Distributed load early warning management method and system based on power grid

Abstract

The invention discloses a distributed load early warning management method and system based on a power grid, relates to the technical field of power system automation, and solves the problem that the prior art neglects the priority state of each region during early warning, so that the early warning efficiency is high when a plurality of regions are subjected to data analysis at the same time. The technical problem of low early warning efficiency caused by the fact that an area in urgent need of early warning is not analyzed timely and effectively is solved. A regional analysis priority is generated according to regional power grid data; constructing a regional power grid topological graph; generating a regional prediction load according to the regional analysis priority and the regional power grid topological graph, and generating equipment adjustment parameters according to the regional prediction load; according to the method, the alarm signal is generated according to the regional prediction load, and the power grid data of each region is analyzed from multiple angles, so that important regions can be preferentially concerned during load early warning, and meanwhile, corresponding equipment parameters are dynamically adjusted according to the load prediction of each region, so that the data accuracy is improved during high-load early warning. And the early warning efficiency and accuracy are improved.

Description

Technical Field

[0001] The present application relates to the field of power system automation technology, and specifically to a distributed load early warning management method and system based on a power grid. Background Art

[0002] With rapid socioeconomic development and the continuous improvement of people's living standards, electricity demand continues to grow, and grid loads are exhibiting new characteristics such as increased peak-to-valley variations, increased volatility, and increased risk of local overloads. Traditional grid operation and management models face significant challenges, particularly given the frequent occurrence of extreme weather, the large-scale integration of new energy sources, and the rapid development of new loads such as electric vehicles. Accurately predicting load trends and promptly identifying potential operational risks have become key issues in ensuring the safe and stable operation of the power grid.

[0003] Existing technologies for power grid load warning ignore the priority status of each area when issuing warnings. As a result, when data analysis is performed on several areas at the same time, areas that urgently need warnings fail to be analyzed in a timely and effective manner, resulting in low warning efficiency. Therefore, further improvements are still needed for power grid load warnings. Summary of the Invention

[0004] The present application aims to solve at least one of the technical problems existing in the prior art; to this end, the present application proposes a distributed load early warning management method and system based on the power grid, which is used to solve the technical problem that the prior art ignores the priority status of each area when making an early warning, so that when there are several areas performing data analysis at the same time, the areas that urgently need early warning cannot be analyzed in a timely and effective manner, resulting in low early warning efficiency.

[0005] To achieve the above objectives, the first aspect of the present application provides a distributed load early warning management method based on a power grid, comprising:

[0006] Acquire regional data; the regional data includes a regional ID and its corresponding regional power grid data and regional environmental data;

[0007] Generating a regional analysis priority based on regional power grid data; the regional analysis priority refers to the priority of analyzing the regional data;

[0008] Construct regional power grid topology;

[0009] Generate regional forecast load based on regional analysis priorities and regional power grid topology. Regional forecast load refers to the load forecast value of a region in the future time period.

[0010] Generate equipment adjustment parameters based on regional forecast load; and assign equipment adjustment parameters to their corresponding equipment parameters; equipment adjustment parameters refer to parameter values ​​of related equipment adjusted according to load forecast values;

[0011] Generate alarm signals based on regional forecast load.

[0012] Through the above steps, this application conducts a comprehensive analysis of the power grid data of each region from multiple dimensions, so that key areas can be prioritized when conducting load warnings; at the same time, based on the load forecast results of each region, the relevant equipment parameters are dynamically adjusted, further improving the data accuracy of the warning under high load conditions, and effectively enhancing the efficiency and accuracy of the warning method.

[0013] Furthermore, generating regional analysis priorities based on regional power grid data includes:

[0014] Acquiring regional power grid data; the regional power grid data includes several power grid parameters;

[0015] Extract data integrity, accuracy, and timeliness corresponding to regional power grid data;

[0016] The regional data quality QSZ is calculated by weighted fusion of data integrity, data accuracy and data timeliness;

[0017] Extract load rate, voltage deviation and fault frequency corresponding to regional power grid data;

[0018] The regional anomaly score QYP is calculated by weighted fusion of load factor, voltage deviation and fault frequency;

[0019] The regional analysis priority is calculated using the formula that satisfies:

[0020] QFY=QSZ α1 ×QYP α2 ; Among them, α1 and α2 represent the adjustment index coefficients respectively, and both α1 and α2 are greater than 0.

[0021] Furthermore, the adjustment index coefficient is generated by a coefficient generation model, including:

[0022] Obtain the average regional data quality, regional data quality standard deviation, regional anomaly score, and regional false alarm rate corresponding to the coefficient generation model and regional ID;

[0023] Integrate the average regional data quality, regional data quality standard deviation, regional anomaly score and regional false alarm rate into regional analysis data;

[0024] Input the regional analysis data corresponding to the regional ID into the coefficient generation model to obtain a number of adjustment index coefficients corresponding to the regional ID;

[0025] The coefficient generation model is constructed through an artificial intelligence model, including:

[0026] Obtain several historical regional analysis data and historical adjustment index coefficients corresponding to the regional ID;

[0027] Divide the historical regional analysis data and historical adjustment index coefficients corresponding to the regional ID into training data, verification data, and test data; and perform data preprocessing on the training data, verification data, and test data to obtain a training set, a verification set, and a test set;

[0028] Select an artificial intelligence model as the base model;

[0029] Train the basic model using the training set, and adjust the learning rate and other hyperparameters on the validation set to obtain the pre-trained model;

[0030] By verifying the pre-trained model on the test set, we finally obtain a coefficient generation model whose input is the regional analysis data corresponding to the regional ID and whose output is a number of adjustment index coefficients corresponding to the regional ID.

[0031] Furthermore, the regional power grid topology map is constructed in the following manner, including:

[0032] Obtain regional power grid data and regional environmental data;

[0033] Defining a grid node based on grid devices in regional grid data; the grid node includes several node attributes;

[0034] Defining edge connections based on device connection relationships in regional power grid data; the edge connections include several connection attributes;

[0035] Based on regional environmental data, it is mapped to the power grid nodes as node attributes;

[0036] Substitute the regional power grid data into several power grid nodes and several edge connections to obtain the regional power grid topology graph.

[0037] This application constructs a regional power grid topology map corresponding to each regional ID and maps the regional environmental data to the regional power grid topology map, so that the regional power grid topology map has stronger information expression capabilities, and can subsequently more accurately estimate the regional load forecast corresponding to each regional ID in conjunction with time series data, thereby improving the accuracy and efficiency of load warning.

[0038] Furthermore, generating a regional forecast load according to the regional analysis priority and the regional power grid topology map includes:

[0039] Obtain regional analysis priorities, regional power grid topology maps, and regional power grid data;

[0040] Sort the regional IDs in descending order according to the regional analysis priority to obtain the analysis sequence;

[0041] Select the region IDs in the analysis sequence in turn;

[0042] Extracting several influencing parameters from the regional power grid data of the regional ID within the sliding time window; the influencing parameters refer to power grid data that can affect future power load forecasting;

[0043] Integrate several influencing parameters within the sliding time window and the regional environmental data corresponding to the sliding time window into a prediction sequence;

[0044] The prediction sequence corresponding to the regional ID and the regional power grid topology map are input into the load forecasting model to obtain the regional forecast load corresponding to the regional ID; the load forecasting model is constructed through an artificial intelligence model.

[0045] This application sorts the regional analysis priorities corresponding to each regional ID, and selects regions with high regional analysis priorities in turn for regional load forecasting, so that load warnings can be carried out in a timely and effective manner according to the actual situation of each region, thereby improving the efficiency of power grid load warnings.

[0046] Furthermore, the load forecasting model is constructed by an artificial intelligence model, including:

[0047] Obtain several historical forecast sequences corresponding to the regional ID, historical regional power grid topology maps, and historical regional forecast loads;

[0048] Divide several historical prediction sequences, historical regional power grid topology maps, and historical regional predicted loads corresponding to regional IDs into training data, verification data, and test data; and perform data preprocessing on the training data, verification data, and test data to obtain a training set, a verification set, and a test set;

[0049] Select an artificial intelligence model as the base model;

[0050] Train the basic model using the training set, and adjust the learning rate and other hyperparameters on the validation set to obtain the pre-trained model;

[0051] By verifying the pre-trained model on the test set, we finally obtain a load forecasting model whose input is the prediction sequence corresponding to the regional ID and the regional power grid topology map, and whose output is the regional predicted load corresponding to the regional ID.

[0052] Furthermore, the generating of equipment adjustment parameters according to regional forecast load includes:

[0053] Obtain regional forecast load;

[0054] When the regional forecast load is greater than or equal to the corresponding regional load threshold, several equipment adjustment parameters in the next time period are set to the maximum values ​​corresponding to the equipment parameters;

[0055] Otherwise, calculate the STC of several equipment adjustment parameters in the next time period by the calculation formula t+1,i , the calculation formula satisfies the following formula:

[0056] Where t represents the time period number, k represents the sensitivity index, k>0; STC t+1,i Indicates the adjustment parameters of several devices in the next time period, i indicates the number of the adjusted device; STC i,max and STC i,min They represent the maximum and minimum values ​​of the device adjustment parameters of the i-th adjustment device respectively; QJF th and QF min They are respectively expressed as emergency load threshold and minimum load benchmark; QYF t+1 Represents the regional forecast load for the next time period.

[0057] This application analyzes the relationship between the regional predicted load and the equipment adjustment parameters of each area, and dynamically adjusts the equipment parameter value corresponding to the area when there is a large change in the regional predicted load. This can improve the accuracy of the power grid data corresponding to the area with high regional predicted load and improve the accuracy of load warning.

[0058] Furthermore, generating an alarm signal according to the regional predicted load includes:

[0059] Get the regional forecast load corresponding to several regional IDs;

[0060] Determine whether the regional predicted load of the region ID is greater than the regional load threshold corresponding to the region ID;

[0061] If yes, an area ID overload alarm signal is generated;

[0062] If not, determine whether the regional predicted load of the region ID is greater than the D times regional load threshold corresponding to the region ID; where D is the proportional coefficient, D∈(0,1);

[0063] If yes, generate a warning signal indicating an overload risk in the regional ID; if no, do nothing.

[0064] Furthermore, the regional load threshold is generated in the following manner, including:

[0065] Get the historical regional load base value QFJ corresponding to the regional ID in the next time period;

[0066] Get the regional forecast load trend QYFQ corresponding to the regional ID in the next time period;

[0067] Obtaining regional environmental data corresponding to the regional ID in the next time period; the regional environmental data includes several environmental parameters;

[0068] The regional environmental trend QHQ corresponding to the regional ID in the next time period is calculated by the calculation formula; the calculation formula satisfies the following formula:

[0069] Where t represents the number of the time period, j represents the number of the environmental parameter, and the total number of environmental parameters is J. t+1,j It is expressed as the specific value of the jth environmental parameter in the next time period; HC base,j Expressed as the historical mean value of the jth environmental parameter; HCQ j Expressed as the weight of the jth environmental parameter;

[0070] According to the nonlinear relationship between the regional load threshold and the historical regional load base value QFJ, the regional forecast load trend QYFQ, the regional environmental trend QHQ, respectively, a regional load threshold calculation function QFYJF(QFJ, QYFQ, QHQ) is constructed.

[0071] The regional load threshold calculation function satisfies the following calculation formula:

[0072] QFJ t+1,n (QFJ t+1,n ,QYFQ t+1,n ,QHQ t+1,n )=QFJ t+1,n ×(1+β1×QYFQ t+1，n +β2×QHQ t+1，n ); where n represents the number corresponding to the area ID, β1 and β2 represent the load trend influence coefficient and the environment trend influence coefficient, respectively, β1 and β2∈(0,1);

[0073] The historical regional load base value, regional forecast load trend and regional environmental trend corresponding to the regional ID are substituted into the regional load calculation function to obtain the regional load threshold corresponding to the regional ID.

[0074] Another aspect of the present invention provides a distributed load early warning management system based on a power grid, comprising: a data acquisition module, a data analysis module and an early warning module; the data acquisition module is connected to the data analysis module; the data analysis module is connected to the early warning module;

[0075] The data acquisition module acquires regional data through a data acquisition device; the regional data includes a regional ID and its corresponding regional power grid data and regional environmental data;

[0076] The data analysis module generates regional analysis priorities based on regional power grid data; constructs a regional power grid topology map; generates regional forecast loads based on the regional analysis priorities and the regional power grid topology map; generates equipment adjustment parameters based on the regional forecast loads, and assigns the equipment adjustment parameters to their corresponding equipment parameters; and generates alarm signals based on the regional forecast loads;

[0077] The early warning module: provides prompts according to the alarm signal.

[0078] Compared with the prior art, the present invention has the following advantages:

[0079] 1. This application generates regional analysis priorities based on regional power grid data; constructs a regional power grid topology; generates regional forecast loads based on the regional analysis priorities and the regional power grid topology; generates equipment adjustment parameters based on the regional forecast loads; and assigns the equipment adjustment parameters to their corresponding equipment parameters; and generates alarm signals based on the regional forecast loads. This application analyzes the power grid data of each region from multiple perspectives, thereby giving priority to important areas when issuing load warnings. At the same time, it dynamically adjusts the corresponding equipment parameters based on the load forecast for each region, thereby improving data accuracy during high-load warnings. This improves the efficiency and accuracy of warnings.

[0080] 2. This application analyzes the power grid data collected from various regions, analyzes the power grid data from multiple angles and quantifies the power grid data. The quantified indicators are then combined with other relevant data to generate an adjustment index coefficient through a pre-trained coefficient generation model to calculate the regional analysis priority of each region. This ensures that when calculating the regional priority, it can be based on the actual power grid data in each region, thereby improving the calculation accuracy and efficiency of the analysis priority of each region and providing strong data support for subsequent load warnings.

[0081] 3. This application dynamically adjusts the regional load threshold corresponding to each regional ID by analyzing the historical power grid data and current power grid data corresponding to each region in different time periods, so that the regional load threshold corresponding to each region in different time periods does not remain consistent for a long time, reducing the occurrence of false alarms of load warnings and improving the accuracy of power grid load warnings. BRIEF DESCRIPTION OF THE DRAWINGS

[0082] In order to more clearly illustrate the embodiments of the present application or the technical solutions in the prior art, the following briefly introduces the drawings required for use in the embodiments or the description of the prior art. Obviously, the drawings described below are only some embodiments of the present application. For ordinary technicians in this field, other drawings can be obtained based on these drawings without any creative work.

[0083] Figure 1This is a flow chart of a distributed load early warning management method based on a power grid in this application;

[0084] Figure 2 This is a schematic diagram of the principle of a distributed load early warning management system based on the power grid in this application. DETAILED DESCRIPTION

[0085] The following will clearly and completely describe the technical solutions of this application in conjunction with the embodiments. Obviously, the embodiments described are only a part of the embodiments of this application, not all of them. Based on the embodiments in this application, all other embodiments obtained by ordinary technicians in this field without making creative efforts are within the scope of protection of this application.

[0086] See also Figure 1 The first embodiment of the present application provides a distributed load early warning management method based on a power grid, comprising:

[0087] Obtain regional data; regional data includes regional ID and its corresponding regional power grid data and regional environmental data;

[0088] Generate regional analysis priorities based on regional power grid data; regional analysis priorities refer to the priority of analyzing regional data;

[0089] Construct regional power grid topology;

[0090] Generate regional forecast load based on regional analysis priorities and regional power grid topology. Regional forecast load refers to the load forecast value of a region in the future time period.

[0091] Generate equipment adjustment parameters based on regional forecast load; and assign equipment adjustment parameters to their corresponding equipment parameters; equipment adjustment parameters refer to parameter values ​​of related equipment adjusted according to load forecast values;

[0092] Generate alarm signals based on regional forecast load.

[0093] In this embodiment, generating regional analysis priorities based on regional power grid data includes:

[0094] Acquire regional power grid data; regional power grid data includes several power grid parameters;

[0095] Extract the data integrity, data accuracy, and data timeliness corresponding to the regional power grid data; data integrity measures the missing ratio of data fields, data accuracy measures the noise level of the data, and data timeliness measures the delay time of the data; in this embodiment, data integrity, data accuracy, and data timeliness are all controlled in the range of 0-1;

[0096] The regional data quality QSZ is calculated by weighted fusion of data integrity, data accuracy and data timeliness;

[0097] The calculation formula for regional data quality satisfies:

[0098] QSZ = γ1 × SW + γ2 × SZ + γ3 × SS; where γ1, γ2, and γ3 represent the weight coefficients corresponding to data integrity SW, data accuracy SZ, and data timeliness SS. γ1, γ2, and γ3 ∈ (0, 1). The specific values ​​are set based on experience. In this embodiment, γ1, γ2, and γ3 are set to 0.3, 0.4, and 0.3, respectively. The larger the weight coefficient, the greater the influence of the corresponding data in calculating regional data quality.

[0099] Extract the load rate, voltage deviation, and fault frequency corresponding to the regional power grid data; the load rate is expressed as the ratio of the current regional load to the rated capacity. The voltage deviation quantifies the voltage stability in the region and is mapped to the range of 0-1 using the historical voltage deviation. The fault frequency refers to the frequency of faults occurring over a period of time. The current fault frequency is mapped to the range of 0-1 using the historical fault frequency. For example, if the fault frequency in a certain area is 3 in the past 24 hours and the historical maximum fault frequency is 5, the corresponding fault frequency mapped to the range of 0-1 is 0.6.

[0100] The regional anomaly score QYP is calculated by weighted fusion of load factor, voltage deviation and fault frequency;

[0101] The calculation formula for regional anomaly score satisfies:

[0102] QSZ = γ4 × FL + γ5 × DP + γ6 × GP; where γ4, γ5, and γ6 represent the weight coefficients corresponding to the load factor FL, voltage deviation DP, and fault frequency GP. γ4, γ5, and γ6 ∈ (0, 1). The specific values ​​are set based on experience. In this embodiment, γ4, γ5, and γ6 are set to 0.6, 0.3, and 0.1, respectively. The larger the weight coefficient, the greater the influence of the corresponding data in calculating the regional anomaly score.

[0103] The regional analysis priority is calculated using the formula that satisfies:

[0104] QFY=QSZ α1 ×QYP α2 ; Among them, α1 and α2 represent the adjustment index coefficients, respectively, and both α1 and α2 are greater than 0; the regional analysis priority increases with the increase of regional data quality and regional anomaly score.

[0105] This embodiment performs multi-angle analysis and quantitative processing on the power grid data collected from each region, combines it with other relevant data, and inputs it into a pre-trained coefficient generation model to calculate the adjustment index coefficient reflecting the characteristics of each region, and then determines the analysis priority of each region. This ensures that the priority calculation is fully based on the actual power grid operation conditions in the region, effectively improves the accuracy and efficiency of regional analysis, and provides strong data support for the subsequent load warning.

[0106] The adjustment index coefficient in this embodiment is generated by a coefficient generation model, including:

[0107] Obtain the average regional data quality, regional data quality standard deviation, regional anomaly score, and regional false alarm rate corresponding to the coefficient generation model and regional ID; the average regional data quality refers to the mean of the historical regional data quality corresponding to the regional ID over the past period of time, and the regional data quality standard deviation measures the fluctuation of data quality within the regional ID over the past period of time; the regional false alarm rate refers to the proportion of false alarms over the past period of time; the specific value of the past period of time is set based on experience. In this embodiment, the past period of time is set to the past 24 hours;

[0108] Integrate the average regional data quality, regional data quality standard deviation, regional anomaly score and regional false alarm rate into regional analysis data;

[0109] Input the regional analysis data corresponding to the regional ID into the coefficient generation model to obtain a number of adjustment index coefficients corresponding to the regional ID;

[0110] The coefficient generation model is constructed through an artificial intelligence model, including:

[0111] Obtain several historical regional analysis data and historical adjustment index coefficients corresponding to the regional ID;

[0112] The historical regional analysis data and historical adjustment index coefficients corresponding to the regional ID are divided into training data, verification data, and test data; the training data, verification data, and test data are preprocessed to obtain the training set, verification set, and test set; the ratio between the training set, test set, and verification set is 7:2:1;

[0113] Select an artificial intelligence model as the basic model; artificial intelligence models include neural network models, etc.

[0114] Train the basic model using the training set, and adjust the learning rate and other hyperparameters on the validation set to obtain the pre-trained model;

[0115] By verifying the pre-trained model on the test set, we finally obtain a coefficient generation model whose input is the regional analysis data corresponding to the regional ID and whose output is a number of adjustment index coefficients corresponding to the regional ID.

[0116] The regional power grid topology diagram in this embodiment is constructed in the following manner, including:

[0117] Obtain regional power grid data and regional environmental data;

[0118] Grid nodes are defined based on grid equipment in regional grid data. Grid nodes include several node attributes. Grid nodes include substations, transformers, and user nodes. Node attributes include node ID, node type, and node coordinates.

[0119] Define edge connections based on device connection relationships in regional power grid data; edge connections include several connection attributes; edge connections include connection relationships; connection attributes include connection type and connection length, etc.;

[0120] Based on the regional environmental data, it is mapped to the power grid node as a node attribute; that is, the regional environmental data corresponding to each power grid node is used as one of the node attributes of the power grid node;

[0121] Substitute the regional power grid data into several power grid nodes and several edge connections to obtain the regional power grid topology graph.

[0122] This embodiment enhances the information expression capability of the power grid topology by constructing a corresponding power grid topology for each regional ID and mapping environmental data to these topology maps. This method enables more accurate prediction of the load conditions corresponding to each regional ID when combined with time series data for analysis, thereby improving the accuracy and efficiency of load warnings. This not only improves the understanding and analysis capabilities of the power grid operating status, but also provides a solid foundation for more scientific and reasonable load management.

[0123] In this embodiment, generating a regional forecast load based on the regional analysis priority and the regional power grid topology map includes:

[0124] Obtain regional analysis priorities, regional power grid topology maps, and regional power grid data;

[0125] Sort the regional IDs in descending order according to the regional analysis priority to obtain the analysis sequence;

[0126] Select the region IDs in the analysis sequence in turn;

[0127] Extract several influencing parameters from the regional power grid data within the sliding time window of the region ID. Influencing parameters refer to power grid data that can affect future power load forecasts. The sliding time window is the time window size used to obtain historical data. In this embodiment, the sliding time window is set to 72 hours. Several influencing parameters include voltage, current, equipment status, and feeder load.

[0128] Integrate several influencing parameters within the sliding time window and the regional environmental data corresponding to the sliding time window into a prediction sequence; that is, integrate several influencing parameters corresponding to a time point and the regional environmental data corresponding to the time point into the analysis data corresponding to the time point, and then integrate several time points within the sliding time window into a prediction sequence in the order of the time points; the regional environmental data includes temperature, humidity, wind speed, etc.

[0129] The prediction sequence corresponding to the regional ID and the regional power grid topology map are input into the load forecasting model to obtain the regional forecast load corresponding to the regional ID; the load forecasting model is constructed through an artificial intelligence model.

[0130] The load forecasting model in this embodiment is constructed using an artificial intelligence model, including:

[0131] Obtain several historical forecast sequences corresponding to the regional ID, historical regional power grid topology maps, and historical regional forecast loads;

[0132] Several historical prediction sequences, historical regional power grid topology maps, and historical regional forecast loads corresponding to regional IDs are divided into training data, validation data, and test data; data preprocessing is performed on the training data, validation data, and test data to obtain training sets, validation sets, and test sets; the ratio of the training set, test set, and validation set is 7:2:1;

[0133] Select an artificial intelligence model as the basic model; the artificial intelligence model used in this embodiment includes a convolutional neural network model and an LSTM model; the convolutional neural network model is used to process the regional power grid topology map, and the LSTM model is used to process the prediction sequence;

[0134] Train the basic model using the training set, and adjust the learning rate and other hyperparameters on the validation set to obtain the pre-trained model;

[0135] By verifying the pre-trained model on the test set, we finally obtain a load forecasting model whose input is the prediction sequence corresponding to the regional ID and the regional power grid topology map, and whose output is the regional predicted load corresponding to the regional ID.

[0136] In this embodiment, the device adjustment parameters are generated based on the regional forecast load, including:

[0137] Obtain regional forecast load;

[0138] When the regional predicted load is greater than or equal to the corresponding regional load threshold, several device adjustment parameters in the next time period are set to the corresponding maximum values ​​of the device parameters. The specific value of the time period is set based on experience. In this embodiment, the regional predicted load is predicted for the next day, so the time period is 1 day.

[0139] Otherwise, calculate the STC of several equipment adjustment parameters in the next time period by the calculation formula t+1,i , the calculation formula satisfies the following formula:

[0140] Wherein, t represents the number of the time period, k represents the sensitivity index, k>0, and the specific value is set according to experience. In this embodiment, k is set to 2; STC t+1,i Indicates the adjustment parameters of several devices in the next time period, i indicates the number of the adjusted device; STC i,max and STC i,min They represent the maximum and minimum values ​​of the device adjustment parameters of the i-th adjustment device respectively; QJF th and QF min They are respectively expressed as emergency load threshold and minimum load benchmark, and the specific values ​​are set according to experience. In this embodiment, QF min Set to 100kW, QJF th Set to 90% of the maximum load reference, the maximum load reference in this embodiment is set to 1000kW, that is, QJF th Set to 900kW; QYF t+1 It represents the regional predicted load for the next time period. The adjustment devices in this embodiment can be divided into three categories, including SCADA devices, smart meter devices, and meteorological API devices. The device adjustment parameter corresponding to the adjustment device is the acquisition frequency. The SCADA device is used to collect bus voltage, current, power factor, etc. The smart meter device is used to collect user load curves and harmonic content in the region. The meteorological API device is used to obtain environmental data. The higher the regional predicted load, the higher the acquisition frequency in the next time period in the region will be, so as to ensure the data accuracy in the next time period.

[0141] This embodiment deeply analyzes the relationship between the predicted load of each area and the equipment adjustment parameters. When a significant change in the predicted load of a certain area is detected, the corresponding equipment parameters of the area are dynamically adjusted. This can improve the accuracy of power grid data in areas with higher predicted loads, thereby enhancing the accuracy of load warnings. It not only optimizes the operating efficiency of the power grid, but also ensures that a more timely and accurate response can be made when there are significant changes in the load.

[0142] In this embodiment, generating an alarm signal based on regional predicted load includes:

[0143] Get the regional forecast load corresponding to several regional IDs;

[0144] Determine whether the regional predicted load of the region ID is greater than the regional load threshold corresponding to the region ID;

[0145] If yes, an area ID overload alarm signal is generated;

[0146] If no, determine whether the regional predicted load of the region ID is greater than D times the regional load threshold corresponding to the region ID; where D is the proportional coefficient, D∈(0,1), and the specific value is set based on experience. In this embodiment, D is set to 0.7;

[0147] If yes, generate a warning signal indicating an overload risk in the regional ID; if no, do nothing.

[0148] In this embodiment, the regional load threshold is generated by:

[0149] Obtain the historical regional load base value QFJ corresponding to the region ID in the next time period. The historical regional load base value is represented by the baseline value of the load threshold corresponding to the region ID in the historical period. In this embodiment, the historical regional load base value corresponding to a region in the region ID is 650kW.

[0150] Obtain the regional forecast load trend QYFQ corresponding to the regional ID in the next time period; the regional forecast load trend is expressed as the trend change of the forecast load in the historical M hours before the forecast time point, QYFQ∈[-1,1]. When QYFQ>0, it indicates that the forecast load is on an upward trend, otherwise it is on a downward trend. M is a positive integer. In this embodiment, M is set to 24 hours.

[0151] Obtain the regional environmental data corresponding to the regional ID in the next time period; the regional environmental data includes several environmental parameters;

[0152] The regional environmental trend QHQ corresponding to the regional ID in the next time period is calculated by the calculation formula; the calculation formula satisfies the following formula:

[0153] Where t represents the number of the time period, j represents the number of the environmental parameter, and the total number of environmental parameters is J. t+1,j It is expressed as the specific value of the jth environmental parameter in the next time period, which can be obtained through weather forecast; HC base,j Expressed as the historical mean value of the jth environmental parameter; HCQ j It is represented as the weight of the jth environmental parameter, and the specific value is set based on experience. In this embodiment, the environmental parameters include temperature, humidity, and wind speed. The weights for temperature, humidity, and wind speed are set to 0.7, 0.2, and 0.1, respectively. The regional environmental trend increases with the increase in the deviation of the environmental parameters from the historical mean value for the same period.

[0154] According to the nonlinear relationship between the regional load threshold and the historical regional load base value QFJ, the regional forecast load trend QYFQ, the regional environmental trend QHQ, respectively, a regional load threshold calculation function QFYJF(QFJ, QYFQ, QHQ) is constructed.

[0155] The regional load threshold calculation function satisfies the following calculation formula:

[0156] QFJ t+1,n (QFJ t+1,n ,QYFQ t+1,n ,QHQ t+1,n )=QFJ t+1,n ×(1+β1×QYFQ t+1，n +β2×QHQ t+1，n ); where n represents the number corresponding to the area ID, β1 and β2 represent the load trend influence coefficient and the environment trend influence coefficient, respectively, β1 and β2∈(0,1); the specific values ​​are set based on experience. In this embodiment, β1 and β2 are set to 0.5 and 0.3 respectively;

[0157] The historical regional load base value, regional forecast load trend and regional environmental trend corresponding to the regional ID are substituted into the regional load calculation function to obtain the regional load threshold corresponding to the regional ID.

[0158] This embodiment analyzes the historical power grid data and current power grid data of each region at different time periods, and dynamically adjusts the load threshold corresponding to each region ID, so that the load threshold of each region can be adaptively adjusted over time according to the actual operating conditions, avoiding the limitations brought by fixed thresholds, effectively reducing false alarms in load warnings, and significantly improving the accuracy and reliability of power grid load warnings.

[0159] See also Figure 2 Another embodiment of the present application provides a distributed load early warning management system based on a power grid, comprising: a data acquisition module, a data analysis module, and an early warning module; the data acquisition module is electrically and / or communicatively connected to the data analysis module; the data analysis module is electrically and / or communicatively connected to the early warning module;

[0160] Data acquisition module: acquires regional data through data acquisition equipment; regional data includes regional ID and its corresponding regional power grid data and regional environmental data; data acquisition equipment includes several sensors, etc.

[0161] Data analysis module: Generates regional analysis priorities based on regional power grid data; constructs a regional power grid topology; generates regional forecast loads based on the regional analysis priorities and the regional power grid topology; generates equipment adjustment parameters based on the regional forecast loads and assigns the equipment adjustment parameters to their corresponding equipment parameters; generates alarm signals based on the regional forecast loads;

[0162] Early warning module: Prompts based on alarm signals, including area ID overload alarm signals and area ID overload risk warning signals.

[0163] Some of the data in the above formula are calculated by removing the dimensions and taking their numerical values. The formula is a formula that is closest to the actual situation obtained by software simulation of a large amount of collected data; the preset parameters and preset thresholds in the formula are set by technical personnel in this field according to actual conditions or obtained through simulation of a large amount of data.

[0164] The working principle of this application is as follows: by acquiring regional data; generating regional analysis priorities based on regional power grid data; constructing a regional power grid topology; generating regional predicted loads based on regional analysis priorities and the regional power grid topology; generating equipment adjustment parameters based on regional predicted loads, and assigning the equipment adjustment parameters to their corresponding equipment parameters; generating alarm signals based on regional predicted loads, analyzing the power grid data of each region from multiple angles, so that when making load warnings, priority can be given to important areas, and at the same time, the corresponding equipment parameters can be dynamically adjusted according to the load forecast of each area, so that the accuracy of the data can be improved during high-load warnings. This improves the efficiency and accuracy of warnings, and avoids the problem that the existing technology ignores the priority status of each area when making warnings, so that when there are several areas that conduct data analysis at the same time, the areas that urgently need warnings fail to conduct timely and effective analysis, resulting in low warning efficiency.

[0165] The above embodiments are only used to illustrate the technical method of the present application and are not intended to limit it. Although the present application has been described in detail with reference to the preferred embodiments, those skilled in the art should understand that the technical method of the present application can be modified or replaced by equivalents without departing from the spirit and scope of the technical method of the present application.

Claims

1. A distributed load early warning management method based on a power grid, characterized in that: include: Acquire regional data; the regional data includes a regional ID and its corresponding regional power grid data and regional environmental data; Generate regional analysis priorities based on regional grid data; The regional analysis priority refers to the priority of analyzing regional data; Construct regional power grid topology; Generate regional forecast load based on regional analysis priorities and regional power grid topology. Regional forecast load refers to the load forecast value of a region in the future time period. Generate equipment adjustment parameters based on regional forecast load; and assign equipment adjustment parameters to their corresponding equipment parameters; equipment adjustment parameters refer to parameter values ​​of related equipment adjusted according to load forecast values; Generate alarm signals based on regional forecast load.

2. A distributed load early warning management method based on a power grid according to claim 1, characterized in that: Generating regional analysis priorities based on regional power grid data includes: Obtain regional power grid data; Extract data integrity, accuracy, and timeliness corresponding to regional power grid data; The regional data quality QSZ is calculated by weighted fusion of data integrity, data accuracy and data timeliness; Extract load rate, voltage deviation and fault frequency corresponding to regional power grid data; The regional anomaly score QYP is calculated by weighted fusion of load factor, voltage deviation and fault frequency; The regional analysis priority is calculated using the formula that satisfies: QFY=QSZ α1 ×QYP α2 ; Among them, α1 and α2 represent the adjustment index coefficients respectively, and both α1 and α2 are greater than 0.

3. A distributed load early warning management method based on a power grid according to claim 2, characterized in that: The adjustment index coefficient is generated by a coefficient generation model, including: Obtain the average regional data quality, regional data quality standard deviation, regional anomaly score, and regional false alarm rate corresponding to the coefficient generation model and regional ID; Integrate the average regional data quality, regional data quality standard deviation, regional anomaly score and regional false alarm rate into regional analysis data; Input the regional analysis data corresponding to the regional ID into the coefficient generation model to obtain a number of adjustment index coefficients corresponding to the regional ID; The coefficient generation model is constructed through an artificial intelligence model, including: Obtain several historical regional analysis data and historical adjustment index coefficients corresponding to the regional ID; Divide the historical regional analysis data and historical adjustment index coefficients corresponding to the regional ID into training data, verification data, and test data; and perform data preprocessing on the training data, verification data, and test data to obtain a training set, a verification set, and a test set; Select an artificial intelligence model as the base model; Train the basic model using the training set, and adjust the learning rate and other hyperparameters on the validation set to obtain the pre-trained model; By verifying the pre-trained model on the test set, we finally obtain a coefficient generation model whose input is the regional analysis data corresponding to the regional ID and output is several adjustment index coefficients corresponding to the regional ID.

4. The distributed load early warning management method based on the power grid according to claim 1, characterized in that: The regional power grid topology map is constructed by the following methods, including: Obtain regional power grid data and regional environmental data; Defining a grid node based on grid devices in regional grid data; the grid node includes several node attributes; Defining edge connections based on device connection relationships in regional power grid data; the edge connections include several connection attributes; Based on regional environmental data, it is mapped to the power grid nodes as node attributes; Substitute the regional power grid data into several power grid nodes and several edge connections to obtain the regional power grid topology graph.

5. The distributed load early warning management method based on the power grid according to claim 1 is characterized in that: Generating regional forecast load according to regional analysis priorities and regional power grid topology diagram includes: Obtain regional analysis priorities, regional power grid topology maps, and regional power grid data; Sort the region IDs in descending order according to the region analysis priority to obtain the analysis sequence; Select the region IDs in the analysis sequence in turn; Extracting several influencing parameters from the regional power grid data of the regional ID within the sliding time window; the influencing parameters refer to power grid data that can affect future power load forecasting; Integrate several influencing parameters within the sliding time window and the regional environmental data corresponding to the sliding time window into a prediction sequence; The prediction sequence corresponding to the regional ID and the regional power grid topology map are input into the load forecasting model to obtain the regional forecast load corresponding to the regional ID; the load forecasting model is constructed through an artificial intelligence model.

6. A distributed load early warning management method based on a power grid according to claim 5, characterized in that: The load forecasting model is constructed using an artificial intelligence model, including: Obtain several historical forecast sequences corresponding to the regional ID, historical regional power grid topology maps, and historical regional forecast loads; Divide several historical prediction sequences, historical regional power grid topology maps, and historical regional predicted loads corresponding to regional IDs into training data, verification data, and test data; and perform data preprocessing on the training data, verification data, and test data to obtain a training set, a verification set, and a test set; Select an artificial intelligence model as the base model; Train the basic model using the training set, and adjust the learning rate and other hyperparameters on the validation set to obtain the pre-trained model; By verifying the pre-trained model on the test set, we finally obtain a load forecasting model whose input is the prediction sequence corresponding to the regional ID and the regional power grid topology map, and whose output is the regional predicted load corresponding to the regional ID.

7. The distributed load early warning management method based on the power grid according to claim 1 is characterized in that: Generating equipment adjustment parameters according to regional forecast load includes: Obtain regional forecast load; When the regional forecast load is greater than or equal to the corresponding regional load threshold, several equipment adjustment parameters in the next time period are set to the maximum values ​​corresponding to the equipment parameters; Otherwise, calculate the STC of several equipment adjustment parameters in the next time period by the calculation formula t+1,i , the calculation formula satisfies the following formula: Where t represents the time period number, k represents the sensitivity index, k>0; STC t+1,i Indicates the adjustment parameters of several devices in the next time period, i indicates the number of the adjusted device; STC i,max and STC i,min They represent the maximum and minimum values ​​of the device adjustment parameters of the i-th adjustment device respectively; QJF th and QF min They are respectively expressed as emergency load threshold and minimum load benchmark; QYF t+1 Represents the regional forecast load for the next time period.

8. The distributed load early warning management method based on the power grid according to claim 1 is characterized in that: Generating an alarm signal according to regional predicted load includes: Get the regional forecast load corresponding to several regional IDs; Determine whether the regional predicted load of the region ID is greater than the regional load threshold corresponding to the region ID; If yes, an area ID overload alarm signal is generated; If not, determine whether the regional predicted load of the region ID is greater than the D times regional load threshold corresponding to the region ID; where D is the proportional coefficient, D∈(0,1); If yes, generate a warning signal indicating an overload risk in the regional ID; if no, do nothing.

9. A distributed load early warning management method based on a power grid according to claim 7 or claim 8, characterized in that: The regional load threshold is generated by: Obtain the historical regional load base value QFJ, regional forecast load trend QYFQ and regional environmental data corresponding to the regional ID in the next time period; the regional environmental data includes several environmental parameters; The regional environmental trend QHQ corresponding to the regional ID in the next time period is calculated by the calculation formula; the calculation formula satisfies the following formula: Where t represents the number of the time period, j represents the number of the environmental parameter, and the total number of environmental parameters is J. t+1,j It is expressed as the specific value of the jth environmental parameter in the next time period; HC base,j Expressed as the historical mean value of the jth environmental parameter; HCQ j Expressed as the weight of the jth environmental parameter; According to the nonlinear relationship between the regional load threshold and the historical regional load base value QFJ, the regional forecast load trend QYFQ, the regional environmental trend QHQ, respectively, a regional load threshold calculation function QFYJF(QFJ, QYFQ, QHQ) is constructed. The historical regional load base value, regional forecast load trend and regional environmental trend corresponding to the regional ID are substituted into the regional load calculation function to obtain the regional load threshold corresponding to the regional ID.

10. A distributed load early warning management system based on a power grid, applied to a distributed load early warning management method based on a power grid according to any one of claims 1 to 9, characterized in that: include: Interconnected data acquisition module and data analysis module; The data acquisition module acquires regional data through a data acquisition device; the regional data includes a regional ID and its corresponding regional power grid data and regional environmental data; The data analysis module generates regional analysis priorities based on regional power grid data; constructs a regional power grid topology map; generates regional predicted loads based on the regional analysis priorities and the regional power grid topology map; generates equipment adjustment parameters based on the regional predicted loads, and assigns the equipment adjustment parameters to their corresponding equipment parameters; and generates an alarm signal based on the regional predicted loads.

Images