terminals, taking the mean of the first normal parameters of all output terminals as the second confidence parameter, fusing the second confidence parameter and the first confidence parameter, and taking the negative correlation mapping result of the fusion result as the abnormal confidence weight; The output anomaly coefficient and the input anomaly coefficient are fused, the fusion result is weighted using the anomaly confidence weight, and the normalized value of the weighted result is used as the anomaly index of the target terminal; the target terminal whose anomaly index is less than the preset index threshold is regarded as a normal terminal.

10. A metering junction box terminal abnormality detection system, characterized in that: The invention comprises a memory, a processor and a computer program stored in the memory and executable on the processor, wherein when the processor executes the computer program, the steps of a method for detecting abnormality of a connection terminal of a metering junction box as claimed in any one of claims 1 to 9 are implemented.

Images