A comprehensive evaluation method for optimizing investment in substation reconstruction
By using a comprehensive evaluation method for transformer substation renovation investment, combined with historical anomalies and future load forecasts, the equipment renovation strategy was optimized. This solved the problem of the importance and load trends not being considered in transformer substation renovation, and improved the accuracy of equipment upgrades and the efficiency of operation and maintenance.
Patent Information
- Authority / Receiving Office
- CN · China
- Patent Type
- Patents(China)
- Current Assignee / Owner
- STATE GRID SHANGHAI MUNICIPAL ELECTRIC POWER CO
- Filing Date
- 2022-11-16
- Publication Date
- 2026-05-29
AI Technical Summary
The existing comprehensive evaluation method for transformer substation renovation investment lacks consideration of the importance of different substations and the prediction of future load trends, resulting in poor comprehensive evaluation results.
By determining the comprehensive evaluation indicators for transformer substation renovation investment, calculating the scores of each indicator, calculating the priority of equipment renovation based on the indicator weights, setting multi-level early warning thresholds and levels, optimizing equipment renovation strategies, and using a load time series forecasting model based on attention mechanisms for load forecasting.
Improve the accuracy of equipment upgrades and renovations, reduce overload and failure rates in transformer substations, optimize operation and maintenance methods and efficiency, and enhance power supply service management capabilities and customer satisfaction.
Smart Images
Figure CN115841389B_ABST
Abstract
Description
Technical Field
[0001] This invention relates to the field of transformer substation renovation technology, and in particular to a comprehensive evaluation method for optimizing investment in transformer substation renovation. Background Technology
[0002] Existing comprehensive evaluation methods for distribution transformer area renovation investment mainly rely on a comprehensive evaluation based on power supply load and anomaly occurrence, such as capacity, load factor, power supply, and three-phase imbalance. Specific examples include existing comprehensive evaluation methods for low-voltage distribution transformer areas in regional smart distribution networks, a low-voltage distribution transformer area evaluation method based on improved principal component analysis, an investment decision-making method for 10KV and below distribution network projects based on basic ledgers, and a comprehensive evaluation method and system for distribution network distribution transformer areas.
[0003] However, most current methods are based on a comprehensive evaluation of historical load conditions and anomaly occurrences. On the one hand, they lack consideration of the importance of different transformer substations, and on the other hand, they lack prediction and evaluation of future load trends. As a result, the comprehensive evaluation is not effective for transformer substation renovation. Summary of the Invention
[0004] This invention aims to at least partially solve one of the technical problems in related technologies. Therefore, one objective of this invention is to provide a comprehensive evaluation method for optimizing investment in transformer substation upgrades, thereby improving the accuracy of equipment upgrades, optimizing equipment investment, reducing substation overload and failure rates, optimizing operation and maintenance methods and efficiency, and continuously improving power supply service management capabilities and response speed, ultimately enhancing customer service satisfaction.
[0005] To achieve the above objectives, the present invention is implemented through the following technical solution:
[0006] A comprehensive evaluation method for optimizing investment in transformer substation renovation includes:
[0007] Several comprehensive evaluation indicators for investment in the renovation of transformer substations were determined;
[0008] Calculate the index scores corresponding to the comprehensive evaluation indicators of the renovation investment of each transformer substation, and calculate the comprehensive score of the priority of equipment renovation of the transformer substation based on the scores of each index and the corresponding index weights.
[0009] Determine the early warning thresholds for equipment upgrades in multiple transformer substations, and determine the early warning level for equipment upgrades in the transformer substations based on the comprehensive score of the equipment upgrade priority and the early warning thresholds for equipment upgrades in the multiple substations.
[0010] The priority of equipment upgrades in a given area is determined based on the early warning level of that area.
[0011] Optionally, several of the comprehensive evaluation indicators for the transformation investment of the transformer substation may include historical abnormality indicators, potential abnormality indicators, and indicators of the degree of abnormal impact.
[0012] Optionally, the steps for calculating historical operational anomaly index scores include:
[0013] Determine the historical heavy overload frequency index, the historical other abnormal frequency index, and the number of days since the last year index for the transformer substation; calculate the historical operational abnormality index score based on the historical heavy overload frequency index, the historical other abnormal frequency index, and the number of days since the last year index for the transformer substation.
[0014] Optionally, the historical operational anomaly index score can be calculated using the following formula:
[0015]
[0016] Among them, S h h is the score for historical abnormal operation indicators. c For the historical heavy overload frequency index of the power station area, o c d represents the number of other historical anomalies in the Taiwan area, and d represents the number of days since today (within the past year).
[0017] Optionally, the historical overload frequency index of the transformer area is used to count the historical overload frequency and the historical heavy load frequency of the transformer area; the historical other abnormal frequency index of the transformer area is used to count the historical three-phase severe imbalance frequency, low voltage frequency, and low power factor frequency of the transformer area; and the number of days from the last year to today index is used to count the number of days from a preset date in the previous year to today.
[0018] Optionally, the steps for calculating the potential anomaly index score include:
[0019] A time series prediction model for transformer area load based on an attention mechanism is constructed, and the transformer area load is predicted according to the time series prediction model to obtain the transformer area load prediction result.
[0020] Based on the load forecast results of the transformer area, the following indicators are statistically analyzed: maximum predicted maximum load rate, mean maximum predicted load rate, upper quartile of maximum predicted load rate, sliding predicted maximum load rate, mean sliding predicted load rate, and upper quartile of sliding predicted load rate.
[0021] The final predicted load of the transformer area is calculated based on the statistical indicators of the transformer area load forecast results.
[0022] Determine multi-level potential anomaly warning thresholds, and calculate potential anomaly index scores based on the final predicted load of the transformer area and the multi-level potential anomaly warning thresholds.
[0023] Optionally, the potential anomaly index score can be calculated using the following formula:
[0024]
[0025] Among them, S p For potential outlier indicators, lr p For the final predicted load of the distribution area, t l1 The threshold for Level 1 potential anomaly early warning, t l2 The threshold for secondary potential anomalies is t. l3 This is the threshold for a Level 3 potential anomaly warning.
[0026] Optionally, the steps for calculating the score of the abnormal impact index include:
[0027] The importance index and the total assembly capacity index of the power distribution area are determined as indicators for evaluating the degree of impact of power distribution area anomalies; the scores of the importance index and the total assembly capacity index of the power distribution area are calculated, and the score of the degree of impact of anomalies is calculated based on the scores of the importance index and the total assembly capacity index of the power distribution area.
[0028] Optionally, the scores for the importance index of the transformer area and the total installation capacity index of the transformer area can be calculated using the following formulas:
[0029]
[0030]
[0031] Where imp represents the importance of the station area, s imp The score represents the importance index of the transformer area, cap represents the total assembly capacity of the transformer area, max(cap) represents the maximum value of the total assembly capacity of all transformer areas, and s cap This indicates the score for the total assembly capacity index of the transformer area.
[0032] Optionally, the score of the abnormal impact index can be calculated using the following formula:
[0033]
[0034] Among them, w imp w represents the weight of the importance of each channel. cap S represents the weight of the total assembly capacity of the distribution area. f The score represents the degree of impact of the abnormality.
[0035] This invention has at least the following technical effects:
[0036] 1. This invention can optimize the upgrading and transformation strategy of distribution network equipment. Specifically, it can continuously improve the effect of equipment upgrading and transformation, and enhance the stability of equipment operation and power supply capacity through the construction of a data-intelligent distribution network equipment management system.
[0037] 2. This invention can improve load forecasting capabilities and regional operation and maintenance efficiency. Specifically, it can accurately predict the distribution of regional load hotspots through in-depth analysis of the load levels of equipment in the distribution network area, provide intelligent decision support, optimize inspection modes, improve operation and maintenance efficiency, and realize proactive services based on data-driven approaches.
[0038] 3. This invention can enhance the lean management of equipment operation status by business personnel. Specifically, it can provide business personnel with detailed equipment operation analysis reports and anomaly diagnosis reports through comprehensive and accurate quantitative analysis of the power supply operation status of equipment in the distribution network area, thereby improving the level of lean management.
[0039] Additional aspects and advantages of the invention will be set forth in part in the description which follows, and in part will be obvious from the description, or may be learned by practice of the invention. Attached Figure Description
[0040] Figure 1 This is a flowchart of a comprehensive evaluation method for optimizing investment in transformer substation renovation, provided as an embodiment of the present invention. Detailed Implementation
[0041] The following describes this embodiment in detail. Examples of the embodiments are shown in the accompanying drawings, wherein the same or similar reference numerals denote the same or similar elements or elements having the same or similar functions throughout. The embodiments described below with reference to the accompanying drawings are exemplary and intended to explain the invention, and should not be construed as limiting the invention.
[0042] To improve the effectiveness of distribution transformer area renovation, this invention proposes a comprehensive evaluation method for distribution transformer area renovation. This method evaluates the transformer area from three dimensions: historical operational anomalies, predicted potential anomalies, and the degree of impact after anomalies occur. The comprehensive evaluation score is used to rank the renovation priorities, thereby improving the accuracy of equipment upgrades, optimizing equipment investment, reducing transformer area overload and failure rates, optimizing operation and maintenance methods and efficiency, and continuously improving power supply service management capabilities and response speed, thus enhancing customer service satisfaction.
[0043] The following describes a comprehensive evaluation method for optimizing investment in transformer substation renovation, with reference to the accompanying drawings.
[0044] Figure 1 This is a flowchart illustrating a comprehensive evaluation method for optimizing investment in transformer substation renovation, provided as an embodiment of the present invention. Figure 1 As shown, the method includes:
[0045] Step S1: Determine several comprehensive evaluation indicators for the investment in the renovation of transformer substations.
[0046] Among them, several comprehensive evaluation indicators for the renovation investment of transformer substations include historical abnormality indicators, potential abnormality indicators, and indicators of the degree of abnormal impact.
[0047] Specifically, the comprehensive evaluation of a transformer substation can be based on three dimensions: historical operational anomalies, potential anomalies, and the degree of impact of anomalies. The comprehensive score can be used to help determine the priority of equipment upgrades in the substation.
[0048] In this embodiment, historical operational anomaly indicators are used to represent abnormal situations that occurred in the equipment over the past year, such as the number of heavy overloads, excessive line losses, and severe three-phase imbalance. Potential anomaly indicators are used to represent possible anomalies that may occur in the equipment in the next year based on load forecast results, such as maximum load rate and load growth rate. Anomaly impact indicators are used to evaluate the impact of anomalies occurring in the distribution network equipment, such as the capacity and importance of the distribution area.
[0049] Step S2: Calculate the index scores corresponding to the comprehensive evaluation indicators of the renovation investment of each transformer substation, and calculate the comprehensive score of the priority of equipment renovation of the transformer substation based on the scores of each index and the corresponding index weights.
[0050] The steps for calculating the scores of historical operational anomaly indicators include:
[0051] Determine the historical heavy overload frequency index, the historical other abnormal frequency index, and the number of days since the last year for the transformer substation; calculate the historical operation abnormality index score based on the historical heavy overload frequency index, the historical other abnormal frequency index, and the number of days since the last year for the transformer substation.
[0052] In this embodiment, the historical overload frequency index of the transformer area is used to count the historical overload frequency and the historical heavy load frequency of the transformer area. The historical other abnormal frequency index of the transformer area is used to count the historical three-phase severe imbalance frequency, low voltage frequency, and low power factor frequency of the transformer area. The number of days from the last year to today index is used to count the number of days from a preset date in the previous year to today.
[0053] Specifically, the above indicators can be determined based on the logic of the comprehensive evaluation of existing business operations. For example, taking the priority evaluation of equipment upgrades in 2019 as an example: Starting from the peak summer season of 2018, basic data on heavy overload was collected for a period of time. The first occurrence of heavy overload was classified as a Class 2 warning. Subsequent occurrences of heavy overload in the same area were classified as Class 1 warnings. If no heavy overload occurred but two or more of the following abnormal conditions were present, it was classified as a Class 3 warning (e.g., excessive line loss, new connected users, load rate increase exceeding 15% for two consecutive months, severe three-phase imbalance, severely low power factor). After accumulating data for more than half a year, a comprehensive evaluation list of distribution areas was generated by February 2019, containing several Class 1, 2, and 3 warning areas. Class 1 warnings required upgrades by specialized departments, while Class 2 and 3 warnings required enhanced monitoring. Monitoring was then conducted using the same method. A comprehensive evaluation list was released three times a year: in February, after the first wave of high temperatures, and in October. Through rolling upgrades, the heavy overload areas were gradually resolved.
[0054] Furthermore, following the logic of existing business operations, statistics can be compiled on the number of overloads, heavy loads, three-phase severe imbalances, low voltages, and low power factors occurring since January 1st of the previous year, as well as the number of days since January 1st of the previous year.
[0055] Among them, the number of overloads and the number of heavy loads are considered as serious anomalies and are merged into the historical heavy overload count of the transformer area; other anomalies are considered as ordinary anomalies and are merged into the historical other anomalies count of the transformer area, and the final indicator is the historical other anomalies count of the transformer area.
[0056] In this embodiment, the symbols for historical operational anomalies and their meanings are shown in Table 1 below:
[0057] Table 1. Symbols and Meanings of Historical Operational Anomalies
[0058] symbol meaning <![CDATA[S h ]]> Historical operational anomaly index score <![CDATA[h c ]]> Historical heavy overload times in the Taiwan area <![CDATA[o c ]]> Other historical anomalies in the Taiwan area d Number of days from today in the past year
[0059] After determining each indicator, the historical operational anomaly indicator score can be calculated using the following formula:
[0060]
[0061] Among them, S h h is the score for historical abnormal operation indicators. c For the historical heavy overload frequency index of the power station area, o c d represents the number of other historical anomalies in the Taiwan area, and d represents the number of days since today (within the past year).
[0062] It should be noted that if the number of heavy overloads is greater than or equal to 2, the historical score is greater than 90 points. The higher the number of heavy overloads or other anomalies, the higher the historical score. When the number of heavy overloads minus 2 reaches 1 / 20 of the statistical days (d) or the number of other anomalies reaches 1 / 10 of the statistical days (d), the historical score reaches 100 points. If it is greater than 100 points, the historical score is set to 100 points.
[0063] If the number of overload events is greater than or equal to 1, the historical score is greater than 70 points. The higher the number of other anomalies, the higher the historical score. When the number of other anomalies reaches 1 / 10 of the statistical days (d), the historical score reaches 90 points. If it is greater than 90 points, the historical score is set to 90 points.
[0064] If no overload or excessive load occurs and there are two or more other anomalies, the historical score is greater than 60 points. The higher the number of other anomalies, the higher the historical score. When the number of other anomalies minus 2 reaches 1 / 10 of the statistical days (d), the historical score reaches 70 points. If it is greater than 70 points, the historical score is set to 70 points.
[0065] If no overload or other abnormality occurs and the number of occurrences is less than 2, the historical score is 0.
[0066] In one embodiment of the present invention, the step of calculating the potential anomaly index score includes:
[0067] A time series load forecasting model for transformer substations based on an attention mechanism is constructed, and the load of the substations is predicted according to the model to obtain the load forecasting results. Based on the load forecasting results, the maximum predicted maximum load rate, the mean predicted maximum load rate, the upper quartile of the maximum predicted load rate, the sliding predicted maximum load rate, the mean sliding predicted load rate, and the upper quartile of the sliding predicted load rate are statistically analyzed. The final predicted load of the substations is calculated based on the statistically analyzed indicators. Multi-level potential anomaly warning thresholds are determined, and the potential anomaly indicator scores are calculated based on the final predicted load of the substations and the multi-level potential anomaly warning thresholds.
[0068] The transformer load time series prediction model in this embodiment is abbreviated as "Attention-CNN+LSTM". This model consists of a preprocessing unit, an Attention & CNN unit, an LSTM unit, and an output unit. The preprocessing unit replaces outlier data with nearest-neighbor mean data; since different input features have varying value ranges, data normalization is required. The Attention & CNN unit extracts multiple partially overlapping continuous subsequences from the original data as its input. The LSTM unit constructs the time series model using the output of the Attention & CNN unit as its input. In this embodiment, the output unit is the output of the last hidden layer of the LSTM unit. Both the Attention & CNN unit and the LSTM unit in this embodiment are convolutional neural network units.
[0069] In this embodiment, the load of the transformer area can be predicted according to the constructed transformer area load time series prediction model to obtain the transformer area load prediction result. Then, based on the transformer area load prediction result, the maximum predicted maximum load rate (the maximum load rate predicted based on the daily maximum load), the maximum predicted load rate mean (the load rate mean predicted based on the daily maximum load), the maximum predicted load rate upper quartile (the load rate upper quartile predicted based on the daily maximum load), the sliding predicted maximum load rate, the sliding predicted load rate mean, and the sliding predicted load rate upper quartile are used as key indicators of potential anomalies.
[0070] The symbols and meanings of the potential anomaly indicators are shown in Table 2 below:
[0071] Table 2. Symbols and meanings of potential anomaly indicators
[0072]
[0073]
[0074] Furthermore, the following weighting is applied to each indicator:
[0075] The default weight of the maximum load factor based on the daily maximum load forecast: max_w m =0.2;
[0076] The default weight of the load factor based on the daily maximum load forecast is: mean_w m =0.1;
[0077] The upper quartile factor of the load rate forecast based on the daily maximum load has a default weight: uq_w m =0.2;
[0078] The default weight for the sliding prediction maximum load factor is: max_w r =0.2;
[0079] The default weight of the moving average load factor is: mean_w r =0.1;
[0080] Default weight of upper quartile factor for sliding forecast load rate: uq_w r =0.2;
[0081] Furthermore, based on the weights configured above and the weighted average of each indicator, the final predicted load lr for the distribution area is calculated. p Then, the potential anomaly index score is calculated using the following formula:
[0082]
[0083] Among them, S p For potential outlier indicators, lr p For the final predicted load of the distribution area, t l1 The threshold for Level 1 potential anomaly early warning, t l2 The threshold for secondary potential anomaly warning, t l3 This refers to the three levels of potential anomaly warning thresholds. Specifically, the first-level potential anomaly warning threshold t... l1 =0.75, Level II potential anomaly early warning threshold t l2 =0.6, Level 3 potential anomaly warning threshold t l3 =0.5.
[0084] It should be noted that if the final predicted load of the transformer area is >= 0.75, the prediction score will be greater than 90 points, and the higher the final load, the higher the prediction score. If the final predicted load of the transformer area is 1, the prediction score reaches 100 points; if it is greater than 100 points, the prediction score is set to 100 points. If the final predicted load of the transformer area is >= 0.6 and less than 0.75, the prediction score will be greater than 70 points, and the higher the final load, the higher the prediction score. When the final predicted load of the transformer area is 0.75, the prediction score reaches 90 points. If the final predicted load of the transformer area is >= 0.5 and less than 0.6, the prediction score will be greater than 60 points, and the higher the final load, the higher the prediction score. When the final predicted load of the transformer area is 0.6, the prediction score reaches 70 points. If the final predicted load of the transformer area is less than 0.5, the prediction score is equal to 0.
[0085] In one embodiment of the present invention, the step of calculating the score of the abnormal impact index includes:
[0086] The importance index and the total assembly capacity index of the power distribution area are determined as indicators for evaluating the degree of impact of anomalies in the power distribution area; the scores of the importance index and the total assembly capacity index of the power distribution area are calculated, and the score of the degree of impact index of anomalies is calculated based on the scores of the importance index and the total assembly capacity index of the power distribution area.
[0087] In this implementation, the symbols and meanings of the abnormal impact degree indicators are shown in Table 3 below:
[0088] Table 3. Symbols and meanings of indicators for the degree of abnormal impact.
[0089] symbol meaning <![CDATA[S f ]]> Abnormal impact score imp Importance of Taiwan <![CDATA[s imp ]]> Importance score of the Taiwan region cap The total assembly capacity of the transformer area <![CDATA[s cap ]]> Unit assembly capacity score
[0090] Specifically, the importance factor of the transformer area and the total assembly capacity factor of the transformer area can be converted into a score range of 0-100. For example, the scores for the importance index and the total assembly capacity index of the transformer area can be calculated using the following formulas:
[0091]
[0092]
[0093] Where imp represents the importance of the station area, s imp The score represents the importance index of the transformer area, cap represents the total assembly capacity of the transformer area, max(cap) represents the maximum value of the total assembly capacity of all transformer areas, and s cap This indicates the score for the total assembly capacity index of the transformer area.
[0094] In this embodiment, if the importance level of the station area is "important", the importance score of the station area is 100; if the importance level of the station area is "average" or empty, the importance score of the station area is 0.
[0095] Furthermore, a weighted average can be calculated using the importance factor of the transformer area and the total assembly capacity factor of the transformer area. Before weighting the average, the weights of each factor are configured as follows: the default weight for the importance score of the transformer area is 0.8, i.e., w imp =0.8, the default weight of the total assembly capacity of the transformer area is 0.2, i.e., w cap =0.2.
[0096] Furthermore, the score for the degree of abnormal impact can be calculated using the following formula:
[0097]
[0098] Among them, w imp w represents the weight of the importance of each channel. cap S represents the weight of the total assembly capacity of the distribution area. f The score represents the degree of impact of the abnormality.
[0099] Step S3: Determine the early warning threshold for multi-level transformer area equipment upgrades. Based on the comprehensive score of transformer area equipment upgrade priority and the early warning threshold for multi-level transformer area equipment upgrades, determine the early warning level for transformer area equipment upgrades.
[0100] Specifically, the above steps can be used to obtain the scores of historical abnormal operation indicators, potential abnormal indicators, and abnormal impact degree indicators. Then, based on the above three scores, the comprehensive evaluation score, namely the comprehensive score of the equipment renovation priority of the transformer area, is calculated. Finally, the early warning level of the equipment renovation of the transformer area is determined based on the comprehensive evaluation score.
[0101] In this embodiment, the symbols and meanings of the indicators for evaluating the overall evaluation score are shown in Table 4 below:
[0102] The symbols and meanings of the indicators in the comprehensive evaluation score
[0103]
[0104] In this embodiment, the scores of the above three dimensions can be weighted and averaged to obtain the comprehensive evaluation score. Before weighting and averaging, the weights of each dimension can be configured, such as the weight of the historical dimension: w h =0.5; Prediction dimension weight: w p =0.4; Influence dimension weight: w f =0.1, and then the comprehensive evaluation score is calculated using the following formula:
[0105]
[0106] Furthermore, the comprehensive evaluation score calculated in the previous step can be converted into an early warning level based on the preset division threshold in the early warning level threshold configuration table, i.e., the multi-level transformer area equipment modification early warning threshold.
[0107]
[0108] Where, wt l1 =90 is the warning threshold for equipment upgrades in the first-level distribution area, wt l2 =80 is the warning threshold for equipment upgrades in the secondary distribution area, wt l3 =70 is the early warning threshold for equipment upgrades in the third-level distribution area.
[0109] Step S4: Determine the priority of equipment upgrades in the transformer substation based on the early warning level of the equipment upgrade in the substation.
[0110] Once the early warning level for equipment upgrades in each transformer substation is determined, the priority for equipment upgrades in each substation can be determined.
[0111] In summary, this invention, through its comprehensive evaluation method for distribution network transformation, can optimize the upgrading and transformation strategy of distribution network equipment. Specifically, it can continuously improve the effectiveness of equipment upgrading and transformation, and enhance equipment operational stability and power supply capacity by constructing a data-driven intelligent distribution network equipment management system. Furthermore, it can improve load forecasting capabilities and regional operation and maintenance efficiency. Specifically, it can accurately predict the distribution of regional load hotspots through in-depth analysis of the load levels of distribution network equipment, provide intelligent decision support, optimize inspection modes, improve operation and maintenance efficiency, and achieve proactive service based on data-driven approaches. Finally, it can strengthen the lean management of equipment operation by business personnel. Specifically, it can provide detailed equipment operation analysis reports and anomaly diagnosis reports for business personnel through comprehensive and accurate quantitative analysis of the power supply operation of distribution network equipment, thereby improving the level of lean management.
[0112] 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.
[0113] Although the present invention has been described in detail through the preferred embodiments above, it should be understood that the above description should not be considered as a limitation of the present invention. Various modifications and substitutions to the present invention will be apparent to those skilled in the art after reading the above description. Therefore, the scope of protection of the present invention should be defined by the appended claims.
Claims
1. A comprehensive evaluation method for optimizing investment in transformer substation renovation, characterized in that, include: Several comprehensive evaluation indicators for the investment in the renovation of transformer substations were determined, including historical abnormality indicators, potential abnormality indicators, and indicators of the degree of abnormal impact. Calculate the index scores corresponding to the comprehensive evaluation indicators of the renovation investment of each transformer substation, and calculate the comprehensive score of the priority of equipment renovation of the transformer substation based on the scores of each index and the corresponding index weights. Determine the early warning thresholds for equipment upgrades in multiple transformer substations, and determine the early warning level for equipment upgrades in the transformer substations based on the comprehensive score of the equipment upgrade priority and the early warning thresholds for equipment upgrades in the multiple substations. The priority of equipment upgrades in the transformer area is determined based on the early warning level of the equipment upgrade in that area. The steps for calculating the potential anomaly index score include: A time series prediction model for transformer area load based on an attention mechanism is constructed, and the transformer area load is predicted according to the time series prediction model to obtain the transformer area load prediction result. Based on the load forecast results of the transformer area, the following indicators are statistically analyzed: maximum predicted maximum load rate, mean maximum predicted load rate, upper quartile of maximum predicted load rate, sliding predicted maximum load rate, mean sliding predicted load rate, and upper quartile of sliding predicted load rate. The final predicted load of the transformer area is calculated based on the statistical indicators of the transformer area load forecast results. A multi-level potential anomaly warning threshold is determined. Based on the final predicted load of the transformer area and the multi-level potential anomaly warning thresholds, the potential anomaly index score is calculated using the following formula: Among them, S p For potential outlier indicators, lr p For the final predicted load of the distribution area, t l1 The threshold for Level 1 potential anomaly early warning, t l2 The threshold for secondary potential anomalies is t. l3 This is the threshold for a Level 3 potential anomaly warning.
2. The comprehensive evaluation method for optimizing investment in transformer substation renovation as described in claim 1, characterized in that, The steps for calculating historical operational anomaly index scores include: Determine the historical heavy overload frequency index, the historical other abnormal frequency index, and the number of days since today in the past year for the transformer area; The historical operational anomaly index score is calculated based on the historical heavy overload frequency index of the transformer area, the historical other anomaly frequency index of the transformer area, and the number of days since the last year.
3. The comprehensive evaluation method for optimizing investment in transformer substation renovation as described in claim 2, characterized in that, The historical operational anomaly index score is calculated using the following formula: Among them, S h h is the score for historical abnormal operation indicators. c For the historical heavy overload frequency index of the power station area, o c d represents the number of other historical anomalies in the Taiwan area, and d represents the number of days since today (within the past year).
4. The comprehensive evaluation method for optimizing investment in transformer substation renovation as described in claim 3, characterized in that, The historical overload frequency index of the transformer area is used to count the historical overload frequency and the historical heavy load frequency of the transformer area. The historical other abnormal frequency index of the transformer area is used to count the historical three-phase severe imbalance frequency, low voltage frequency, and low power factor frequency of the transformer area. The number of days from the last year to today index is used to count the number of days from a preset date in the previous year to today.
5. The comprehensive evaluation method for optimizing investment in transformer substation renovation as described in claim 1, characterized in that, The steps for calculating the score of the anomaly impact index include: The importance index and the total assembly capacity index of the transformer area are determined as indicators for evaluating the degree of impact of transformer area anomalies; Calculate the importance index score and the total assembly capacity index score of the transformer area, and calculate the abnormal impact index score based on the importance index score and the total assembly capacity index score of the transformer area.
6. The comprehensive evaluation method for optimizing investment in transformer substation renovation as described in claim 5, characterized in that, The importance score of the transformer area and the total installation capacity score of the transformer area are calculated using the following formulas: Where imp represents the importance of the station area, s imp The score represents the importance index of the transformer area, cap represents the total assembly capacity of the transformer area, max(cap) represents the maximum value of the total assembly capacity of all transformer areas, and s cap This indicates the score for the total assembly capacity index of the transformer area.
7. The comprehensive evaluation method for optimizing investment in transformer substation renovation as described in claim 6, characterized in that, The score of the abnormal impact level index is calculated using the following formula: Among them, w imp w represents the weight of the importance of each channel. cap S represents the weight of the total assembly capacity of the distribution area. f The score represents the degree of impact of the abnormality.