Transformer area low-voltage cause diagnosis method, device, equipment and medium
By utilizing historical data and Pareto principle sorting methods in the diagnosis of low voltage in transformer substations, the causes of low voltage can be accurately located, solving the problems of low efficiency, high cost, and poor targeting in existing technologies, and achieving efficient and accurate cause diagnosis.
Patent Information
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2025-12-04
- Publication Date
- 2026-03-10
AI Technical Summary
Existing technologies are inefficient, costly, and lack specificity in diagnosing low voltage causes in transformer substations. They are difficult to accurately pinpoint the main causes and rely on detailed line parameters, making them difficult to promote. Additionally, the addition of data acquisition terminals increases implementation costs.
By using historical normal data of the target transformer area to calculate quantitative indicators of various causes of low voltage, and combining them with the Pareto principle for sorting, abnormal indicator values are determined, comparable object nodes are divided for horizontal comparison, and the target causes of low voltage are screened out.
It achieves efficient and accurate diagnosis of low voltage causes, reduces implementation costs, avoids the one-sidedness of single-dimensional judgment, and improves the accuracy and efficiency of cause location.
Smart Images
Figure CN121637332A_ABST
Abstract
Description
TECHNICAL FIELD
[0001] The present application relates to the technical field of power distribution network, in particular to a low-voltage cause diagnosis method, device, equipment and medium for a transformer area. BACKGROUND
[0002] Low-voltage problem is a common and stubborn problem in the operation of transformer area, and its causes are complex and diverse, which can be mainly summarized into four categories: excessive user load, low power factor, unreasonable network structure (such as too long power supply radius and too thin wire diameter), and three-phase imbalance. At present, the diagnosis of low-voltage causes mainly depends on the personal experience of operation and maintenance personnel, or only a simple single threshold (such as whether the voltage value is lower than 198V) is used for judgment. Such methods have the following disadvantages: first, the efficiency is low and the accuracy is insufficient, it is difficult to accurately locate the dominant cause, resulting in low pertinence of treatment measures; second, although the traditional model analysis method based on power flow calculation can perform quantitative analysis, it often depends on detailed line parameters, and these parameters (especially 0.4kV low-voltage line parameters) are often difficult to obtain comprehensively in actual engineering, which limits its popularization and application; in addition, some methods may need to additionally install a collection terminal, which increases the implementation cost.
[0003] In summary, how to optimize the low-voltage cause diagnosis method for a transformer area to solve the problems of low diagnosis efficiency, high cost and poor pertinence is a technical problem to be solved at present. SUMMARY
[0004] Therefore, the purpose of the present application is to provide a low-voltage cause diagnosis method, device, equipment and medium for a transformer area, which can optimize the low-voltage cause diagnosis method for a transformer area to solve the problems of low diagnosis efficiency, high cost and poor pertinence. The specific solutions are as follows:
[0005] In the first aspect, the present application provides a low-voltage cause diagnosis method for a transformer area, comprising:
[0006] If it is detected that the current target transformer area meets the preset low-voltage condition, the historical index values of each preset quantitative index are calculated using the historical normal data of the target transformer area; the historical normal data are historical original data when the preset low-voltage condition is not met; the preset quantitative index is a quantitative index corresponding to each type of low-voltage cause;
[0007] For each monitored object node of the target transformer area, the corresponding average index value is calculated using the historical index value, and based on the current index value of the monitored object node and the corresponding average index value, it is judged whether the current index value is abnormal relative to its history;
[0008] dividing the monitored object node into several groups of comparable object nodes, and sorting the current indicator value of the same preset quantitative indicator for each of the several groups of comparable object nodes respectively to determine whether the current indicator value is abnormal relative to the target area according to a first sorting result obtained;
[0009] If the current indicator value is abnormal relative to both the historical data and the target area, it is determined that the current indicator value is an abnormal indicator value, and the preset quantitative indicators are sorted based on the abnormal indicator value using the Pareto principle to determine the target low-voltage cause of the target area from the various low-voltage causes according to a second sorting result obtained.
[0010] Optionally, the monitored object node includes user nodes and line nodes in the target area, and a district node representing the target area; the preset quantitative indicators include load current, average power factor, unit power voltage loss coefficient, and three-phase current unbalance degree, the load current being a quantitative indicator corresponding to a low-voltage cause related to user load, the average power factor being a quantitative indicator corresponding to a low-voltage cause related to power factor, the unit power voltage loss coefficient being a quantitative indicator corresponding to a low-voltage cause related to network architecture rationality, and the three-phase current unbalance degree being a quantitative indicator corresponding to a low-voltage cause related to three-phase unbalance degree; the load current, the unit power voltage loss coefficient, and the three-phase current unbalance degree are all of high-value abnormal type, the high-value abnormal type being a type in which the numerical value is positively correlated with the probability of abnormality; the average power factor is of low-value abnormal type, the low-value abnormal type being a type in which the numerical value is negatively correlated with the probability of abnormality.
[0011] Optionally, if it is detected that the current target area meets a preset low-voltage condition, historical indicator values of each preset quantitative indicator are calculated using historical normal data of the target area, including:
[0012] If it is detected that the user voltage of the current target area is lower than a preset voltage threshold, it is determined that the target area meets a preset low-voltage condition, and historical normal data of the target area in a preset historical time period is determined; the historical normal data is historical original data corresponding to a user voltage not lower than the preset voltage threshold;
[0013] At the node level of the user nodes, current data and average power factor data in the historical normal data are determined, the current data is determined as the historical indicator value of the load current, and the average power factor data is determined as the historical indicator value of the average power factor;
[0014] At the node level of the line node, line-related data in the historical normal data is determined, and a preset unit power voltage loss calculation formula is calculated based on the line-related data and the average power factor data, so that a calculation result obtained is determined as a historical index value of the unit power voltage loss coefficient;
[0015] At the node level of the transformer node, three-phase current-related data in the historical normal data is determined, and a preset three-phase current unbalance degree calculation formula is calculated based on the three-phase current-related data, so that a calculation result obtained is determined as a historical index value of the three-phase current unbalance degree.
[0016] Optionally, the judging whether the current index value is abnormal relative to the historical index value of the monitored object node includes:
[0017] Real-time original data of the target transformer area is collected, and for each monitored object node, a current index value of each preset quantitative index is calculated based on the real-time original data;
[0018] If the index type of the current index value is the high-value abnormal type, when the current index value is greater than a first preset multiple of the corresponding average index value, it is determined that the current index value is abnormal relative to the historical index value.
[0019] If the index type of the current index value is the low-value abnormal type, when the current index value is less than a second preset multiple of the corresponding average index value, it is determined that the current index value is abnormal relative to the historical index value. The second preset multiple is the reciprocal of the first preset multiple.
[0020] Optionally, the dividing the monitored object nodes into a plurality of comparable object nodes includes:
[0021] The user nodes in a plurality of the monitored object nodes located in the same power supply branch are divided into the same group of comparable object nodes, the line nodes located in the same branch are divided into the same group of comparable object nodes, and the line nodes located in the same main road are divided into the same group of comparable object nodes, so as to obtain a plurality of groups of comparable object nodes.
[0022] Optionally, the sorting the current index values of the same preset quantitative index includes:
[0023] If the index type of the preset quantitative index is the high-value abnormal type, the current index values of the same preset quantitative index are sorted in descending order to obtain a corresponding first sorting result, a target quantile is determined according to the first sorting result, and a target index value higher than the target quantile in the current index values is determined to be abnormal relative to the target station area;
[0024] If the index type of the preset quantitative index is the low-value abnormal type, the current index values of the same preset quantitative index are sorted in ascending order to obtain a corresponding first sorting result, a target quantile is determined according to the first sorting result, and a target index value lower than the target quantile in the current index values is determined to be abnormal relative to the target station area.
[0025] Optionally, the sorting of the preset quantitative index based on the abnormal index value according to the Pareto principle to determine the target low-voltage cause of the target station area from the various low-voltage causes according to the obtained second sorting result includes:
[0026] The abnormal occurrence times corresponding to the preset quantitative index are counted based on the abnormal index value, and the preset quantitative index is sorted according to the abnormal occurrence times to obtain a corresponding second sorting result;
[0027] The abnormal contribution proportion corresponding to each preset quantitative index is calculated using the abnormal occurrence times, and the abnormal contribution proportions are sequentially accumulated according to the second sorting result to obtain a corresponding cumulative contribution proportion, until the cumulative contribution proportion is higher than a preset proportion threshold;
[0028] The index combination of the preset quantitative index corresponding to the cumulative contribution proportion is determined, and the low-voltage cause corresponding to the index combination is determined as the target low-voltage cause of the target station area.
[0029] In a second aspect, the present application provides a low-voltage cause diagnosis device for a station area, comprising:
[0030] An index value calculation module is configured to calculate historical index values of each preset quantitative index using historical normal data of a target station area if it is detected that the target station area satisfies a preset low-voltage condition; the historical normal data is historical original data when the preset low-voltage condition is not satisfied; and the preset quantitative index is a quantitative index corresponding to various low-voltage causes;
[0031] anomaly judgment module, configured to, for each monitored object node of the target transformer area, calculate a corresponding average index value by using the historical index value, and judge whether the current index value is abnormal relative to the historical index value of the monitored object node based on the current index value of the monitored object node and the corresponding average index value;
[0032] index value sorting module, configured to divide the monitored object nodes into several groups of comparable object nodes, and sort the current index values of the same preset quantitative index for each group of the several groups of comparable object nodes respectively, to determine whether the current index value is abnormal relative to the target transformer area according to a first sorting result obtained;
[0033] index sorting module, configured to, if the current index value is abnormal relative to the historical index value and the target transformer area, determine that the current index value is an abnormal index value, sort the preset quantitative index based on the abnormal index value by using a Pareto principle, and determine a target low-voltage cause of the target transformer area from the various low-voltage causes according to a second sorting result obtained.
[0034] In a third aspect, the present application provides an electronic device, comprising:
[0035] a memory, configured to save a computer program;
[0036] a processor, configured to execute the computer program to implement the transformer low-voltage cause diagnosis method described above.
[0037] In a fourth aspect, the present application provides a computer readable storage medium, configured to save a computer program; wherein the computer program is executed by a processor to implement the transformer low-voltage cause diagnosis method described above.
[0038] In the present application, if it is detected that the current target station area satisfies the preset low-voltage condition, the historical index values of each preset quantitative index corresponding to various low-voltage causes are calculated using the historical normal data of the target station area; the historical normal data are historical original data when the preset low-voltage condition is not satisfied; the preset quantitative index is a quantitative index corresponding to each type of low-voltage cause; for each monitored object node of the target station area, the corresponding average index value is calculated using the historical index values, and based on the current index value of the monitored object node and the corresponding average index value, it is judged whether the current index value is abnormal relative to its own history; the monitored object nodes are divided into several groups of comparable object nodes, and for each group of the several groups of comparable object nodes, the current index values of the same preset quantitative index are sorted respectively, so as to judge whether the current index value is abnormal relative to the target station area according to the obtained first sorting result; if the current index value is abnormal relative to its own history and the target station area, it is determined that the current index value is an abnormal index value, and the preset quantitative index is sorted based on the abnormal index value using the Pareto principle, so as to determine the target low-voltage cause of the target station area from the various low-voltage causes according to the obtained second sorting result. As can be seen from the above, when it is detected that the target station area satisfies the preset low-voltage condition, the historical index values of each preset quantitative index corresponding to various low-voltage causes are calculated using the historical normal data of the target station area when the preset low-voltage condition is not satisfied, for each monitored object node of the target station area, the corresponding average index value is obtained through the historical index values, and the current index value of the monitored object node is compared with the average index value to judge whether the current index value is abnormal relative to its own history, at the same time, the monitored object nodes are divided into several groups of comparable object nodes, the current index values of the same preset quantitative index in each group are sorted, and it is judged whether the current index value is abnormal relative to the target station area according to the first sorting result, if the current index value is abnormal in both the comparison with its own history and the comparison within the station area, it is determined to be an abnormal index value, and then the preset quantitative index is sorted according to the abnormal index value using the Pareto principle, and the target low-voltage cause of the target station area is determined from the various low-voltage causes according to the second sorting result. In this way, through the above process of the present application, combined with the double abnormality judgment logic of the self-history data and the horizontal comparison within the station area, and combined with the core cause screening mechanism of the Pareto principle, the key index deviating from the normal level and different from the overall state of the station area can be accurately positioned, and the core low-voltage cause can be focused through priority sorting, the one-dimensional judgment is effectively avoided, the accuracy and efficiency of cause positioning are improved, and the low-voltage cause diagnosis method of the station area is optimized to solve the problems of low efficiency, high cost and poor pertinence of cause diagnosis. BRIEF DESCRIPTION OF DRAWINGS
[0039] To more clearly illustrate the technical solutions in the embodiments of the present invention or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are only embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on the provided drawings without creative effort.
[0040] Figure 1 This application discloses a flowchart of a method for diagnosing the causes of low voltage in a transformer substation.
[0041] Figure 2 This is a flowchart and timing diagram of a method for diagnosing the causes of low voltage in a transformer substation disclosed in this application.
[0042] Figure 3 This is a schematic diagram of a low voltage cause diagnosis device for a transformer substation disclosed in this application.
[0043] Figure 4 This is a structural diagram of an electronic device disclosed in this application. Detailed Implementation
[0044] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.
[0045] Currently, the diagnosis of low voltage causes largely relies on the personal experience of maintenance personnel or is based solely on a simple threshold (such as whether the voltage value is below 198V). These methods have the following drawbacks: First, they are inefficient and inaccurate, making it difficult to precisely pinpoint the dominant cause, resulting in less targeted remedial measures. Second, while traditional power flow calculation-based model analysis methods can perform quantitative analysis, they often depend on detailed line parameters, which (especially for 0.4kV low-voltage lines) are often difficult to obtain comprehensively in actual engineering projects, limiting their widespread application. Furthermore, some methods may require additional data acquisition terminals, increasing implementation costs.
[0046] To overcome the above-mentioned technical problems, this application provides a method for diagnosing the causes of low voltage in transformer substations, which can optimize the method for diagnosing the causes of low voltage in transformer substations and solve the problems of low efficiency, high cost and poor pertinence in cause diagnosis.
[0047] See Figure 1 As shown, this embodiment of the invention discloses a method for diagnosing the causes of low voltage in a transformer substation, including:
[0048] Step S11: If the current target transformer area is detected to meet the preset low voltage conditions, the historical index value of each preset quantitative index is calculated using the historical normal data of the target transformer area; the historical normal data is the historical raw data when the preset low voltage conditions are not met; the preset quantitative index is the quantitative index corresponding to various causes of low voltage.
[0049] In this embodiment, when the current target transformer area is detected to meet the preset low voltage conditions, the historical raw data (i.e., historical normal data) when the target transformer area does not meet the preset low voltage conditions is used as the basis to calculate the historical index value of each preset quantitative indicator corresponding to various causes of low voltage. These various causes of low voltage include user load, power factor, network architecture rationality, and three-phase imbalance. For example... Figure 2 The diagram shown is a flowchart and timing diagram of a method for diagnosing the causes of low voltage in a transformer substation provided in this application.
[0050] It should be noted that the monitored nodes include user nodes and line nodes within the target distribution area, as well as the distribution area nodes that characterize the target distribution area. Before conducting low voltage cause diagnosis in the distribution area, it is necessary to first construct an indicator system covering four major categories of core low voltage causes to determine the causes of low voltage. The indicators for low voltage causes are modeled, and specific, operable quantitative evaluation indicators and calculation formulas are defined for each type of cause. The preset quantitative indicators include load current, average power factor, voltage loss coefficient per unit power, and three-phase current imbalance. Load current is a quantitative indicator corresponding to low voltage causes related to user load; average power factor is a quantitative indicator corresponding to low voltage causes related to power factor; voltage loss coefficient per unit power is a quantitative indicator corresponding to low voltage causes related to network architecture rationality; and three-phase current imbalance is a quantitative indicator corresponding to low voltage causes related to three-phase imbalance. The load current, voltage loss coefficient per unit power, and three-phase current imbalance indicators are all of the high-value anomaly type, where the value is positively correlated with the probability of an anomaly. The average power factor indicator is of the low-value anomaly type, where the value is negatively correlated with the probability of an anomaly. In other words, regarding user load, in rural areas, due to seasonal load surges (such as during busy farming seasons and holidays), distribution transformers frequently experience overload. Some distribution transformers are fully loaded or even overloaded during the busy farming season, causing a significant voltage drop. Load current is directly related to power; therefore, it can be selected as the load factor, i.e., a quantitative indicator of the user load, specifically a high-value anomaly. The formula for expressing the load current is as follows:
[0051] ;
[0052] Wherein, I represents the load current; P represents the total active power; and U represents the voltage. The power factor angle, The power factor is characterized. For the power factor, the average power factor is directly used as the indicator. This indicator reflects the proportion of reactive power in the system. When the proportion of inductive load is large, this value is low, which leads to increased reactive power flow and increased voltage loss. In other words, the average power factor is considered an abnormally low value. The specific formula for expressing the power factor is as follows:
[0053] ;
[0054] Here, Q represents the total reactive power. Regarding the rationality of the network architecture, its core parameters are the power supply radius and conductor diameter. Voltage drop is positively correlated with transmission power, power supply radius, and line impedance (directly related to conductor diameter). Therefore, the voltage loss coefficient per unit power can be selected as a quantitative indicator for network structure modeling and analysis, and it is a high-value anomaly type. Regarding three-phase imbalance, the common three-phase current imbalance is used as an indicator, which is also a high-value anomaly type. Three-phase load imbalance leads to neutral point potential shift, forming an asymmetrical operating state and exacerbating voltage deviation.
[0055] Specifically, if the user voltage in the current target distribution area is detected to be lower than a preset voltage threshold, the target distribution area is determined to meet the preset low voltage condition, and the historical normal data of the target distribution area within a preset historical time period is determined; the historical normal data is the historical raw data of the corresponding user voltage not lower than the preset voltage threshold; at the node level of the user node, the current data and average power factor data in the historical normal data are determined, and the current data is determined as the historical index value of the load current, and the average power factor data is determined as the historical index value of the average power factor; at the node level of the line node, the line-related data in the historical normal data are determined, and a preset unit power voltage loss calculation formula is calculated based on the line-related data and the average power factor data, so that the obtained calculation result is determined as the historical index value of the unit power voltage loss coefficient; at the node level of the distribution area node, the three-phase current-related data in the historical normal data are determined, and a preset three-phase current imbalance calculation formula is calculated based on the three-phase current-related data, so that the obtained calculation result is determined as the historical index value of the three-phase current imbalance. That is, when the user voltage in the current target transformer area is detected to be lower than a preset voltage threshold, such as 198V, it is determined that the target transformer area meets the preset low voltage condition, and historical raw data of user voltage not being lower than the threshold within a preset historical time period is extracted as historical normal data. Subsequently, for indicators directly related to user electricity consumption behavior, calculations are performed at the user node level. That is, at the node level of the user node, current data and average power factor data are determined from the historical normal data and used as historical indicator values of the load current and the average power factor, respectively. At the node level of the line node, line-related data are extracted from the historical normal data, and combined with the average power factor data, the historical indicator value of the unit power voltage loss coefficient is obtained through a preset unit power voltage loss calculation formula. The specific preset unit power voltage loss calculation formula is as follows:
[0056] ;
[0057] in, Characterizes the voltage loss coefficient per unit power; Characterizing voltage loss. The specific formula for expressing voltage loss is as follows:
[0058] ;
[0059] Where R is resistance and X is reactance; The power factor angle; The line is represented by the voltage at the beginning of the line; L is the line length; r represents the resistance per unit length; and x represents the reactance per unit length. At the node level of the transformer substation, the three-phase current related data from the historical normal data are obtained. The historical index value of the three-phase current imbalance is determined using a preset three-phase current imbalance calculation formula, which is as follows:
[0060] ;
[0061] in, The three-phase current imbalance; This represents the maximum value of the three-phase current; This represents the average value of the three-phase current. The specific formula for expressing the average value of the three-phase current is as follows:
[0062] ;
[0063] in, , , These represent the currents of phases A, B, and C in a three-phase circuit, respectively. In this way, the historical values of each preset quantitative indicator calculated in this embodiment provide a historical normal benchmark for subsequent analysis of low-voltage issues in the target transformer area. This ensures that the quantitative indicators corresponding to various causes of low voltage have comparable reference standards, effectively avoiding biases in cause analysis due to a lack of historical references. By extracting data hierarchically at multiple nodes (user, line, transformer area) and calculating the historical values of core indicators in a targeted manner, the key influencing dimensions related to low voltage can be comprehensively covered, making the identification of subsequent abnormal indicators and the location of low-voltage causes more targeted and reliable.
[0064] Step S12: For each monitored node in the target area, calculate the corresponding average index value using the historical index value, and determine whether the current index value is abnormal relative to its own history based on the current index value of the monitored node and the corresponding average index value.
[0065] In this embodiment, for each monitored node in the target area, the corresponding average index value is first calculated using the historical index values as a normal benchmark for judging whether the index is abnormal during low voltage periods, providing a basis for subsequent comparisons. Then, by comparing the current index value of the node with the corresponding average index value, it is determined whether the current index value is abnormal relative to its own historical operating status.
[0066] It should be noted that the process for determining whether the current indicator value is abnormal relative to its historical value is as follows: Real-time raw data of the target area is collected, and for each monitored node, the current indicator value of each preset quantitative indicator is calculated based on the real-time raw data. If the indicator type of the current indicator value is the high-value anomaly type, then when the current indicator value is greater than a first preset multiple of the corresponding average indicator value, it is determined that the current indicator value is abnormal relative to its historical value. If the indicator type of the current indicator value is the low-value anomaly type, then when the current indicator value is less than a second preset multiple of the corresponding average indicator value, it is determined that the current indicator value is abnormal relative to its historical value. The second preset multiple is the reciprocal of the first preset multiple. In other words, real-time raw data of the target area is collected, and for each monitored node, the current indicator value of each preset quantitative indicator is calculated based on the real-time raw data. The current indicator value is compared with the corresponding average indicator value. If there is a significant deterioration (e.g., the real-time value far exceeds the historical normal range), it is recorded as abnormal in that dimension. Specifically, for the current index value of the high-value anomaly type, such as the current index value of the load current, the voltage loss coefficient per unit power, and the current imbalance of the three-phase current, if it is greater than a first preset multiple of the corresponding average index value, for example, 1.1 times, then the current index value is determined to be abnormal relative to its own historical data. For the current index value of the low-value anomaly type, such as the current index value of the average power factor, if it is less than a second preset multiple of the corresponding average index value, which is the reciprocal of the first preset multiple, for example, 1 / 1.1, then it is determined to be abnormal relative to its own historical data. In this way, this embodiment uses the node's own historical average level as a reference for anomaly determination, which can accurately capture changes in the node's index that deviate from its normal operating pattern, avoid misjudgments caused by differences in the basic conditions of different nodes, and make anomaly identification more in line with the actual operating characteristics of the node. By combining the differentiating judgment thresholds of index types, it can adapt to the abnormal characteristics of different quantitative indicators, accurately identify different types of deviations such as high values exceeding the standard and low values being too low, avoid judgment bias caused by a single threshold, and make the anomaly identification of its own historical dimension more targeted and accurate.
[0067] Step S13: Divide the monitored object node into several groups of comparable object nodes, and for each of the several groups of comparable object nodes, sort the current index value of the same preset quantitative index, so as to determine whether the current index value is abnormal relative to the target area based on the obtained first sorting result.
[0068] In this embodiment, the monitored target nodes of the target area are divided into several groups of comparable target nodes. For each group of comparable target nodes, the current index value of the same preset quantitative indicator is sorted at the same time section. Based on the obtained first sorting result, it is determined whether the current index value is abnormal relative to the entire target area, so as to determine whether the index of the node is in a disadvantageous position in the overall area during the same period.
[0069] It should be noted that the process of dividing the monitored nodes into several groups of comparable nodes is as follows: User nodes located under the same power supply branch are divided into one group of comparable nodes; line nodes located on the same branch are divided into another group of comparable nodes; and line nodes located on the same main road are divided into yet another group of comparable nodes, thus obtaining several groups of comparable nodes. That is, the monitored nodes of the target transformer area are grouped according to their power supply topology: user nodes under the same power supply branch form one group, line nodes on the same branch form one group, and line nodes on the same main road form one group, ultimately forming several groups of comparable nodes. It is understood that if no comparable nodes exist for the transformer area, that is, the three-phase current imbalance does not require anomaly judgment at the target transformer area level.
[0070] It should be further pointed out that the processing procedure for determining whether the current indicator value is abnormal relative to the target distribution area is as follows: If the indicator type of the preset quantitative indicator is the high-value abnormality type, the current indicator values of the same preset quantitative indicator are sorted in descending order to obtain the corresponding first sorting result. Based on the first sorting result, the target quantile is determined, and the target indicator values among the current indicator values that are higher than the target quantile are identified, thus determining that the target indicator value is abnormal relative to the target distribution area. If the indicator type of the preset quantitative indicator is the low-value abnormality type, the current indicator values of the same preset quantitative indicator are sorted in ascending order to obtain the corresponding first sorting result. Based on the first sorting result, the target quantile is determined, and the target indicator values among the current indicator values that are lower than the target quantile are identified, thus determining that the target indicator value is abnormal relative to the target distribution area. That is, a specific quantile is taken as the horizontal distribution area threshold for sorting the current indicator values. If the severity of an indicator value of a certain node ranks among the top in the entire distribution area (e.g., entering the top 10%), it is recorded as abnormal in that dimension. Specifically, if the preset quantitative indicator is a high-value anomaly type, the current indicator values of the same indicator are sorted in descending order to obtain a first sorting result. Based on this result, a target quantile is determined, for example, 10%. Current indicator values higher than the target quantile are determined to be anomalies relative to the target area. If it is a low-value anomaly type, the current indicator values of the same indicator are sorted in ascending order to obtain a first sorting result. After determining the target quantile, for example, 10%, current indicator values lower than the target quantile are determined to be anomalies relative to the target area. In this way, by grouping and classifying comparable nodes and performing horizontal sorting and comparison, this embodiment can effectively eliminate the interference of basic differences between different types of nodes, focus on the distribution patterns of indicators of the same type of nodes, and accurately identify abnormal indicators that significantly deviate from the normal level of the same type of nodes within the transformer area. This provides anomaly evidence in the horizontal dimension of transformer area for the analysis of the causes of low voltage, complementing the anomaly judgment in its own historical dimension, and further improving the comprehensiveness and reliability of anomaly identification. The node grouping method based on the power supply topology structure can ensure that the nodes in each group have similar power supply environment and operating characteristics, eliminating the interference of basic differences between nodes in different topological locations to the greatest extent, making the horizontal comparison of indicators within the same group more reasonable and valuable for reference. The differentiated sorting of indicator anomaly types and the quantile screening can accurately capture extreme indicators that significantly deviate from the overall distribution among the same type of nodes within the transformer area, avoiding the limitations of absolute value comparison.
[0071] Step S14: If the current indicator value is abnormal relative to its own history and the target transformer area, the current indicator value is determined to be an abnormal indicator value. Using the Pareto principle, the preset quantitative indicators are sorted based on the abnormal indicator values, so as to determine the target low voltage cause of the target transformer area from the various low voltage causes according to the obtained second sorting result.
[0072] In this embodiment, when the current indicator value is abnormal relative to both its own historical state and the overall level of the target transformer area, the current indicator value is determined to be an abnormal indicator value. Then, the Pareto principle is applied to sort the preset quantitative indicators based on the abnormal indicator values, and the core low voltage cause of the target transformer area is locked from the various low voltage causes according to the obtained second sorting result.
[0073] It should be noted that the process of sorting the preset quantitative indicators to determine the target low voltage cause of the target transformer area from the various low voltage causes based on the obtained second sorting result is as follows: Based on the abnormal indicator values, the number of abnormal occurrences corresponding to the preset quantitative indicators is counted, and the preset quantitative indicators are sorted according to the number of abnormal occurrences to obtain the corresponding second sorting result; the abnormal contribution ratio corresponding to each preset quantitative indicator is calculated using the number of abnormal occurrences, and the abnormal contribution ratio is sequentially accumulated according to the second sorting result to obtain the corresponding cumulative contribution ratio, until the cumulative contribution ratio is higher than a preset ratio threshold; the indicator combination of the preset quantitative indicators corresponding to the cumulative contribution ratio is determined, and the low voltage cause corresponding to the indicator combination is determined as the target low voltage cause of the target transformer area. That is, based on the abnormal indicator values, the number of abnormal occurrences of each of the preset quantitative indicators for each user or line is counted. The preset quantitative indicators are sorted according to the number of abnormal occurrences to obtain a second sorting result. Then, the abnormal contribution ratio of each preset quantitative indicator is calculated according to the number of abnormal occurrences. The abnormal contribution ratios are accumulated sequentially according to the second sorting result to obtain a cumulative contribution ratio until the cumulative contribution ratio is higher than a preset ratio threshold, such as 80%. The low voltage cause associated with the preset quantitative indicator combination corresponding to the cumulative contribution ratio at this time is determined as the target low voltage cause of the target transformer area. In addition, this embodiment can also sort according to the severity of the abnormality. In this way, this embodiment uses dual anomaly verification to screen core indicators. Through logical judgment rules, it eliminates the randomness of single-dimensional anomalies and effectively filters out those cases where the anomaly itself has deteriorated historically but the overall situation in the transformer area is still acceptable, or those cases where the anomaly is relatively poor in the transformer area but has remained so without worsening. This makes the anomaly judgment more accurate. By sorting the frequency of anomalies and accumulating the contribution ratio, the core indicator combination is precisely implemented. This can identify the key cause combination that contributes the most to the low voltage problem, avoid the blindness of cause analysis, improve the efficiency and accuracy of low voltage cause diagnosis, avoid the waste of resources in decentralized governance, and provide a precise decision-making basis for subsequent centralized resource development of efficient governance plans and rapid improvement of transformer area power supply quality.
[0074] As can be seen from the above, when the target transformer area is detected to meet the preset low voltage conditions, the historical normal data when the target transformer area does not meet the preset low voltage conditions are used to calculate the historical index value of each preset quantitative index corresponding to various causes of low voltage. For each monitored node of the target transformer area, the corresponding average index value is obtained through the historical index value. By comparing the current index value of the monitored node with the average index value, it is determined whether the current index value is abnormal relative to its own history. At the same time, the monitored node is divided into several groups of comparable nodes, and the current index value of the same preset quantitative index in each group is sorted. Based on the first sorting result, it is determined whether the current index value is abnormal relative to the target transformer area. If the current index value is abnormal in both its own historical comparison and the comparison within the transformer area, it is determined to be an abnormal index value. Then, the Pareto principle is used to sort the preset quantitative indexes according to the abnormal index values. Based on the second sorting result, the target low voltage cause of the target transformer area is determined from the various causes of low voltage. In this way, through the above-described process of this application embodiment, combined with the dual anomaly judgment logic of its own historical data and horizontal comparison within the transformer area, and with the core cause screening mechanism of the Pareto principle, it can not only accurately locate key indicators that deviate from its own normal level and are different from the overall state of the transformer area, but also focus on the core causes of low voltage through priority ranking, effectively avoiding the one-sidedness of single-dimensional judgment, and improving the accuracy and efficiency of cause location; it mainly relies on existing power consumption information collection systems, distribution automation systems and other data sources, without the need to add a large number of dedicated collection terminals. It can achieve in-depth analysis using existing data, reducing implementation costs, and thus optimizing the transformer area low voltage cause diagnosis method to solve the problems of low efficiency, high cost and poor targeting in cause diagnosis.
[0075] Accordingly, see Figure 3 As shown in the figure, this application embodiment also provides a low voltage cause diagnosis device for a transformer substation, including:
[0076] The indicator value calculation module 11 is used to calculate the historical indicator value of each preset quantitative indicator by using the historical normal data of the target area if the current target area is detected to meet the preset low voltage conditions; the historical normal data is the historical raw data when the preset low voltage conditions are not met; the preset quantitative indicators are the quantitative indicators corresponding to various causes of low voltage.
[0077] The anomaly detection module 12 is used to calculate the corresponding average index value for each monitored object node in the target area using the historical index value, and to determine whether the current index value is abnormal relative to its own history based on the current index value of the monitored object node and the corresponding average index value.
[0078] The indicator value sorting module 13 is used to divide the monitored object node into several groups of comparable object nodes, and sort the current indicator value of the same preset quantitative indicator for each of the several groups of comparable object nodes, so as to determine whether the current indicator value is abnormal relative to the target area based on the obtained first sorting result.
[0079] The indicator sorting module 14 is used to determine that the current indicator value is an abnormal indicator value if the current indicator value is abnormal relative to its own history and the target transformer area. Using the Pareto principle, the preset quantitative indicators are sorted based on the abnormal indicator values, so as to determine the target low voltage cause of the target transformer area from the various low voltage causes according to the obtained second sorting result.
[0080] In some specific implementations, the monitored target nodes include user nodes and line nodes within the target distribution area, as well as distribution area nodes characterizing the target distribution area; the preset quantitative indicators include load current, average power factor, voltage loss coefficient per unit power, and three-phase current imbalance. The load current is a quantitative indicator corresponding to low voltage causes related to user load, the average power factor is a quantitative indicator corresponding to low voltage causes related to power factor, the voltage loss coefficient per unit power is a quantitative indicator corresponding to low voltage causes related to network architecture rationality, and the three-phase current imbalance is a quantitative indicator corresponding to low voltage causes related to three-phase imbalance. The load current, the voltage loss coefficient per unit power, and the three-phase current imbalance are all high-value anomaly types, where the value is positively correlated with the probability of an anomaly. The average power factor is a low-value anomaly type, where the value is negatively correlated with the probability of an anomaly.
[0081] In some specific embodiments, the index value calculation module 11 may specifically include:
[0082] The data determination unit is used to determine that the target substation meets the preset low voltage condition if the user voltage of the current target substation is detected to be lower than the preset voltage threshold, and to determine the historical normal data of the target substation within a preset historical time period; the historical normal data is the historical raw data of the corresponding user voltage not lower than the preset voltage threshold.
[0083] The indicator value determination unit is used to determine the current data and average power factor data in the historical normal data at the node level of the user node, and to determine the current data as the historical indicator value of the load current and the average power factor data as the historical indicator value of the average power factor.
[0084] The first formula calculation unit is used to determine the line-related data in the historical normal data at the node level of the line node, and calculate the preset unit power voltage loss calculation formula based on the line-related data and the average power factor data, so as to determine the obtained calculation result as the historical index value of the unit power voltage loss coefficient.
[0085] The second formula calculation unit is used to determine the three-phase current related data in the historical normal data at the node level of the transformer area node, and calculate the preset three-phase current imbalance calculation formula based on the three-phase current related data, so as to determine the obtained calculation result as the historical index value of the three-phase current imbalance.
[0086] In some specific embodiments, the anomaly detection module 12 may specifically include:
[0087] The indicator value calculation unit is used to collect the real-time raw data of the target area and calculate the current indicator value of each preset quantitative indicator based on the real-time raw data for each monitored object node.
[0088] The first anomaly determination unit is used to determine that the current indicator value is abnormal relative to its own history if the indicator type of the current indicator value is the high value anomaly type and the current indicator value is greater than the first preset multiple of the corresponding average indicator value.
[0089] The second anomaly determination unit is used to determine that the current indicator value is abnormal relative to its own history if the indicator type of the current indicator value is the low value anomaly type and the current indicator value is less than the second preset multiple of the corresponding average indicator value; the second preset multiple is the reciprocal of the first preset multiple.
[0090] In some specific embodiments, the index value sorting module 13 may specifically include:
[0091] The node division unit is used to divide the user nodes located under the same power supply branch into the same group of comparable object nodes, the line nodes located on the same branch into the same group of comparable object nodes, and the line nodes located on the same main road into the same group of comparable object nodes, so as to obtain several groups of comparable object nodes.
[0092] In some specific embodiments, the index value sorting module 13 may specifically include:
[0093] The third anomaly determination unit is used to sort the current index values of the same preset quantitative index in descending order if the index type of the preset quantitative index is the high value anomaly type, to obtain the corresponding first sorting result, and to determine the target quantile based on the first sorting result, and to determine the target index value among the current index values that is higher than the target quantile, and to determine that the target index value is anomaly relative to the target station area.
[0094] The fourth anomaly determination unit is used to sort the current index values of the same preset quantitative index in ascending order if the index type of the preset quantitative index is the low value anomaly type, to obtain the corresponding first sorting result, and to determine the target quantile based on the first sorting result, and to determine the target index value among the current index values that is lower than the target quantile, and to determine that the target index value is anomaly relative to the target station area.
[0095] In some specific embodiments, the index ranking module 14 may specifically include:
[0096] The indicator sorting unit is used to count the number of times the preset quantitative indicator appears based on the abnormal indicator value, and sort the preset quantitative indicator according to the number of times the abnormal indicator appears to obtain a corresponding second sorting result.
[0097] The percentage accumulation unit is used to calculate the percentage of abnormal contribution corresponding to each preset quantitative indicator using the number of abnormal occurrences, and to accumulate the percentage of abnormal contribution in sequence according to the second sorting result to obtain the corresponding cumulative contribution percentage until the cumulative contribution percentage is higher than the preset percentage threshold.
[0098] The cause determination unit is used to determine the combination of preset quantitative indicators corresponding to the cumulative contribution ratio, and to determine the low voltage cause corresponding to the indicator combination as the target low voltage cause of the target transformer area.
[0099] Furthermore, embodiments of this application also disclose an electronic device, Figure 4 This is a structural diagram of an electronic device 20 according to an exemplary embodiment. The content of the diagram should not be construed as limiting the scope of this application. The electronic device 20 may specifically include: at least one processor 21, at least one memory 22, a power supply 23, a communication interface 24, an input / output interface 25, and a communication bus 26. The memory 22 stores a computer program, which is loaded and executed by the processor 21 to implement the relevant steps in the low-voltage cause diagnosis method for transformer substations disclosed in any of the foregoing embodiments. Furthermore, the electronic device 20 in this embodiment may specifically be a computer.
[0100] In this embodiment, the power supply 23 is used to provide operating voltage for each hardware device on the electronic device 20; the communication interface 24 can create a data transmission channel between the electronic device 20 and external devices, and the communication protocol it follows can be any communication protocol applicable to the technical solution of this application, and is not specifically limited here; the input / output interface 25 is used to acquire external input data or output data to the outside world, and its specific interface type can be selected according to specific application needs, and is not specifically limited here.
[0101] In addition, the memory 22, as a carrier for resource storage, can be a read-only memory, random access memory, disk or optical disk, etc. The resources stored thereon can include operating system 221, computer program 222, etc., and the storage method can be temporary storage or permanent storage.
[0102] The operating system 221 is used to manage and control the various hardware devices on the electronic device 20 and the computer program 222, which may be Windows Server, Netware, Unix, Linux, etc. In addition to including a computer program capable of performing the low-voltage cause diagnosis method for the transformer substation disclosed in any of the foregoing embodiments, the computer program 222 may further include a computer program capable of performing other specific tasks.
[0103] Furthermore, this application also discloses a computer-readable storage medium for storing a computer program; wherein, when the computer program is executed by a processor, it implements the aforementioned method for diagnosing the causes of low voltage in transformer substations. Specific steps of this method can be found in the corresponding content disclosed in the foregoing embodiments, and will not be repeated here.
[0104] The various embodiments in this specification are described in a progressive manner, with each embodiment focusing on its differences from other embodiments. Similar or identical parts between embodiments can be referred to interchangeably. For the apparatus disclosed in the embodiments, since it corresponds to the method disclosed in the embodiments, the description is relatively simple; relevant parts can be referred to in the method section.
[0105] Those skilled in the art will further recognize that the units and algorithm steps of the various examples described in conjunction with the embodiments disclosed herein can be implemented in electronic hardware, computer software, or a combination of both. To clearly illustrate the interchangeability of hardware and software, the components and steps of the various examples have been generally described in terms of functionality in the foregoing description. Whether these functions are implemented in hardware or software depends on the specific application and design constraints of the technical solution. Those skilled in the art can use different methods to implement the described functions for each specific application, but such implementation should not be considered beyond the scope of this application.
[0106] The steps of the methods or algorithms described in conjunction with the embodiments disclosed herein can be implemented directly by hardware, a software module executed by a processor, or a combination of both. The software module can be located in random access memory (RAM), main memory, read-only memory (ROM), electrically programmable ROM, electrically erasable programmable ROM, registers, hard disk, removable disk, CD-ROM, or any other form of storage medium known in the art.
[0107] Finally, it should be noted that in this document, relational terms such as "first" and "second" are used only to distinguish one entity or operation from another, and do not necessarily require or imply any such actual relationship or order between these entities or operations. Furthermore, the terms "comprising," "including," or any other variations thereof are intended to cover non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements includes not only those elements but also other elements not expressly listed, or elements inherent to such a process, method, article, or apparatus. Without further limitations, an element defined by the phrase "comprising one..." does not exclude the presence of other identical elements in the process, method, article, or apparatus that includes said element.
[0108] The technical solutions provided in this application have been described in detail above. Specific examples have been used to illustrate the principles and implementation methods of this application. The descriptions of the above embodiments are only for the purpose of helping to understand the methods and core ideas of this application. At the same time, for those skilled in the art, there will be changes in the specific implementation methods and application scope based on the ideas of this application. Therefore, the content of this specification should not be construed as a limitation of this application.
Claims
1. A low-voltage cause diagnosis method for a transformer area, characterized by, The method comprises the following steps: If it is detected that the current target substation area satisfies a preset low-voltage condition, historical index values of each preset quantitative index are calculated respectively by using historical normal data of the target substation area; The historical normal data are historical original data when the preset low-voltage condition is not satisfied; and the preset quantitative index is a quantitative index corresponding to each type of low-voltage cause; For each monitored object node of the target substation area, a corresponding average index value is calculated by using the historical index values, and it is judged whether the current index value is abnormal relative to the historical index value of the monitored object node based on the current index value and the corresponding average index value; The monitored object nodes are divided into several groups of comparable object nodes, and for each group of the several groups of comparable object nodes, the current index values of the same preset quantitative index are sorted respectively to determine whether the current index values are abnormal relative to the target substation area according to a first sorting result obtained; If the current index values are abnormal relative to the historical index values and the target substation area, the current index values are determined as abnormal index values, the preset quantitative index is sorted based on the abnormal index values by using the Pareto principle, and a target low-voltage cause of the target substation area is determined from the types of low-voltage causes according to a second sorting result obtained.
2. The diagnostic method of low voltage causes in a transformer district as claimed in claim 1, wherein, The monitored object nodes include user nodes and line nodes in the target substation area and a substation node representing the target substation area; the preset quantitative index includes a load current, an average power factor, a unit power voltage loss coefficient, and a three-phase current unbalance degree, the load current is a quantitative index corresponding to a low-voltage cause related to user load, the average power factor is a quantitative index corresponding to a low-voltage cause related to a power factor, the unit power voltage loss coefficient is a quantitative index corresponding to a low-voltage cause related to network architecture rationality, and the three-phase current unbalance degree is a quantitative index corresponding to a low-voltage cause related to three-phase unbalance degree; the index types of the load current, the unit power voltage loss coefficient, and the three-phase current unbalance degree are all high-value abnormal types, the high-value abnormal type is a type in which the numerical value is positively correlated with the probability of abnormality; and the index type of the average power factor is a low-value abnormal type, the low-value abnormal type is a type in which the numerical value is negatively correlated with the probability of abnormality.
3. The low-voltage cause diagnosis method for a district according to claim 2, characterized by, If it is detected that the current target substation area satisfies a preset low-voltage condition, historical index values of each preset quantitative index are calculated respectively by using historical normal data of the target substation area, which comprises the following steps: If it is detected that the user voltage of the current target substation area is lower than a preset voltage threshold, it is determined that the target substation area satisfies a preset low-voltage condition, and historical normal data of the target substation area in a preset historical time period are determined; the historical normal data are historical original data corresponding to a user voltage not lower than the preset voltage threshold; At the node level of the user node, current data and average power factor data in the historical normal data are determined, the current data is determined as a historical index value of the load current, and the average power factor data is determined as a historical index value of the average power factor; At the node level of the line node, line-related data in the historical normal data are determined, and a preset unit power voltage loss calculation formula is calculated based on the line-related data and the average power factor data, so that a calculation result obtained is determined as a historical index value of the unit power voltage loss coefficient; At the node level of the transformer area node, three-phase current-related data in the historical normal data are determined, and a preset three-phase current unbalance degree calculation formula is calculated based on the three-phase current-related data, so that a calculation result obtained is determined as a historical index value of the three-phase current unbalance degree.
4. The diagnostic method of low voltage causes in a transformer district as claimed in claim 2, wherein, The judgment of whether the current index value is abnormal relative to its history based on the current index value of the monitored object node and the corresponding average index value comprises: Real-time raw data of the target transformer area are collected, and for each monitored object node, the current index value of each preset quantitative index is calculated based on the real-time raw data; If the index type of the current index value is the high-value abnormal type, when the current index value is greater than the first preset multiple of the corresponding average index value, it is determined that the current index value is abnormal relative to its history; If the index type of the current index value is the low-value abnormal type, when the current index value is less than the second preset multiple of the corresponding average index value, it is determined that the current index value is abnormal relative to its history; the second preset multiple is the reciprocal of the first preset multiple.
5. The low-voltage cause diagnosis method for a district according to Claim 2, characterized by, The division of the monitored object nodes into several comparable object nodes comprises: The user nodes in several monitored object nodes located in the same power supply branch are divided into the same group of comparable object nodes, the line nodes located in the same branch are divided into the same group of comparable object nodes, and the line nodes located in the same main road are divided into the same group of comparable object nodes, to obtain several groups of comparable object nodes.
6. The low-voltage cause diagnosis method for a district according to Claim 2, characterized by, The sorting of the current index values of the same preset quantitative index respectively, to determine whether the current index value is abnormal relative to the target transformer area according to the obtained first sorting result, comprises: If the index type of the preset quantitative index is the high-value abnormal type, the current index values of the same preset quantitative index are sorted in descending order to obtain the corresponding first sorting result, a target fractile is determined according to the first sorting result, and a target index value higher than the target fractile in the current index value is determined, and it is determined that the target index value is abnormal relative to the target transformer area; If the index type of the preset quantitative index is the low-value abnormal type, the current index values of the same preset quantitative index are sorted in ascending order to obtain a corresponding first sorting result, a target quantile is determined according to the first sorting result, and a target index value lower than the target quantile is determined from the current index values, and it is determined that the target index value is abnormal relative to the target area.
7. The low-voltage cause diagnosis method for a transformer area according to any one of claims 1 to 6, characterized by, The preset quantitative index is sorted based on the abnormal index value by using the Pareto principle, so as to determine a target low-voltage cause of the target area from the low-voltage causes according to a second sorting result obtained. The abnormal index value is counted to obtain the abnormal occurrence times of the preset quantitative index, and the preset quantitative index is sorted according to the abnormal occurrence times to obtain a corresponding second sorting result. The abnormal contribution proportion corresponding to each preset quantitative index is calculated by using the abnormal occurrence times, and the abnormal contribution proportions are sequentially accumulated according to the second sorting result to obtain a corresponding cumulative contribution proportion, until the cumulative contribution proportion is higher than a preset proportion threshold. The index combination of the preset quantitative index corresponding to the cumulative contribution proportion is determined, and the low-voltage cause corresponding to the index combination is determined as the target low-voltage cause of the target area.
8. A low-voltage cause diagnosis device for a district, characterized by comprising: It includes: The index value calculation module is configured to calculate the historical index value of each preset quantitative index by using the historical normal data of the target area if it is detected that the current target area meets the preset low-voltage condition; the historical normal data is the historical original data when the preset low-voltage condition is not met; and the preset quantitative index is a quantitative index corresponding to each low-voltage cause; The abnormality judgment module is configured to calculate the average index value corresponding to each monitored object node of the target area by using the historical index value, and determine whether the current index value of the monitored object node is abnormal relative to the historical data based on the current index value and the average index value corresponding thereto. The index value sorting module is configured to divide the monitored object nodes into a plurality of comparable object nodes, and sort the current index values of the same preset quantitative index for each group of comparable object nodes to determine whether the current index values are abnormal relative to the target area according to a first sorting result obtained. The index sorting module is configured to determine that the current index value is an abnormal index value if the current index value is abnormal relative to the historical data and the target area, sort the preset quantitative index based on the abnormal index value by using the Pareto principle, and determine a target low-voltage cause of the target area from the low-voltage causes according to a second sorting result obtained.
9. An electronic device, comprising: It includes: The memory is configured to save a computer program; The processor is configured to execute the computer program to implement the low-voltage cause diagnosis method of the area as claimed in any one of claims 1 to 7.
10. A computer-readable storage medium, characterized in that, A computer program product for saving a computer program; wherein the computer program, when executed by a processor, implements the low-voltage cause diagnosis method for a transformer area as claimed in any one of claims 1 to 7.