Intelligent power grid line loss trend early warning system and layered regulation and control method
By building a smart grid line loss trend warning system and combining neural networks and graph theory topological sorting algorithms, accurate prediction and real-time regulation of grid line losses are achieved, solving the problems of low line loss prediction accuracy and lagging regulation strategies in existing technologies, and improving the efficiency and economy of grid operation.
Patent Information
- Application Number
- CN202510950011.4
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2025-07-10
- Publication Date
- 2025-10-14
AI Technical Summary
The existing smart grid system lacks multi-dimensional data fusion analysis, resulting in low line loss prediction accuracy, lack of dynamic and hierarchical management capabilities in the control strategy, and inability to cope with complex operating scenarios, leading to high line loss rates and economic losses.
Build a smart grid line loss trend warning system. By collecting power grid and meteorological data in real time, combining neural network algorithms to establish nonlinear mapping relationships, construct a prediction model for meteorological indicators and line loss parameters, and combine graph theory topological sorting and particle swarm optimization algorithms for real-time regulation to achieve accurate warning and dynamic regulation.
The line loss trend prediction error was reduced by 30%, the control response time was shortened from minutes to seconds, and the equipment adjustment frequency was reduced by 40%, which balanced the grid operation efficiency and equipment life, and reduced the line loss rate and economic losses.
Smart Images
Figure CN120782067A_ABST
Abstract
Description
TECHNICAL FIELD
[0001] The present application relates to the technical field of power grid monitoring, more particularly to an intelligent power grid line loss trend early warning system and hierarchical regulation method. BACKGROUND
[0002] At present, intelligent power grids are accelerating the evolution towards digital and fine management. As a core indicator for measuring the operational efficiency of power grids, the accurate control of line loss rate is crucial for reducing energy loss and improving power supply reliability. The current intelligent power grid line loss management technology faces multiple challenges: Firstly, the existing system lacks multi-dimensional fusion analysis of power grid operation data and meteorological data, making it difficult to accurately quantify the impact of environmental factors on line loss. Research shows that extreme high-temperature weather can increase line resistance, and the leakage current fluctuation amplitude of insulators under strong wind conditions can reach more than 20%. However, the traditional system only relies on historical data of power grids for line loss calculation, without considering meteorological factors such as temperature, wind speed, and humidity in the dynamic model, resulting in a loss estimation deviation rate of up to 15-20%.
[0003] Secondly, the existing regulation system mostly adopts a static threshold triggering mechanism. In the face of complex operating scenarios such as dynamic changes in power grid topology caused by large-scale access of new energy (such as the reversal of power flow direction due to the switching of distributed power sources), and the severe fluctuations of load during peak and valley periods, it cannot achieve real-time adaptive adjustment. The traditional regulation strategy in areas with high penetration rate of photovoltaic access has a warning response delay of more than 30 minutes, resulting in an increase of 40% in the average duration of line loss abnormal period, directly causing annual economic losses of more than one million yuan.
[0004] This technical bottleneck of "early warning lag and ineffective regulation strategy" is essentially due to the lack of a dynamic and adaptive line loss prediction and regulation model. The result not only leads to a long-term high line loss rate, but also seriously restricts the intelligent upgrading process of power grids, making it difficult to balance the improvement of clean energy consumption capacity and the economic efficiency of power supply. In view of this, we propose an intelligent power grid line loss trend early warning system and hierarchical regulation method. SUMMARY
[0005] The purpose of the present application is to provide an intelligent power grid line loss trend early warning system and hierarchical regulation method to solve the technical problems of low line loss prediction accuracy, lack of dynamic regulation, and insufficient hierarchical management capability in the prior art.
[0006] To solve the above technical problems, the present application provides the following technical solution: an intelligent power grid line loss trend early warning system, comprising: an early warning and prediction mechanism for collecting and analyzing data and predicting line loss trend and early warning level; An execution regulation system is responsible for converting the early warning prediction result into actual regulation action; An intelligent regulation terminal is used for driving and executing the regulation instruction, and comprises a regulation driving unit and a line loss management terminal. The regulation driving unit is used for receiving the early warning information sent by the line loss early warning execution unit, and driving the line loss management terminal to execute line loss management operation. The line loss management terminal acquires the line loss management information of the line loss early warning execution unit, and performs line loss management adjustment on the power grid. The line loss management information of the line loss early warning execution unit is acquired, and the line loss management adjustment is performed on the power grid, specifically including: acquiring the early warning type and grade, determining the equipment adjustment scheme and the line loss management control parameter, generating the regulation information and transmitting it to the power grid, and realizing the line loss management.
[0007] The early warning prediction mechanism is used to collect power grid basic data, power flow data and meteorological historical and forecast data in real time, and a nonlinear mapping relationship between meteorological indexes and line loss parameters is constructed by combining a neural network algorithm, so that the problem of single data dimension and insufficient prediction model precision in the prior art is solved. By meteorological fitting, key factors such as temperature and wind speed are extracted, and a prediction model is generated by linear fitting of grid line loss historical data, so that the line loss trend prediction error is reduced by more than 30%, and the regulation preposition is realized.
[0008] Preferably, the early warning prediction mechanism comprises: A data acquisition unit is used to collect power grid data information and meteorological data information in real time. The power grid data information covers power grid basic data and power grid power flow data, and the meteorological data information includes meteorological basic data and meteorological forecast data. A data analysis unit is used to deeply analyze the power grid basic data and the meteorological basic data to obtain grid line loss parameters. A trend analysis unit obtains grid line loss trend parameters through meteorological fitting and trend analysis. A trend classification early warning prediction unit classifies the grades based on the corresponding relationship between the meteorological indexes and the grid line loss index parameters, and establishes a line loss regulation prediction model. A trend regulation prediction unit establishes the corresponding relationship between the meteorological indexes and the grid line loss index parameters, and the line loss regulation prediction relationship, and provides data support for subsequent regulation.
[0009] Preferably, the power grid basic data and the meteorological basic data are deeply analyzed to obtain grid line loss parameters, and the specific process is as follows: A1: The power grid basic data is counted, historical data and operation data containing historical line loss data are extracted, and line loss calculation data and line loss historical data are counted respectively. A2: statistical meteorological basic data, including meteorological historical data and meteorological running data.
[0010] Preferably, the grid line loss trend parameter is derived by meteorological fitting and trend analysis, and the specific steps are as follows: B1: fitting the meteorological basic data and the prediction data to obtain the key parameter factor affecting the line loss; B1: linear fitting based on the grid line loss historical data and the above-mentioned parameter factor to establish a line loss prediction model.
[0011] Preferably, the line loss control prediction model is established based on the corresponding relationship between the meteorological index and the grid line loss index parameter, and the specific steps include the following steps: C1: analyzing the line loss index parameter, obtaining the grid line loss parameter curve, and setting multiple levels of line loss parameter early warning levels by using clustering algorithm; C2: using similar method to analyze the meteorological index parameter, obtaining the meteorological parameter curve, setting the meteorological parameter early warning level and fluctuation interval, and storing the meteorological parameter interval calibration parameter; C3: establishing a line loss control prediction model, and constructing a grid line loss control prediction graph according to the line loss control prediction relationship.
[0012] Preferably, when establishing the corresponding relationship between the meteorological index and the grid line loss index parameter, a neural network algorithm is used, a multilayer perceptron model is constructed, meteorological index parameters are input, nonlinear transformation of the hidden layer is performed, and corresponding grid line loss index parameter prediction values are output. The model training uses back propagation algorithm to update network weights, and the optimization objective function is the mean square error of the prediction value and the actual value.
[0013] Preferably, the execution control system comprises: The line loss early warning execution unit establishes a network topology structure, and constructs a power grid prediction network topology structure according to the grid line loss control prediction graph and the power grid flow parameter; The trend control execution unit obtains the grid line loss control factor in the grid line loss control prediction graph, and generates the management and control control information.
[0014] Preferably, the network topology structure is established, and the power grid prediction network topology structure is constructed according to the grid line loss control prediction graph and the power grid flow parameter, and the specific steps are as follows: D1: obtaining the line loss management and control interval and performing parameter calibration to obtain the line loss management and control interval parameter, and setting the line loss management and control parameter interval; D2: based on the line loss regulation prediction graph of the grid line loss regulation interval, the line loss interval is constructed, the topological structure modeling is carried out by using the topological sorting algorithm in graph theory, the line loss topological structure graph is obtained, the line loss topological interval graph, the line loss control parameter interval and the power grid power flow parameter are combined to construct the power grid prediction network topology, the topological sorting algorithm ensures the correct dependence relationship of the nodes in the network, and a loop-free topological sequence is outputted; D3: the line loss risk level in the grid line loss regulation prediction graph is acquired, the grid control is executed, and the control information is transmitted to the trend regulation execution unit.
[0015] Preferably, the grid line loss regulation factor in the grid line loss regulation prediction graph is acquired, and the control regulation information is generated, and the specific process is as follows: E1: set the regulation interval and establish the regulation interval graph, which is used in cooperation with the line loss control interval parameter; E2: analyze the parameters affecting the line loss regulation factor, and set the line loss regulation factor interval graph; E3: determine the line control regulation interval based on the grid line loss regulation prediction graph, construct the line control regulation interval graph, combine the line loss control interval graph to establish the line control regulation relationship graph, generate the control regulation information by using the decision tree algorithm, and transmit the control regulation information to the trend regulation prediction unit, so as to realize the regulation of the line loss trend.
[0016] A kind of intelligent power grid line loss hierarchical regulation method is driven based on the line loss control terminal of above-mentioned intelligent regulation terminal, including the following steps: S1: driving regulation driving unit receives early warning information, obtains the power grid prediction network topology constructed based on grid line loss regulation prediction graph and power grid power flow parameter; S2: line loss control terminal obtains early warning type and early warning level; S3: determine equipment adjustment scheme according to early warning level, determine line loss control parameter by grid line loss regulation prediction graph, and optimize parameter by using particle swarm optimization algorithm; S4: generate regulation information according to line control regulation relationship graph and regulation parameter, including adjusting equipment, adjusting time, adjusting parameter and adjusting numerical value, and transmit to power grid; S5: power grid receives regulation information and executes line loss control operation.
[0017] Compared with the prior art, the beneficial effects of the present application are: 1. This invention uses a warning and forecasting mechanism to collect basic grid data, power flow data, and historical and forecast meteorological data in real time. It then uses a neural network algorithm to construct a nonlinear mapping relationship between meteorological indicators and line loss parameters. This solves the existing problems of single-dimensional data and insufficient prediction model accuracy. Key factors such as temperature and wind speed are extracted through meteorological fitting, and then linearly fitted with historical grid line loss data to generate a prediction model. This reduces line loss trend prediction errors by over 30%, achieving "early warning and precise grading" for pre-emptive regulation.
[0018] 2. This invention leverages the line loss warning execution unit of the control system and employs a graph-theoretic topological sorting algorithm to construct a grid prediction network topology. This algorithm combines the line loss control prediction graph with power flow parameters to dynamically update topological node dependencies, further resolving the issue of traditional control strategies lagging behind the grid's actual operating status. When new distributed power sources (DGs) cause topological changes in the grid, the system rapidly generates a loop-free control sequence through topological sorting, reducing control command response time from minutes to seconds and ensuring real-time alignment of control actions with line loss trends.
[0019] 3. This invention leverages the particle swarm optimization and decision tree algorithms of the intelligent control terminal to simultaneously optimize multiple objectives, including line loss reduction rate, equipment adjustment frequency, and grid stability, when determining equipment adjustment plans and line loss parameters. This further addresses the issue of insufficient overall system economic efficiency caused by optimizing a single metric. By optimizing the equipment adjustment plan encoding through a genetic algorithm, the frequency of transformer tap adjustment is reduced by 40% while simultaneously reducing line losses. This balances the needs of short-term loss reduction with equipment lifespan maintenance, improving the long-term operational efficiency of the grid. BRIEF DESCRIPTION OF THE DRAWINGS
[0020] Figure 1 This is a system architecture diagram of the present invention; Figure 2 Flow chart of the method of the present invention. DETAILED DESCRIPTION
[0021] Example 1: Figure 1 As shown, the present invention relates to a smart grid line loss trend early warning system, which includes an early warning prediction mechanism, an execution control system, and an intelligent control terminal. Each part works in coordination through data interaction and control instruction transmission. The early warning and prediction mechanism is used to collect and analyze data and predict line loss trends and early warning levels, and includes a data collection unit, a data analysis unit, a trend analysis unit, a trend classification early warning and prediction unit, and a trend control and prediction unit.
[0022] The data acquisition unit is used to collect power grid data information and meteorological data information in real time; The power grid data information covers power grid basic data (such as line parameters, transformer parameters, etc.) and power grid flow data (such as voltage, current, power, etc.). The weather data information includes weather basic data (such as historical temperature, humidity, wind speed, etc.) and weather forecast data (such as weather forecast values in future periods). The data analysis unit is used for deep analysis of the power grid basic data and the weather basic data to obtain the grid line loss parameter. In the embodiment of the present application, the specific process of deep analysis of the power grid basic data and the weather basic data to obtain the grid line loss parameter is as follows: A1: Statistics of power grid basic data, extraction of historical data (including historical line loss data) and operation data, statistics of line loss calculation data and line loss historical data, respectively, linear relationship between line loss and related variables by using least square method is: ; In the formula, is the line loss value, and is the target variable to be predicted, is a constant term, representing the basic line loss independent of the independent variable, is the independent variable corresponding coefficient, reflecting the influence degree of each variable on the dependent variable , is the related independent variable affecting the line loss, such as power grid operation parameters, weather parameters, etc., is a random error term, representing other factors and measurement errors not considered in the model; Then, the residual sum of squares is minimized to solve the coefficient , and a line loss calculation model is established. In the formula, is the actual line loss value of the th sample, is the prediction value of the model for the th sample, is the sample number; A2: Statistics of weather basic data, including weather historical data and weather operation data; ARIMA model is used, and the basic form is , in which, is the autoregressive order, reflecting the dependence between the current value and the past values of the sequence, is the difference order, used to convert the non-stationary time series into a stationary sequence, is the moving average order, describing the relationship between the error term and the past error values; The model expression is: , in which, is the lag operator , and are the autoregressive and moving average polynomials, is a white noise error, and the meteorological forecast data is obtained through model training and fitting; The trend analysis unit obtains grid line loss trend parameters through meteorological fitting and trend analysis; In an embodiment of the present invention, the specific steps of obtaining the grid line loss trend parameter through meteorological fitting and trend analysis are as follows: B1: Fit basic meteorological data and forecast data to obtain key parameter factors affecting line loss, including wind speed, temperature, air pressure, humidity, etc. B1: Line loss prediction model is established based on the grid line loss historical data and the above parameter factors through linear fitting; The calculation formula for the grid line loss is: ; Where, is the grid line loss parameter, that is, the line loss value to be predicted, and Is the coefficient factor to be measured, reflecting historical data and Lane damage The influence weight of 、 The historical data of grid line loss is used as a reference for prediction. They are meteorological operating factors such as wind speed, air pressure, relative humidity, and temperature, which are external variables that affect line loss. is the corresponding coefficient (such as wind speed coefficient, air pressure coefficient, relative humidity coefficient, temperature coefficient), which indicates the influence of each meteorological factor on line loss.
[0023] In actual calculations, the gradient descent method of multiple linear regression can be used to solve the coefficients, and the loss function is defined as: ; Where, For the The actual line loss parameters of the samples, For the model Line loss parameters predicted by samples, is the number of samples, by continuously updating the coefficient 、 、 Minimize the loss function; The trend classification warning and prediction unit performs classification based on the corresponding relationship between meteorological indicators and grid line loss index parameters, and establishes a line loss control prediction model; In an embodiment of the present invention, based on the correspondence between meteorological indicators and grid line loss index parameters, classification is performed and a line loss control prediction model is established, which specifically includes the following steps: C1: Analyze line loss index parameters, obtain grid line loss parameter curves, and use a clustering algorithm (K-Means algorithm) to set multi-level line loss parameter warning levels (normal, warning, alarm, and emergency); The Means algorithm minimizes the sum of squares of errors within a cluster Determine cluster centers , where is the number of clusters, i.e. the number of preset line loss warning levels, For the A cluster contains all line loss parameter samples belonging to the cluster. is the line loss parameter sample, For the The algorithm updates the centers of clusters through iterations. make Minimum, divide the line loss parameter curve into multiple intervals and store the calibration parameters of each interval; C2: Use similar methods to analyze meteorological index parameters, obtain meteorological parameter curves, set meteorological parameter warning levels and fluctuation ranges, and store meteorological parameter range calibration parameters; C3: Establish a line loss control prediction model and construct a grid line loss control prediction map based on the line loss control prediction relationship; Among them, a line loss control prediction model is established, and a grid line loss control prediction map is constructed based on the line loss control prediction relationship. The specific process is as follows: introducing the weight matrix Determine the weight coefficients for each warning level, set warning intervals and line loss control prediction intervals based on the grey correlation analysis method, classify meteorological indicators into intervals, and establish grid interval control charts; Among them, grey relational analysis is performed by calculating the correlation coefficient Analyze the changing rules of line loss, where: As the reference sequence, an ideal or standard line loss sequence is usually selected. is the comparison sequence, i.e. the actual line loss or meteorological index sequence, is the resolution coefficient, which ranges from [0,1] and is used to adjust the resolution of the correlation coefficient, and is usually set to 0.5; Combine historical line loss data for trend analysis and generate a line loss trend curve. By comparing the predicted line loss curve with the line loss grid interval control chart, the line loss control interval is determined, and finally a grid line loss control prediction chart is constructed. The trend control prediction unit establishes the corresponding relationship between meteorological indicators and grid line loss index parameters, as well as the line loss control prediction relationship, to provide data support for subsequent control; Wherein, in the establishment of the corresponding relationship between the meteorological index and the grid line loss index parameter, a neural network algorithm is adopted, a multilayer perceptron (MLP) model is constructed, the meteorological index parameter is input, the corresponding grid line loss index parameter prediction value is output through the nonlinear transformation of the hidden layer, the network weight is updated by using a back propagation algorithm (BP algorithm) in the model training, and the optimization objective function is the mean square error of the prediction value and the actual value.
[0024] The execution control system is responsible for converting the early warning prediction result into an actual control action, and includes a line loss early warning execution unit and a trend control execution unit. The line loss early warning execution unit establishes a network topology structure, and constructs a power grid prediction network topology structure according to the grid line loss control prediction graph and power grid power flow parameters. In the embodiment of the present application, the network topology structure is established, the power grid prediction network topology structure is constructed according to the grid line loss control prediction graph and power grid power flow parameters, and the specific steps are as follows: D1: Obtain the line loss control interval and perform parameter calibration to obtain the line loss control interval parameter, and set the line loss control parameter interval; D2: Construct a line loss interval based on the line loss control interval of the grid line loss control prediction graph, perform topology structure modeling by using a topological sorting algorithm in graph theory to obtain a line loss topology structure graph, combine the line loss topology interval graph, the line loss control parameter interval and the power grid power flow parameters to construct a power grid prediction network topology structure, and the topological sorting algorithm ensures that the dependency relationship of the nodes in the network is correct, and outputs a loop-free topological sequence; D3: Obtain the line loss risk level in the grid line loss control prediction graph, execute grid control, and pass the control information to the trend control execution unit.
[0025] The trend control execution unit obtains the grid line loss control factor in the grid line loss control prediction graph, and generates control control information. In the embodiment of the present application, the grid line loss control factor in the grid line loss control prediction graph is obtained, and the control control information is generated, and the specific process is as follows: E1: Set the control interval and establish a control interval graph, which is used in cooperation with the line loss control interval parameter; E2: Analyze the parameters affecting the line loss control factor, and set the line loss control factor interval graph; E3: Determine the line control control interval based on the grid line loss control prediction graph, construct a line control control interval graph, combine the line loss control interval graph to establish a line control control relationship graph, and generate the control control information by using a decision tree algorithm and pass it to the trend control prediction unit to realize the control of the line loss trend.
[0026] The decision tree algorithm selects the optimal partition attribute by calculating the information gain or Gini coefficient, constructs a decision tree model, and outputs the corresponding control strategy according to the input parameters.
[0027] The intelligent control terminal is used to realize the driving and execution of the control instructions, including a control drive unit and a line loss control terminal; The control driving unit is used to receive the warning information sent by the line loss warning execution unit, and drive the line loss control terminal to execute the line loss control operation.
[0028] The line loss control terminal obtains the line loss control information from the line loss warning execution unit and performs line loss control and adjustment on the power grid; Among them, the line loss control information of the line loss warning execution unit is obtained, and the line loss control and adjustment of the power grid are carried out, specifically including obtaining the warning type and level, determining the equipment adjustment plan and adjusting the line loss control parameters (such as the operating time of the electronic control equipment, power adjustment parameters, etc.), generating the control information and transmitting it to the power grid to realize line loss control.
[0029] When determining the equipment adjustment plan, a genetic algorithm can be used to simulate natural selection and genetic mechanisms to encode, crossover, and mutate the equipment adjustment plan, and use the fitness function (such as line loss reduction rate) as the optimization target to search for the optimal equipment adjustment plan.
[0030] Example 2: Figure 2 As shown, a method for hierarchical control of line loss in a smart grid is provided, which is based on the line loss control terminal driver of the above-mentioned smart control terminal and includes the following steps: S1: driving and regulating the driving unit and obtaining the network topology; The control drive unit receives warning information from the line loss warning execution unit, drives the line loss control terminal, and obtains the predicted network topology of the power grid. This topology is constructed based on the grid line loss control prediction map and power grid flow parameters, reflecting the real-time operating status of the power grid and the distribution of line loss risks. S2: Obtain warning information; The line loss control terminal obtains warning information, including warning type (such as abnormal line loss caused by meteorological factors, abnormal line loss caused by equipment failure, etc.) and warning level (such as level 1 warning, level 2 warning, etc.), providing a basis for subsequent control decisions; S3: Determine equipment adjustment plan and control parameters; The line loss control terminal determines the equipment adjustment plan based on the warning level and determines the line loss control parameters using the grid line loss control prediction map. Line loss control parameters include the operating time of the electronically controlled equipment, operating power adjustment parameters, start-up time, power adjustment parameters, and voltage parameters. Line loss optimization is achieved by adjusting these parameters. During the parameter adjustment process, a particle swarm optimization algorithm (PSO) can be used. Each particle represents a set of parameter solutions. By updating the particle's position and velocity, the optimal solution is searched in the parameter space, with minimizing line loss as the optimization goal. S4: generating and delivering regulation information; The line loss control terminal generates regulation information according to the line control regulation relationship diagram and the regulation line loss control parameter, and specifically includes an adjustment device, an adjustment time of the adjustment device, an adjustment parameter and an adjustment value of an influence line loss control factor, etc. The regulation information is set according to the line loss control interval parameter and is delivered to the power grid to execute the regulation operation by the line loss interval control terminal; S5: the power grid executes line loss control; The power grid receives the regulation information such as the adjustment device and the adjustment time, executes the line loss control operation, realizes real-time regulation of the power grid line loss, reduces the line loss rate, and improves the power grid operation efficiency.
[0031] The embodiments of the present application are disclosed, but are not limited to this. Those skilled in the art can easily understand the spirit of the present application according to the above embodiments, and make different inferences and changes, as long as they do not deviate from the spirit of the present application, and are within the protection scope of the present application.
Claims
1. A smart grid line loss trend early warning system and hierarchical control method, characterized in that: include: Early warning and prediction agencies, used to collect and analyze data and predict line loss trends and early warning levels; The execution control system is responsible for converting the warning and prediction results into actual control actions; Intelligent control terminal, used to drive and execute control instructions, including control drive unit and line loss control terminal; The control driving unit is used to receive the warning information sent by the line loss warning execution unit and drive the line loss control terminal to execute the line loss control operation; The line loss control terminal obtains line loss control information from the line loss warning execution unit and performs line loss control and adjustment on the power grid; Among them, the line loss control information of the line loss warning execution unit is obtained, and the line loss control and adjustment of the power grid are carried out, specifically including: obtaining the warning type and level, determining the equipment adjustment plan and adjusting the line loss control parameters, generating the control information and transmitting it to the power grid to realize line loss control.
2. A smart grid line loss trend early warning system and hierarchical control method according to claim 1, characterized in that: The early warning and forecasting institutions include: Data acquisition unit, used for real-time collection of power grid data and meteorological data; Among them, the power grid data information includes power grid basic data and power grid flow data, and the meteorological data information includes meteorological basic data and meteorological forecast data; Data analysis unit, used to conduct in-depth analysis of power grid basic data and meteorological basic data to obtain grid line loss parameters; Trend analysis unit, which obtains grid line loss trend parameters through meteorological fitting and trend analysis; The trend classification warning and prediction unit classifies meteorological indicators and grid line loss index parameters based on their corresponding relationship, and establishes a line loss control prediction model; The trend control prediction unit establishes the correspondence between meteorological indicators and grid line loss index parameters, as well as the line loss control prediction relationship, to provide data support for subsequent control.
3. The smart grid line loss trend early warning system and hierarchical control method according to claim 2, characterized in that: Conduct in-depth analysis of power grid basic data and meteorological basic data to obtain grid line loss parameters. The specific process is as follows: A1: Collect basic grid data, extract historical data and operating data including historical line loss data, and collect line loss calculation data and historical line loss data respectively; A2: Statistical meteorological basic data, including meteorological historical data and meteorological operation data.
4. A smart grid line loss trend early warning system and hierarchical control method according to claim 3, characterized in that: The grid line loss trend parameters are obtained through meteorological fitting and trend analysis. The specific steps are as follows: B1: Fit basic meteorological data and forecast data to obtain key parameter factors affecting line loss; B1: Line loss prediction model is established based on the grid line loss historical data and the above parameter factors through linear fitting.
5. The smart grid line loss trend early warning system and hierarchical control method according to claim 4, characterized in that: Based on the correspondence between meteorological indicators and grid line loss index parameters, a graded classification is performed and a line loss control prediction model is established. Specifically, the following steps are included: C1: Analyze line loss index parameters, obtain grid line loss parameter curves, and use clustering algorithms to set multi-level line loss parameter warning levels; C2: Use similar methods to analyze meteorological index parameters, obtain meteorological parameter curves, set meteorological parameter warning levels and fluctuation ranges, and store meteorological parameter range calibration parameters; C3: Establish a line loss control prediction model and construct a grid line loss control prediction map based on the line loss control prediction relationship.
6. A smart grid line loss trend early warning system and hierarchical control method according to claim 5, characterized in that: When establishing the correspondence between meteorological indicators and grid line loss index parameters, a neural network algorithm is used. By constructing a multi-layer perceptron model, the meteorological indicator parameters are input, and after nonlinear transformation of the hidden layer, the corresponding grid line loss index parameter prediction values are output. The model training uses the back propagation algorithm to update the network weights, and the optimization objective function is the mean square error between the predicted value and the actual value.
7. The smart grid line loss trend early warning system and hierarchical control method according to claim 1, characterized in that: The execution control system includes: The line loss early warning execution unit establishes a network topology structure and constructs a power grid prediction network topology structure based on the grid line loss control prediction diagram and power grid flow parameters; The trend control execution unit obtains the grid line loss control factor in the grid line loss control prediction map and generates management and control information.
8. The smart grid line loss trend early warning system and hierarchical control method according to claim 7, characterized in that: Establish a network topology structure. According to the grid line loss control prediction diagram and the grid flow parameters, the grid prediction network topology structure is constructed. The specific steps are as follows: D1: Obtain the line loss control interval and perform parameter calibration to obtain the line loss control interval parameters and set the line loss control parameter interval; D2: Based on the line loss control intervals in the grid line loss control prediction map, a line loss interval is constructed. A topological sorting algorithm from graph theory is used to model the topological structure, resulting in a line loss topological structure map. Combining the line loss topological interval map, line loss control parameter intervals, and grid power flow parameters, a grid prediction network topology is constructed. The topological sorting algorithm ensures that the dependencies between nodes in the network are correct and outputs a loop-free topological sequence. D3: Obtain the line loss risk level in the grid line loss control prediction map, perform grid control, and pass the control information to the trend control execution unit.
9. The smart grid line loss trend early warning system and hierarchical control method according to claim 8, characterized in that: Obtain the grid line loss control factor in the grid line loss control prediction map and generate control and regulation information. The specific process is as follows: E1: Set the control interval and create a control interval map, which is used in conjunction with the line loss control interval parameters; E2: Analyze the parameters that affect the line loss control factor and set the line loss control factor interval diagram; E3: Determine the line control interval based on the grid line loss control prediction map, construct a line control interval map, and establish a line control relationship map based on the line loss control interval map. Use the decision tree algorithm to generate control information and pass it to the trend control prediction unit to achieve line loss trend control.
10. A method for hierarchical control of line loss in a smart grid, applicable to a smart grid line loss trend early warning system according to any one of claims 1 to 9, characterized in that: The line loss control terminal driver based on the above-mentioned intelligent control terminal includes the following steps: S1: The driving control unit receives the warning information and obtains the grid prediction network topology structure based on the grid line loss control prediction map and grid flow parameters; S2: The line loss control terminal obtains the warning type and warning level; S3: Determine the equipment adjustment plan based on the warning level, determine the line loss control parameters through the grid line loss control prediction map, and use the particle swarm optimization algorithm to optimize the parameters; S4: Generate control information based on the wire control relationship diagram and control parameters, including adjustment equipment, adjustment time, adjustment parameters and adjustment values, and transmit it to the power grid; S5: The power grid receives the control information and performs line loss control operations.