A line loss management method and system for seasonal power consumption of a transformer area

Abstract

The application relates to the technical field of power system management, and discloses a line loss treatment method and system for seasonal power consumption of a transformer area, the method comprising the following steps: acquiring transformer area load data and constructing a load distribution feature map; establishing a load-structure coupling relationship matrix according to the feature map; identifying high-loss risks and formulating a path adjustment scheme according to the matrix; further evaluating bearing pressure according to the scheme, and dynamically optimizing the matrix when a fault early warning is triggered to obtain an updated risk control matrix; and finally reevaluating risks in combination with real-time load data to obtain a line loss fine control scheme. The method can accurately identify high-loss areas, significantly improve power grid operation efficiency, reduce line loss rate, and ensure power supply stability and equipment safety.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of power system management, and in particular to a line loss management method and system for seasonal power consumption in a transformer area. BACKGROUND

[0002] At present, in the field of power system management, transformer area line loss management is an important link to ensure the efficiency of power grid operation. Line loss directly affects the effectiveness of power transmission, especially in the case of seasonal power consumption fluctuation, reasonable control of line loss can improve power supply reliability. With the development of smart grid, a large amount of industrial big data is generated in power grid operation. How to use these data to analyze and respond to complex power grid environment and variable power demand has become a difficult problem to be solved in line loss management.

[0003] In one prior art, the main implementation is to develop management strategies from a single dimension, for example, it may only focus on local device optimization or mainly based on historical load data for static analysis. This method ignores the deep connection between power grid structure and load distribution, resulting in lack of flexibility and accuracy of management strategies. Especially in the face of seasonal power consumption peak or valley, this limitation makes line loss problems occur frequently, and even may cause instability of local power grid.

[0004] In summary, there is a problem in the prior art that it is difficult to accurately identify high loss risk areas and perform dynamic optimization. SUMMARY

[0005] The present application provides a line loss management method and system for seasonal power consumption in a transformer area to solve the problem of difficulty in accurately identifying high loss risk areas and performing dynamic optimization.

[0006] In a first aspect, to solve the above technical problems, the present application provides a line loss management method for seasonal power consumption in a transformer area, comprising:

[0007] Obtain real-time load data and historical seasonal power consumption records of the transformer area, and perform time series change analysis and feature labeling to obtain a load distribution feature map;

[0008] According to the load distribution feature map, perform node interaction and structure feature fusion processing to obtain a load-structure coupling relationship matrix;

[0009] According to the load-structure coupling relationship matrix, perform risk area identification and screening to obtain a high loss risk distribution list;

[0010] According to the high loss risk distribution list, perform path bottleneck analysis and optimization to obtain a path adjustment scheme;

[0011] According to the path adjustment scheme, a bearing pressure risk assessment is performed, it is judged whether a device failure warning is triggered, if triggered, a matrix dynamic optimization is performed, and an updated risk control matrix is obtained;

[0012] According to the updated risk control matrix and the real-time load data, a risk area reevaluation is performed, and a line loss fine control scheme is obtained.

[0013] Preferably, the real-time load data and historical seasonal electricity consumption records of the transformer area are obtained, and time series change analysis and feature labeling are performed to obtain a load distribution feature map, including:

[0014] The real-time load data and historical seasonal electricity consumption records of the transformer area are collected from the smart meter;

[0015] According to the real-time load data and the historical seasonal electricity consumption records, denoising processing is performed to obtain cleaned load time series data;

[0016] According to the cleaned load time series data, a connection relationship graph is constructed to obtain a preliminary connection relationship graph;

[0017] According to the preliminary connection relationship graph, load weight calculation and feature labeling are performed to obtain a load distribution feature map.

[0018] Preferably, the line loss management method for seasonal electricity consumption of the transformer area, according to the load distribution feature map, node interaction and structure feature fusion processing are performed to obtain a load-structure coupling relationship matrix, including:

[0019] According to the load distribution feature map, interactive data extraction and impedance characteristic analysis are performed to obtain a load interaction data set and a line impedance feature set;

[0020] According to the load interaction data and the line impedance feature set, fusion processing is performed to obtain a fused load and structure correlation data set;

[0021] According to the load and structure correlation data set, a matrix is generated to obtain a load-structure coupling relationship matrix.

[0022] Preferably, the line loss management method for seasonal electricity consumption of the transformer area, according to the load-structure coupling relationship matrix, risk area identification and screening are performed to obtain a high loss risk distribution list, including:

[0023] According to the load-structure coupling relationship matrix, the correlation strength is compared, and the nodes with correlation strength greater than the preset threshold value of correlation strength are identified as to-be-analyzed nodes to obtain a to-be-analyzed node set;

[0024] The node set to be analyzed is subjected to risk condition judgment with a preset risk feature library, and nodes meeting the condition are screened as high-loss risk nodes to generate a high-loss risk node list;

[0025] According to the high-loss risk node list, multi-dimensional feature extraction and integration are performed to obtain a high-loss risk distribution list.

[0026] Preferably, in the line loss governance method for seasonal power consumption of a transformer area, according to the high-loss risk distribution list, path bottleneck analysis and optimization are performed to obtain a path adjustment scheme, which includes:

[0027] According to the high-loss risk distribution list, time series data extraction and constraint characteristic analysis are performed to obtain a load fluctuation correlation set.

[0028] According to the load fluctuation correlation set, path bottleneck evaluation and marking are performed to obtain a key path list.

[0029] According to the key path list, constraint data acquisition and feasibility judgment are performed to obtain a priority ranking result.

[0030] According to the priority ranking result, a path scheme is generated to determine the path adjustment scheme.

[0031] Preferably, in the line loss governance method for seasonal power consumption of a transformer area, according to the path adjustment scheme, bearing pressure risk evaluation is performed to determine whether a device fault warning is triggered, and if triggered, matrix dynamic optimization is performed to obtain an updated risk control matrix, which includes:

[0032] According to the path adjustment scheme, overload signal acquisition is performed to obtain a classified signal data set.

[0033] According to the classified signal data set, bearing pressure index calculation is performed to obtain a pressure index set.

[0034] The pressure index set is compared with a preset overload warning threshold, and if the pressure index set exceeds the overload warning threshold, a triggered device fault warning signal is determined.

[0035] According to the triggered device fault warning signal, parameter deviation positioning is performed to obtain a parameter range that needs to be adjusted.

[0036] According to the parameter range that needs to be adjusted, matrix optimization and iterative calculation are performed to obtain an updated risk control matrix.

[0037] Preferably, in the line loss governance method for seasonal power consumption of a transformer area, according to the updated risk control matrix and the real-time load data, risk area reevaluation is performed to obtain a line loss fine control scheme, which includes:

[0038] According to the real-time load data and the updated risk control matrix, synchronous comparison and error correction are performed to obtain an adjusted data matching precision and a synchronous risk control matrix;

[0039] According to the data matching precision and the synchronous risk control matrix, regional division adjustment is performed to determine a potential high-loss area range;

[0040] According to the potential high-loss area range, risk area reevaluation is performed to obtain an updated risk area division result;

[0041] According to the updated risk area division result, a control scheme is generated to obtain a line loss fine control scheme.

[0042] In a second aspect, the present application provides a line loss management system for seasonal electricity consumption in a transformer area, comprising:

[0043] A graph construction module is configured to obtain real-time load data and historical seasonal electricity consumption records of the transformer area, and perform time series change analysis and feature labeling to obtain a load distribution feature graph;

[0044] A coupling relationship analysis module is configured to perform node interaction and structure feature fusion processing according to the load distribution feature graph to obtain a load-structure coupling relationship matrix;

[0045] A risk area identification module is configured to perform risk area identification and screening according to the load-structure coupling relationship matrix to obtain a high-loss risk distribution list;

[0046] A path optimization module is configured to perform path bottleneck analysis and optimization according to the high-loss risk distribution list to obtain a path adjustment scheme;

[0047] An evaluation and optimization module is configured to perform bearing pressure risk evaluation according to the path adjustment scheme to determine whether to trigger a device fault warning, and if triggered, to perform matrix dynamic optimization to obtain an updated risk control matrix;

[0048] A control scheme generation module is configured to perform risk area reevaluation according to the updated risk control matrix and the real-time load data to obtain a line loss fine control scheme.

[0049] Compared with the prior art, the present application has the following beneficial effects:

[0050] (1) The present application constructs a load distribution characteristic map by collecting real-time load data and historical seasonal power consumption records in the transformer area, and determines the coupling relationship matrix of the load and structure in the power grid by using neighborhood information aggregation and structure feature extraction technology (such as analyzing line impedance). This way deeply integrates and quantifies the dynamic load change and the power grid topology structure, solves the problem of existing technology that only analyzes from a single dimension and ignores the deep correlation between the two, so as to accurately capture the inherent characteristics of the power grid structure and its complex interaction with dynamic load, and provides a scientific basis for accurately identifying line loss sources.

[0051] (2) The present application determines the high loss risk area by using risk feature recognition technology based on the coupling relationship matrix, and uses dynamic interactive modeling and path impact analysis to evaluate the bottleneck for the list, so as to determine the optimized path adjustment scheme. This method can locate the risk based on the coupling strength of load and structure, and simulate the impact of different paths combined with time sequence correlation, so as to realize accurate identification and dynamic path optimization of high loss risk area, make the treatment strategy more flexible and targeted, and effectively reduce the line loss caused by structure or load bottleneck.

[0052] (3) The present application obtains overload signals from the equipment state monitoring system, and starts the matrix update trigger mechanism when triggering the equipment fault warning, dynamically adjusts the coupling relationship matrix combined with historical data backtracking and feedback iterative calculation, and finally reevaluates the risk area through real-time data synchronization. This builds a closed-loop feedback optimization process, so that the line loss treatment scheme can be real-time corrected and iteratively optimized according to the actual bearing pressure of the equipment and sudden warning, ensuring the long-term effectiveness and adaptability of the control strategy, and significantly improving the power supply stability and equipment safety. BRIEF DESCRIPTION OF DRAWINGS

[0053] Figure 1 is a line loss treatment method flowchart for seasonal power consumption in a transformer area provided by the first embodiment of the present application;

[0054] Figure 2 is a line loss treatment system structure schematic diagram for seasonal power consumption in a transformer area provided by the second embodiment of the present application. DETAILED DESCRIPTION

[0055] The technical solutions in the embodiments of the present application will be described clearly and completely below with reference to the drawings in the embodiments of the present application. Obviously, the described embodiments are only part of the embodiments of the present application, not all. Based on the embodiments in the present application, all other embodiments obtained by those skilled in the art without creative labor are within the scope of protection of the present application.

[0056] REFERENCE Figure 1The first embodiment of the present application provides a line loss management method for seasonal power consumption of a transformer area, comprising the following steps:

[0057] S11, real-time load data and historical seasonal power consumption records of the transformer area are acquired, time series change analysis and feature labeling are performed, and a load distribution feature map is obtained;

[0058] S12, according to the load distribution feature map, node interaction and structure feature fusion processing are performed, and a load-structure coupling relationship matrix is obtained;

[0059] S13, according to the load-structure coupling relationship matrix, risk area identification and screening are performed, and a high loss risk distribution list is obtained;

[0060] S14, according to the high loss risk distribution list, path bottleneck analysis and optimization are performed, and a path adjustment scheme is obtained;

[0061] S15, according to the path adjustment scheme, bearing pressure risk assessment is performed, it is judged whether device fault early warning is triggered, if triggered, matrix dynamic optimization is performed, and an updated risk control matrix is obtained;

[0062] S16, according to the updated risk control matrix and the real-time load data, risk area reevaluation is performed, and a line loss fine control scheme is obtained.

[0063] In step S11, real-time load data and historical seasonal power consumption records of the transformer area are acquired, time series change analysis and feature labeling are performed, and a load distribution feature map is obtained, comprising:

[0064] The real-time load data and the historical seasonal power consumption records of the transformer area are collected from the smart meter;

[0065] According to the real-time load data and the historical seasonal power consumption records, denoising processing is performed, and cleaned load time series data is obtained;

[0066] According to the cleaned load time series data, connection relationship graph construction is performed, and a preliminary connection relationship map is obtained;

[0067] According to the preliminary connection relationship map, load weight calculation and feature labeling are performed, and a load distribution feature map is obtained.

[0068] In one implementation, firstly, real-time load data for the distribution area is acquired at fixed collection intervals (e.g., every 15 minutes or 1 hour) using smart meters or corresponding concentrators deployed within the area. The real-time load data is preferably a set of time-series data, with each record containing a timestamp, active power, reactive power, voltage, and current. Simultaneously, historical seasonal electricity consumption records are retrieved from the electricity marketing system or the master station database. The electricity marketing system stores long-term electricity consumption information and settlement data for each user within the distribution area. The data source for this database is frozen data (e.g., daily frozen, monthly frozen data) periodically reported by the Advanced Metering Architecture (AMI) system, and is updated with the settlement cycle. The integration operation aligns the real-time load data with the historical seasonal electricity consumption records according to the distribution area identifier and timestamp, aggregating them into a unified dataset to construct an initial time-series change dataset.

[0069] The next step is to perform data cleaning and noise reduction on the initial time-series variation dataset. In one implementation, firstly, the mean of the load data for a specific time period (e.g., all historical summer midday peak periods) in the dataset is calculated. and standard deviation Then, iterate through the data points in the dataset. Defined as the active power value collected at time t. Judgment Is it an outlier? If a certain data point satisfy or If a data point is identified as an outlier, it is determined to be either an outlier or noise. Data points identified as outliers are then processed. One approach is to mark them as invalid data and remove them. Another approach is to use linear interpolation for correction; for example, if... If an error is detected, find its preceding valid data point. (at time ta) and the next valid data point (At time t+b). Correction value The calculation is as follows:

[0070]

[0071] Where a is the time interval between time t and time ta, and b is the time interval between time t+b and time t. After processing all outliers, the resulting dataset is the cleaned load time series data.

[0072] Then, a connection relationship graph of the transformer area is constructed according to the cleaned load time series data. First, electrical equipment account books (such as transformers, lines, switches) and physical connection information in the transformer area are extracted from a transformer area asset management system or a geographic information system (GIS). Then, a graph is constructed. Specifically, transformer areas, main feeder branch points, or large users are defined as nodes (Node); and electrical connection lines between nodes are defined as edges (Edge). Each edge (Edge) is associated with its physical parameters, such as line type, length, and material, which are used to calculate line impedance. The constructed graph is a preliminary connection relationship graph. The preliminary connection relationship graph can be mathematically represented as wherein is a node set, is an edge set, and the graph reflects the physical electrical connection structure of the transformer area.

[0073] For each node in the preliminary connection relationship graph , load distribution weight calculation is performed in combination with the cleaned load time series data. In an implementation manner, the load distribution weight of the node is defined as the proportion of the average load of the node in a specific analysis period (for example, a historical seasonal peak period) to the total average load of the transformer area. The calculation formula is as follows,

[0074]

[0075] wherein, is the load distribution weight of the node , is the average active load of the node in the period, which is calculated according to the cleaned load time series data, is the total number of nodes, is the sum of the average active loads of all nodes in the transformer area. To avoid a zero denominator, it is necessary to check before calculation. If the total load of the transformer area is zero, the load distribution weight of all nodes is set to 0.

[0076] Then, a load weight preset threshold is set. The setting of the load weight preset threshold is based on the identification of a key node with the highest contribution to the load of the transformer area. One setting method is to set the 90th percentile of the distribution of all nodes in the history of the transformer area as the load weight preset threshold ​The 90th percentile is selected because it can statistically effectively screen the top 10% of nodes with heavy load, which are usually the main contributors to load distribution imbalance and line loss. Concentrating analysis resources on these key nodes can improve governance efficiency.

[0077] Then, feature labeling is performed. All nodes are traversed and compared with . If , the node is labeled as a “high-load node”. If , it is labeled as a “regular-load node”. Finally, the load distribution weight and the feature label (such as “high-load node”) are attached to the corresponding node of the preliminary connection graph as attributes, obtaining a load distribution feature graph.

[0078] In step S12, according to the load distribution feature graph, node interaction and structural feature fusion processing is performed to obtain a load-structure coupling relationship matrix, including:

[0079] According to the load distribution feature graph, interactive data extraction and impedance characteristic analysis are performed to obtain a load interaction data and line impedance feature set;

[0080] According to the load interaction data and the line impedance feature set, fusion processing is performed to obtain a fused load and structure correlation data set;

[0081] According to the load and structure correlation data set, a matrix is generated to obtain a load-structure coupling relationship matrix.

[0082] In one implementation, first, impedance characteristic analysis is performed. Each edge (line) in the load distribution feature graph is traversed , and according to the line type and line length of the line obtained in S11 , a preset line parameter table is queried. The line parameter table is constructed according to the Electrical Engineering Handbook or the equipment manufacturer's specification book, and its content maps the line type and the corresponding unit length resistance (Ω / km) and unit length reactance (Ω / km). The line resistance is obtained by calculating , the line reactance is obtained by calculating , and all are combined into a line impedance feature set.

[0083] Then, the interactive data extraction is performed, which is accomplished by constructing and solving a base power flow model. The construction of the base power flow model is first to calculate the line admittance matrix according to the power grid topology of S11 and the set of line impedance characteristics obtained in this step, and to construct the node admittance matrix of (wherein is the total number of nodes). The matrix is the core parameter of the subsequent power flow calculation. At the same time, according to the cleaned load time series data obtained in S11, the average active load and the average reactive load of each node i are extracted, and are set as the active injection power and the reactive injection power of the node. In addition, a node (for example, the main node of the transformer in the area) is selected as the slack bus, and its voltage amplitude (for example, 1.0 p.u.) and phase angle (for example, 0 degrees) are set as the base.

[0084] The mathematical core of the base power flow model is the nonlinear power system power flow equation set. For any node i except the slack bus, the injection power and satisfy:

[0085] ,

[0086] ,

[0087] wherein and are the voltage amplitude and phase angle to be solved, and are the voltage amplitude and phase angle of node k, if node k is the slack bus, the voltage amplitude and phase angle are the set values and , if node k is a non-slack bus, the voltage amplitude and phase angle are the current iteration values is an element of the node admittance matrix , is the conductance, is the susceptance.

[0088] Then, the power flow calculation, for example, the Newton-Raphson method, is applied to iteratively solve the equation set. First, the of all non-slack nodes is set to 1.0 p.u., is set to 0 degrees, and the initial is 0; then in the th iteration, according to the current and Calculate the computational power of the node. and and with the set injection power and By comparison, the power imbalance is obtained. and Determine if the maximum imbalance is less than the preset convergence accuracy. (For example If the condition is met, the iteration converges. If it does not converge, the Jacobian matrix for the current state is constructed. Solve the system of linear corrected equations:

[0089]

[0090] in, and These are the vectors of active and reactive power imbalance, respectively. and These are the correction vectors for the phase angle and voltage magnitude, respectively. After solving for the corrections, the state is updated. and Return to the next iteration. (The superscript is missing from the original text.) Indicates the current iteration number. Indicates the next iteration number; and They represent the first The node voltage phase angle vector and voltage magnitude vector at the next iteration; and They represent the first The phase angle correction and voltage amplitude correction calculated in the next iteration; and They represent the updated version, used for the first... The node voltage phase angle vector and voltage magnitude vector for each iteration. When the iteration converges, the final state obtained is the average voltage magnitude of each node. and phase angle And based on these results, further calculations were performed for each line. Average active power of the upward flow With average reactive power The calculated average voltage amplitude, average active power, and average reactive power are combined to form the load interaction data.

[0091] Then, a fusion process is performed. This step involves calculating the loss of each line and filtering out high-loss lines. Specifically, all lines are traversed. Combined with the aforementioned load interaction data (average active power) , average reactive power and average voltage amplitude ) and the line impedance characteristic set (line resistance ), the active power loss of the line is calculated , the formula of which is,

[0092]

[0093] To avoid the denominator being zero, if the average voltage amplitude is zero, the active power loss is set to 0. At the same time, the input active power of the line is calculated, the formula of which is,

[0094]

[0095] wherein, here, refers to the power flowing to the j node, is the loss on the line, and the sum of the two is the power injected from the i node into the line. According to the above two calculation results, the influence degree of the line is calculated, which is defined as the active loss rate, and the formula of which is,

[0096]

[0097] To avoid the denominator being zero, if the input active power is zero, the influence degree is set to 0.

[0098] Then, a line influence degree threshold is set, which is used to screen out lines with abnormal loss rate. The threshold is preferably set to 30%, which is a typical engineering value in power system analysis for distinguishing between normal technical loss and high loss that needs to be focused on; in actual application, it can be flexibly adjusted according to the voltage level of the specific transformer area, historical line loss statistics, and operation management requirements, for example, for transformer areas with strict operation standards, it can be set to 15% to 20%; for transformer areas with large load fluctuations and general line conditions, it can be appropriately relaxed to 30% to 40%, the purpose of which is to screen out lines with abnormally high loss rate relative to the average or normal level of the transformer area. Subsequently, the influence degree is compared with a line influence degree threshold , and according to the comparison result, the correlation value of the line is determined, including the following cases, case one, if the influence degree is greater than the line influence degree threshold, the line is determined to be a high coupling line, and its correlation value set the active power loss Case two, if the influence degree is not greater than the line influence degree threshold, it is determined that the line is a non-critical line, and the associated value of the line is set to 0. Finally, all edges and the corresponding associated values are summarized to obtain a fused load and structure associated data set.

[0099] Finally, a load-structure coupling relationship matrix M is constructed according to the fused load and structure associated data set. Specifically, first, set the total number of nodes in the transformer area to (i.e. the number of nodes in the load distribution feature map), and construct a zero matrix M of size. Traverse each element in the fused load and structure associated data set, and set the element in the i-th row and j-th column of M and the element in the j-th row and i-th column of M to the associated value . The final M is the load-structure coupling relationship matrix, which is a symmetric matrix, and the element quantitatively describes the active power loss value between nodes i and j caused by the combined action of high load interaction and high impedance.

[0100] It should be noted that in the present embodiment, the load-structure coupling relationship matrix M is constructed as a symmetric matrix, which is a simplified and preferred implementation based on the radial network structure of the transformer area and the calculation characteristics of the active power loss (which is a scalar and has considered the bidirectional symmetric parameter of line resistance during calculation); this simplification helps to reduce the complexity of the model and has met the accuracy requirements of most transformer area line loss analysis; in applications that require more accurate bidirectional influence, an asymmetric matrix can also be constructed, for example, defining and as the power loss from node to node and from node to node respectively.

[0101] In step S13, according to the load-structure coupling relationship matrix, risk area identification and screening are performed to obtain a high loss risk distribution list, including:

[0102] According to the load-structure coupling relationship matrix, the correlation strength is compared, and nodes with a correlation strength greater than a preset correlation strength threshold are identified as to-be-analyzed nodes to obtain a to-be-analyzed node set;

[0103] The set of nodes to be analyzed is subjected to risk condition judgment with a preset risk feature library, and nodes meeting the conditions are screened as high-loss risk nodes to generate a high-loss risk node list;

[0104] According to the high-loss risk node list, multi-dimensional feature extraction and integration are performed to obtain a high-loss risk distribution list.

[0105] First, according to the load-structure coupling relationship matrix M generated in the foregoing steps, the correlation strength is compared and the nodes are identified. Then, a correlation strength preset threshold is set . The basis for setting the threshold is to statistically screen out the key connections that contribute most to the total loss of the transformer area. One setting method is to statistically analyze the distribution of all non-zero elements (representing the loss values of the screened high-impact lines) in the load-structure coupling relationship matrix M, and set the 90th percentile of these values as the correlation strength preset threshold . The 90th percentile is selected because it can effectively separate the "heavy tail" part of the loss distribution in statistics, that is, locate the 10% lines that contribute most of the loss, so as to concentrate resources for management.

[0106] Subsequently, all elements in the load-structure coupling relationship matrix M are traversed , compared with the correlation strength preset threshold , if , nodes i and j are identified as nodes to be analyzed; if , no marking is performed. All marked nodes to be analyzed are summarized to form a set of nodes to be analyzed.

[0107] Then, the set of nodes to be analyzed is compared with the preset risk feature library to determine the risk conditions, and nodes meeting the conditions are screened as high-loss risk nodes. First, a pre-established risk feature library is needed. The feature library is constructed based on analysis of historical fault logs (obtained from an operation and maintenance management system O&M) and historical operation data (obtained from a SCADA database, which is dynamically updated with the operation of the O&M system and the SCADA system). The feature library stores a series of structured risk conditions, such as: {“feature”: “historical cumulative fault times”, “judgment”: “>”, “value”: 2} or {“feature”: “voltage fluctuation amplitude”, “judgment”: “>”, “value”: 5%}. The feature library is updated with the input of new operation data and fault analysis reports (for example, every quarter). Then, each node in the set of nodes to be analyzed is traversed, and its corresponding feature data (for example, the actual historical fault times of a certain node are 3 times) is obtained from the SCADA system or the O&M system. The extracted feature data is compared and screened with the risk conditions in the risk feature library. If the feature data meets any risk condition in the library (such as 3>2), the node is determined to be a confirmed high-loss risk node; if it does not meet any risk condition, it is excluded from the high-risk list. All confirmed nodes are aggregated to form a high-loss risk node list.

[0108] Finally, according to the high-loss risk node list, multi-dimensional feature extraction and integration are performed. This integration operation creates a structured data record for each node (or node pair) in the list. Specifically, the identification of the node (or node pair), the calculated correlation strength, and the hit risk condition (such as “historical cumulative fault times > 2”) are used as basic fields, and additional multi-dimensional features are extracted and filled from multiple data sources, such as load characteristics (such as load peak value) from the SCADA system, asset characteristics (such as line aging degree, which can be calculated by subtracting the commissioning date from the current date) from the asset management system (which stores the equipment inventory, model, commissioning date, and maintenance records), and environmental characteristics (such as whether the node is located in a high-temperature and high-humidity area) by spatially superimposing device coordinates with weather data (such as historical temperature and humidity distribution maps) from the geographic information system (GIS). All the above fields are integrated into a structured data record, and all records are aggregated to form a final high-loss risk distribution list.

[0109] In step S14, according to the high-loss risk distribution list, path bottleneck analysis and optimization are performed to obtain a path adjustment scheme, including:

[0110] According to the high-loss risk distribution list, time series data extraction and constraint characteristic analysis are performed to obtain a load fluctuation correlation set;

[0111] According to the load fluctuation association set, path bottleneck assessment and marking are performed to obtain a key path list;

[0112] According to the key path list, constraint data acquisition and feasibility judgment are performed to obtain a priority ranking result;

[0113] According to the priority ranking result, path scheme generation is performed to determine a path adjustment scheme.

[0114] First, according to the high loss risk distribution list obtained in S13, time series data extraction and constraint characteristic analysis are performed. First, each high loss node or line (for example, "line A-B") in the list is traversed, and the corresponding historical load fluctuation curve (for example, typical daily 24-hour load data) is extracted from the cleaned load time series data obtained in S11 as the load change time series data, taking the identification as the key. At the same time, the structural constraint characteristics of the line are extracted from the asset management system or the power flow calculation result, and the structural constraint characteristics preferably refer to the rated carrying capacity (for example, the rated carrying capacity of line A-B ). These extracted data are integrated into a record, for example: {identification: "line A-B", time series data: [...], rated carrying capacity: 500A}. All records are summarized to form a load fluctuation association set.

[0115] Next, according to the load fluctuation association set, path bottleneck assessment and marking are performed. First, a path bottleneck preset threshold is set. The threshold is set according to the N-1 safety criterion and the operation margin requirement of the power system. In one implementation, the threshold is set to 90% of the rated carrying capacity , that is . The reason for selecting 90% is to reserve 10% of the operation standby margin to ensure that when the adjacent line or device fails (N-1 event), the line can temporarily carry the transferred load without immediately exceeding its emergency overload upper limit, thereby preventing cascading reactions. Traverse each item in the load fluctuation association set, and compare the peak load in the load change time series data with the path bottleneck preset threshold . The following cases may occur. Case one, if (that is, the peak load exceeds 90% of the rated carrying capacity), it is determined that the path has a bottleneck effect, and it is marked as a key path. Case two, if , it is determined that the path has no bottleneck. All marked key paths are summarized to form a key path list.

[0116] Then, according to the key path list, constraint data acquisition and feasibility judgment are performed. This operation traverses each path (for example, "line A-B") in the key path list, and finds the standby line or parallel line (for example, "line A-C-B") connected thereto and having a transferable load capacity in the preliminary connection relationship graph in S11. The current load of the standby line is queried from the power flow calculation result in S12, and the idle capacity (that is ) thereof is calculated, which is the constraint data related to flow optimization. Subsequently, the path adjustment feasibility range of the key path is judged: if there is a standby line with sufficient idle capacity (greater than the load amount that needs to be transferred by the key path, that is ), it is determined that the feasibility is "high"; if there is no standby path (such as a single radial line), it is determined that the feasibility is "low". Finally, the key path list is sorted, and the basis for sorting is the severity of the bottleneck (the difference value of , the greater the difference value, the higher the priority) and the path adjustment feasibility (priority of the feasibility "high"), to obtain a priority sorting result.

[0117] Finally, according to the priority sorting result, a path scheme is generated and simulation verification is performed. This operation extracts the key path with the highest priority and generates a preliminary path adjustment scheme (for example, "transfer X megawatts of load of line A-B to line A-C-B") for it. Simulation verification is completed by performing a new power flow calculation. First, the benchmark power flow model used in S12 (including the topology and load data of S11) is loaded; then, the model parameters are modified according to the preliminary path adjustment scheme (for example, adjust the load distribution ratio of lines A-B and A-C-B or change the switch state to enable the standby path); then, the power flow calculation (such as the Newton-Raphson method) is run to solve the steady state of the modified model. The verification process includes checking whether the load of the original key path (line A-B) in the new power flow result has been reduced to below the path bottleneck preset threshold. Secondly, check whether the load of the standby path (line A-C-B) and other related equipment does not exceed their respective path bottleneck preset thresholds. If both conditions are met, the verification is passed, and the scheme is determined as the final path adjustment scheme; if either condition is not met (such as causing the standby path to be overloaded), the adjustment scheme (for example, reduce the transferred load X) is adjusted and the simulation verification is repeated until a feasible solution is found.

[0118] In step S15, according to the path adjustment scheme, the carrying pressure risk is evaluated, it is judged whether to trigger the equipment failure warning, if triggered, the matrix dynamic optimization is performed, and the updated risk control matrix is obtained, including:

[0119] According to the path adjustment scheme, an overload signal is collected, and a classified signal data set is obtained;

[0120] According to the classified signal data set, a bearing pressure index calculation is performed to obtain a pressure index set;

[0121] The pressure index set is compared with a preset overload early warning threshold value, and if the overload early warning threshold value is exceeded, it is determined that a triggered device failure early warning signal has been triggered;

[0122] According to the triggered device failure early warning signal, parameter deviation positioning is performed to obtain a parameter range that needs to be adjusted;

[0123] According to the parameter range that needs to be adjusted, matrix optimization and iterative calculation are performed to obtain an updated risk control matrix.

[0124] First, according to the path adjustment scheme determined in S14, the key devices involved in the scheme (for example, "line A-C-B") are locked. Overload signal collection refers to the acquisition of real-time operation signals (such as current, voltage, and device body temperature) of these key devices through device state monitoring systems (such as PMU or intelligent circuit breaker) for their specific identification (ID). Organizational operation is to map and associate the collected raw signal data stream with the device ID with the device name ("line A-C-B") in the S14 scheme, so as to obtain a classified signal data set, in which each data is clearly attributed to a monitored device.

[0125] Then, according to the classified signal data set, a bearing pressure index calculation is performed. This step is based on the search and calculation process of the device identification, and each device (such as "line A-C-B") in the data set is traversed, using its identification as a key, to find its corresponding rated bearing capacity in the device account obtained in S11 step (asset management system) (the ability definition is seen in S14). Then, the real-time signal value (for example, real-time current ) of the device is calculated with (for example, rated current ) to generate a pressure index , and the calculation formula is . The pressure indexes of all related devices are summarized to obtain a pressure index set.

[0126] Then, according to the pressure index set, a risk assessment is performed. This operation sets an overload early warning threshold value The threshold is set based on the operating safety procedures, and is preferably set to 95% of the rated load capacity. Choosing 95% provides an early warning before exceeding the 90% planning margin in S14, but before reaching the 100% continuous operating limit, to prevent accelerated equipment aging. Specifically, the evaluation process first iterates through the set of pressure indicators, and... With the overload warning threshold (i.e., 95%) was compared, and the comparison and results are as follows: Case 1, if If the pressure is within a controllable range, then the monitoring of the equipment in this step is complete, and the process is normal. Scenario 2: If (For example, actual measurement) If the load rate of the line predicted by the model in S14 is about 88%, then it is determined to be a triggered equipment fault warning signal.

[0127] Based on the triggered equipment fault warning signal (i.e., entering situation two), this warning indicates a deviation between the baseline power flow model constructed in S12 and reality. Immediately initiate the parameter deviation localization operation, and retrieve the signal values ​​actually observed in S15 that lead to high pressure (e.g., from...). The actual active power loss obtained by reverse calculation The corresponding values ​​(i.e., correlation values) calculated by the model in S12 This value (stored in the load-structure coupling matrix) is compared. The deviation ( The model parameters corresponding to (e.g., the line resistance used in S12) The load timing data after cleaning (or S11) is determined to be inaccurate and is marked as a parameter range that needs to be adjusted.

[0128] Finally, matrix optimization and iterative calculations are performed based on the required parameter range. First, based on the deviation... The parameter values ​​within the range that need adjustment are corrected. Specifically, if the deviation... Positive (i.e.) If the model underestimates the losses, it indicates that the corresponding parameters in the model (such as line resistance) are inaccurate. If the parameter is underestimated, then in the next iteration... Increase the step size by one (e.g., increase by 1%); conversely, if the deviation is negative, decrease the step size by one. Then, using the corrected parameters, re-execute the power flow calculation and fusion processing steps in S12 to obtain a new temporary matrix. Next, calculate the elements of the new matrix. Compared with the actual observed value of S15 relative error between The calculation formula is: The relative error is compared with the preset convergence accuracy .

[0129] The setting is based on the engineering acceptable error range recognized in the field of power system measurement and simulation. A common setting method is to set it to 5%. This value is generally considered as a highly consistent boundary standard between the model (simulation value) and the physical reality (measured value), and is significantly lower than the 10% operating margin in S14, which can ensure that the updated model has sufficient prediction accuracy. The convergence condition is that if (i.e. less than 5%), it is determined that the iterative calculation converges, and the is the updated risk control matrix. If , it is determined that it does not converge, and returns to the first step (parameter correction) and continues to perform recalculation of S12 until the convergence condition is met or the preset maximum number of iterations (e.g. 20 times) is reached. The updated matrix will replace the old matrix and be used for risk reevaluation in S16.

[0130] In step S16, according to the updated risk control matrix and the real-time load data, the risk region is reevaluated to obtain a line loss refinement control scheme, including:

[0131] According to the real-time load data and the updated risk control matrix, synchronous comparison and error correction are performed to obtain an adjusted data matching accuracy and a synchronous risk control matrix;

[0132] According to the data matching accuracy and the synchronous risk control matrix, regional division adjustment is performed to determine the range of potential high loss areas;

[0133] According to the range of potential high loss areas, risk region reevaluation is performed to obtain an updated risk region division result;

[0134] According to the updated risk region division result, a control scheme is generated to obtain a line loss refinement control scheme.

[0135] First, the real-time load data (i.e. the load snapshot at the current time) is obtained (collected), and according to the updated risk control matrix obtained in S15 , synchronous comparison and error correction are performed. Specifically, the updated risk control matrix and its underlying corrected parameters (e.g. the adjusted in S15) are used, and the real-time load data is used as the new input boundary to perform a real-time power flow simulation (this simulation uses the power flow calculation method defined in S12). The simulation solution obtains a model predicted loss value Simultaneously, the actual real-time loss of the line at the same moment is collected from the equipment condition monitoring system. Calculate the synchronization error. Its calculation formula is,

[0136]

[0137] The synchronization error Convergence accuracy as defined in S15 (For example, 5%) for comparison. If If the model still deviates from reality, the system will trigger the iterative correction process in S15 again (i.e., adjusting parameters and recalculating power flow), and re-execute the synchronization comparison in step (S16) using the corrected new matrix, until... .like (Or, after modification, if this standard is met), then the model is determined to be synchronized with reality. At this point, the... The synchronization state is the adjusted data matching accuracy; and the matrix that has been synchronized (i.e., the one passed in S15) The synchronous risk control matrix is ​​the final version of the matrix (or the version modified in Case 1).

[0138] Next, based on the data matching accuracy (i.e., the model's confirmed synchronization with reality) and the synchronization risk control matrix (i.e., the matrix that has passed synchronization comparison), regional division adjustments and risk area reassessments are performed. This operation is a combined process of determining the potential high-loss region range and obtaining updated risk area division results. Specifically, using the synchronization risk control matrix, the risk identification process in S13 is completely repeated. First, the preset threshold for association strength defined in S13 is applied to the synchronization risk control matrix to filter out node pairs with currently excessive association strength. Then, these node pairs are compared with the risk feature database established in S13. Finally, all confirmed regions are summarized to obtain updated risk area division results. This result reflects the distribution of high-loss regions after calibration based on the latest real-world data.

[0139] Finally, according to the updated risk area division result, a control scheme is generated. This scheme is a fine control that is different from the path adjustment (topology change) in S14. One implementation is that, for the newly identified risks in the updated risk area division result (for example, a certain line voltage fluctuation is re-evaluated as a high risk), a fine control instruction is automatically generated and issued. The instruction preferably includes sending an adjustment tap instruction to the on-load tap changer (OLTC) in the transformer area to stabilize the voltage of the area; or sending a switching instruction to the reactive power compensation device (such as a capacitor bank) in the area to optimize the power factor of the area, thereby realizing real-time fine control of line loss. All generated instructions are summarized, which is the final line loss fine control scheme.

[0140] In summary, the present application accurately identifies high-loss risk areas by constructing a load distribution feature map and a coupling relationship matrix, dynamically updates the risk assessment model based on path optimization and device early warning, forms a closed-loop fine control scheme, and realizes the improvement of power grid operation efficiency and the reduction of line loss rate, ensuring the stability of power supply and the safety of equipment.

[0141] Referring to Figure 2 The second embodiment of the present application provides a line loss management system for seasonal power consumption in a transformer area, comprising:

[0142] a map construction module for obtaining real-time load data and historical seasonal power consumption records of the transformer area, and performing time series change analysis and feature labeling to obtain a load distribution feature map;

[0143] a coupling relationship analysis module for performing node interaction and structure feature fusion processing based on the load distribution feature map to obtain a load-structure coupling relationship matrix;

[0144] a risk area identification module for identifying and screening risk areas based on the load-structure coupling relationship matrix to obtain a high-loss risk distribution list;

[0145] a path optimization module for performing path bottleneck analysis and optimization based on the high-loss risk distribution list to obtain a path adjustment scheme;

[0146] an evaluation and optimization module for performing load pressure risk assessment based on the path adjustment scheme to determine whether to trigger a device fault warning, and if so, performing matrix dynamic optimization to obtain an updated risk control matrix;

[0147] a control scheme generation module for re-evaluating the risk area based on the updated risk control matrix and the real-time load data to obtain a line loss fine control scheme.

[0148] It should be noted that the line loss management system for seasonal power consumption of a transformer area provided in the embodiments of the present application is used to execute all process steps of the line loss management method for seasonal power consumption of a transformer area provided in the embodiments, and the working principles and beneficial effects of the two are one-to-one correspondence, thus not being described again.

[0149] The embodiments of the present application further provide an electronic device. The electronic device comprises a processor, a memory, and a computer program stored in the memory and executable on the processor, for example, a line loss management program for seasonal power consumption of a transformer area. The processor implements the steps in the line loss management method for seasonal power consumption of a transformer area provided in the embodiments when executing the computer program, for example Figure 1 The processor implements the functions of the modules / units in the systems provided in the embodiments when executing the computer program, for example, a graph construction module.

[0150] For example, the computer program can be divided into one or more modules / units, which are stored in the memory and executed by the processor to complete the present application. The one or more modules / units can be a series of computer program instruction segments capable of completing a specific function, which are used to describe the execution process of the computer program in the electronic device.

[0151] The electronic device can be a desktop computer, a notebook, a palm computer, a smart tablet and the like. The electronic device can include, but is not limited to, a processor, a memory. Those skilled in the art can understand that the above components are only examples of the electronic device and do not constitute a limitation on the electronic device, and can include more or fewer components than the above, or combine certain components, or different components, for example, the electronic device can also include an input / output device, a network access device, a bus and the like.

[0152] The processor can be a central processing unit (CPU), and can also be other general-purpose processors, digital signal processors (DSP), application specific integrated circuits (ASIC), field-programmable gate arrays (FPGA) or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components, etc. The general-purpose processor can be a microprocessor or the processor can also be any conventional processor and the like, and the processor is the control center of the electronic device, which connects all parts of the electronic device through various interfaces and lines.

[0153] The memory can be used to store the computer program and / or modules, and the processor realizes various functions of the electronic device by running or executing the computer program and / or modules stored in the memory, and calling data stored in the memory. The memory can mainly include a program storage area and a data storage area, wherein the program storage area can store an operating system, at least one application program required by a function (such as a sound playing function, an image playing function, etc.), and the like; and the data storage area can store data created according to the use of the mobile phone (such as audio data, a phone book, etc.), and the like. In addition, the memory can include a high-speed random access memory, and can also include a nonvolatile memory, such as a hard disk, a memory, a plug-in hard disk, a smart media card (SMC), a secure digital (SD) card, a flash card, at least one disk storage device, a flash memory device, or other volatile solid-state memory devices.

[0154] The modules / units integrated in the electronic device can be stored in a computer readable storage medium if they are realized in the form of software function units and sold or used as independent products. Based on this understanding, all or part of the processes in the above-mentioned embodiment methods can also be completed by a computer program instructing related hardware, and the computer program can be stored in a computer readable storage medium. The computer program can implement the steps of each method embodiment when executed by a processor. The computer program includes computer program code, which can be in the form of source code, object code, executable files, or some intermediate forms, etc. The computer readable medium can include any entity or system that can carry the computer program code, recording medium, U disk, mobile hard disk, magnetic disk, optical disk, computer memory, read-only memory (ROM), random access memory (RAM), electrical carrier signal, telecommunication signal, and software distribution medium, etc. It should be noted that the contents included in the computer readable medium can be appropriately increased or decreased according to the requirements of legislation and patent practice in the jurisdiction, for example, in some jurisdictions, according to legislation and patent practice, the computer readable medium does not include electrical carrier signals and telecommunication signals.

[0155] It should be noted that the system embodiments described above are merely illustrative, wherein the units described as separate components can or can not be physically separated, and the components displayed as units can or can not be physical units, i.e. can be located in one place or distributed to multiple network units. Part or all of the modules can be selected to achieve the purpose of the embodiment according to actual needs. In addition, the connection relationship between the modules in the system embodiment provided by the present application indicates that there is a communication connection between them, which can be implemented as one or more communication buses or signal lines. Those skilled in the art can understand and implement it without creative labor.

[0156] The above specific embodiments further illustrate the purpose, technical scheme and beneficial effects of the present application. It should be understood that the above description is only a specific embodiment of the present application and is not intended to limit the protection scope of the present application. It is particularly pointed out that any modification, equivalent replacement, improvement, etc. made by those skilled in the art within the spirit and principles of the present application shall be included in the protection scope of the present application.

Claims

1. A method for mitigating line losses in transformer substations during seasonal power consumption, characterized in that, include: Real-time load data and historical seasonal electricity consumption records of the transformer area are obtained, and time-series change analysis and feature annotation are performed to obtain a load distribution feature map; Based on the load distribution feature map, node interaction and structural feature fusion processing are performed to obtain the load-structure coupling relationship matrix; Based on the load-structure coupling relationship matrix, risk areas are identified and screened to obtain a list of high-loss risk distributions; Based on the high-loss risk distribution list, path bottleneck analysis and optimization are performed to obtain a path adjustment plan; Based on the path adjustment scheme, a load-bearing pressure risk assessment is performed to determine whether an equipment fault warning is triggered. If triggered, the matrix is ​​dynamically optimized to obtain an updated risk control matrix. Based on the updated risk control matrix and the real-time load data, the risk areas are reassessed to obtain a refined line loss control scheme; The step of performing node interaction and structural feature fusion processing based on the load distribution feature map to obtain the load-structure coupling relationship matrix includes: Based on the load distribution characteristic map, impedance characteristic analysis is performed to obtain the line impedance characteristic set; Based on the line impedance characteristic set, a benchmark power flow model is constructed and solved, interactive data is extracted, and power flow calculation is applied for iterative solution to obtain the average voltage amplitude, average active power and average reactive power combined into load interactive data, and the load interactive data and line impedance characteristic set are obtained. Based on the load interaction data and the line impedance feature set, a fusion process is performed to obtain the fused load and structure association dataset; Based on the load and structure association dataset, a matrix is ​​generated to obtain the load-structure coupling relationship matrix.

2. The method for managing line losses in seasonal power consumption areas according to claim 1, characterized in that, The process of acquiring real-time load data and historical seasonal electricity consumption records for the transformer substation, performing time-series change analysis and feature annotation, and obtaining a load distribution feature map includes: Real-time load data and historical seasonal electricity consumption records of the distribution area are collected from smart meters; Based on the real-time load data and the historical seasonal electricity consumption records, noise reduction processing is performed to obtain cleaned load time-series data; Based on the cleaned load time series data, a connection relationship graph is constructed to obtain a preliminary connection relationship map; Based on the preliminary connection relationship map, load weight calculation and feature annotation are performed to obtain the load distribution feature map.

3. The method for mitigating line losses in seasonal power consumption areas according to claim 1, characterized in that, The process of identifying and filtering risk areas based on the load-structure coupling matrix yields a high-loss risk distribution list, including: Based on the load-structure coupling relationship matrix, the correlation strength is compared, and nodes with a correlation strength greater than a preset threshold are identified as nodes to be analyzed, thus obtaining a set of nodes to be analyzed. The set of nodes to be analyzed is compared with a preset risk feature library to determine risk conditions. Nodes that meet the conditions are selected as high-loss risk nodes, and a list of high-loss risk nodes is generated. Based on the list of high-loss risk nodes, multi-dimensional feature extraction and integration are performed to obtain a high-loss risk distribution list.

4. The method for managing line losses in seasonal power consumption areas according to claim 1, characterized in that, The step of performing path bottleneck analysis and optimization based on the high-loss risk distribution list to obtain a path adjustment scheme includes: Based on the high-loss risk distribution list, time-series data extraction and constraint characteristic analysis are performed to obtain the load fluctuation correlation set. Based on the load fluctuation correlation set, path bottlenecks are assessed and marked to obtain a list of key paths; Based on the list of key paths, constraint data is acquired and feasibility is assessed to obtain priority ranking results; Based on the priority ranking results, a path scheme is generated, and a path adjustment scheme is determined.

5. The method for managing line losses in seasonal power consumption areas according to claim 1, characterized in that, The step involves assessing the load-bearing pressure risk based on the path adjustment scheme, determining whether an equipment fault warning is triggered, and if so, performing dynamic matrix optimization to obtain an updated risk control matrix, including: Based on the path adjustment scheme, overload signal acquisition is performed to obtain a classified signal dataset; Based on the classified signal dataset, the load-bearing pressure index is calculated to obtain the pressure index set. The pressure index set is compared with the preset overload warning threshold. If it exceeds the overload warning threshold, it is determined to be a triggered equipment fault warning signal. Based on the triggered equipment fault warning signal, parameter deviation is located to obtain the parameter range that needs to be adjusted; Based on the parameter range that needs to be adjusted, matrix optimization and iterative calculations are performed to obtain the updated risk control matrix.

6. The method for managing line losses in seasonal power consumption areas according to claim 1, characterized in that, The step of reassessing risk areas based on the updated risk control matrix and the real-time load data to obtain a refined line loss control scheme includes: Based on the real-time load data and the updated risk control matrix, a synchronous comparison and error correction are performed to obtain the adjusted data matching accuracy and synchronous risk control matrix; Based on the data matching accuracy and the synchronous risk control matrix, regional division adjustments are made to determine the potential high-loss area range; Based on the range of the potential high-loss areas, the risk areas are reassessed to obtain updated risk area division results; Based on the updated risk area division results, a control scheme is generated to obtain a refined line loss control scheme.

7. A line loss management system for seasonal electricity consumption in transformer substations, used to implement the line loss management method for seasonal electricity consumption in transformer substations as described in any one of claims 1-6, characterized in that, include: The map construction module is used to acquire real-time load data and historical seasonal electricity consumption records of the distribution area, and to perform time-series change analysis and feature annotation to obtain a load distribution feature map. The coupling relationship analysis module is used to perform node interaction and structural feature fusion processing based on the load distribution feature map to obtain the load-structure coupling relationship matrix; The risk area identification module is used to identify and filter risk areas based on the load-structure coupling relationship matrix to obtain a list of high-loss risk distributions. The path optimization module is used to perform path bottleneck analysis and optimization based on the high-loss risk distribution list to obtain a path adjustment scheme. The assessment and optimization module is used to assess the load pressure risk based on the path adjustment scheme, determine whether to trigger equipment failure warning, and if so, perform matrix dynamic optimization to obtain an updated risk control matrix. The control scheme generation module is used to reassess the risk area based on the updated risk control matrix and the real-time load data to obtain a refined line loss control scheme.

Images