Power distribution network intelligent regulation and control system for high and low voltage power transmission

Abstract

The invention relates to the technical field of power distribution network regulation and control, in particular to a power distribution network intelligent regulation and control system for high and low voltage power transmission, comprising a global optimization layer module which is responsible for global regulation and control of the whole power distribution network and comprises a global data acquisition unit, a global data processing unit and a global control unit; the regional optimization layer module is used for performing intra-regional optimization regulation and control on the plurality of divided regions according to geographic distribution and load characteristics of the power distribution network, and the regional optimization layer module comprises a plurality of regional data acquisition units, a regional data processing unit and a regional control unit; and the adaptive learning module adaptively optimizes global and regional regulation and control strategies through a machine learning algorithm according to historical data and real-time data in operation of the power distribution network, and the adaptive learning module comprises a data training unit and a model updating unit. According to the invention, the multi-level regulation and control mode can effectively reduce the power loss in the region, and improves the overall operation efficiency of the power grid.

Description

Technical Field

[0001] The present invention relates to the technical field of distribution network control, and in particular to an intelligent control system for distribution networks for high and low voltage power transmission. Background Art

[0002] With the development of social economy and the advancement of science and technology, the demand for electricity is increasing day by day. As an important part of the power system, the intelligent and efficient management of the distribution network has become an important issue in the power industry. The existing distribution network control system has the following main problems when facing high and low voltage power transmission:

[0003] Insufficient coordination between global and regional control: Existing control systems often lack effective coordination between global and regional levels, resulting in inconsistency between the control strategies of each region and the overall optimization goals of the entire network, unreasonable resource allocation, and affecting the overall operating efficiency of the distribution network.

[0004] Low load forecasting and control accuracy: The load forecasting and control methods in traditional distribution network systems are relatively simple and difficult to cope with complex and changeable load demands, resulting in insufficient load forecasting accuracy, poor control strategy effects, and difficulty in achieving efficient power distribution and resource optimization.

[0005] Limited data processing and real-time optimization capabilities: The existing system has bottlenecks in data collection, processing and analysis, and is unable to respond to changes in the grid’s operating status in real time. It lacks dynamic optimization and adaptive control capabilities, which affects the system’s intelligent management level. Summary of the invention

[0006] Based on the above objectives, the present invention provides an intelligent control system for distribution networks for high and low voltage power transmission.

[0007] Intelligent control system for distribution network for high and low voltage power transmission, including:

[0008] Global optimization layer module: responsible for the global regulation of the entire distribution network. The global optimization layer module includes a global data acquisition unit, a global data processing unit and a global control unit. The global data acquisition unit is used to collect the operation data of the entire network. The global data processing unit analyzes the collected data and generates a global load balancing and resource optimization configuration plan. The global control unit generates a global control instruction according to the optimization plan. The global control instruction is sent to the regional optimization layer module to ensure the stability and efficiency of the overall operation of the system;

[0009] Regional optimization layer module: According to the geographical distribution and load characteristics of the distribution network, the divided multiple regions are optimized and regulated within the region. The regional optimization layer module includes multiple regional data acquisition units, regional data processing units and regional control units. The regional data acquisition units are used to collect the operation data of each region. The regional data processing units analyze the data in each region and generate the load distribution and voltage regulation scheme within the region. The regional control unit generates regional control instructions according to the load distribution and voltage regulation scheme within the region and the global control instructions, so as to reduce the power loss within the region and improve the stability of regional power supply;

[0010] Adaptive learning module: Based on the historical data and real-time data in the operation of the distribution network, the global and regional control strategies are adaptively optimized through machine learning algorithms. The adaptive learning module includes a data training unit and a model updating unit. The data training unit trains the optimization model based on the collected multi-source heterogeneous data. The model updating unit dynamically adjusts the optimization model according to the actual operating effects of the global optimization layer and the regional optimization layer to improve the accuracy and adaptability of the control strategy.

[0011] Optionally, the global data acquisition unit collects operation data of the entire network, and the global data acquisition unit obtains voltage, current, power, load, equipment status and environmental parameter data in real time through sensors, smart meters and monitoring devices arranged at each node of the distribution network, and the data is transmitted to the global data processing unit;

[0012] The global data processing unit analyzes the collected operation data, removes abnormal values ​​and noise data, and performs load forecasting, trend analysis and status evaluation based on historical data and real-time data, adopts optimization algorithms, combines overall optimization objectives and constraints, and generates global load balancing and resource optimization configuration solutions;

[0013] The global control unit generates a global control instruction according to the optimization plan. The global control unit includes a global instruction generation submodule and a global instruction sending submodule. The global instruction generation submodule formulates a specific control strategy according to the plan generated by the optimization calculation module. The global instruction sending submodule sends the control instruction to each regional optimization layer module through the communication network, and directs the execution of corresponding control operations to ensure the overall operation stability and efficiency of the distribution network.

[0014] Optionally, the global data processing unit uses a support vector regression model to predict future load demand by extracting features from historical data and real-time data, generates load forecasting results, and uses a time series decomposition method to decompose the load data into trend, seasonal and residual components, analyzes the changing characteristics of each component, and evaluates the load trend.

[0015] Optionally, the optimization algorithm adopts a linear programming method to perform global load balancing calculations based on the load demand and power supply capacity of the entire network to ensure the power supply stability of each load point and the optimal allocation of resources. Through a genetic algorithm, the utilization efficiency and configuration strategy of each resource are comprehensively considered, and the resource allocation plan is dynamically adjusted to achieve the optimal configuration of resources in the distribution network.

[0016] Optionally, the global instruction generation submodule formulates a specific control strategy according to the optimization calculation results. The control strategy includes load transfer and power regulation. The strategy is expressed as:

[0017] Strategy = {P ij |i∈ load point, j∈ power supply point}, where P ij represents the power transmitted from the power supply point j to the load point i;

[0018] The instructions sent by the global instruction sending submodule include specific load transfer instructions and power adjustment instructions.

[0019] Optionally, the regional data acquisition unit collects operating data of each area. The regional data acquisition unit obtains voltage, current, power, load, equipment status and environmental parameter data in the area in real time through sensors, smart meters and monitoring devices arranged in each area, and transmits the data to the regional data processing unit;

[0020] The regional data processing unit performs regional optimization calculation on the collected data within the region. The regional data processing unit performs load forecasting, trend analysis and status evaluation based on regional historical data and regional real-time data, and generates load distribution and voltage regulation solutions within the region in combination with the geographical distribution and load characteristics of the region;

[0021] The regional control unit generates regional control instructions according to the optimization plan. The regional control unit includes a regional instruction generation submodule and a regional instruction execution submodule. The regional instruction generation submodule formulates a specific regional control strategy according to the load distribution and voltage regulation plan in the region and the global control instructions issued by the global optimization layer module. The regional instruction execution submodule sends the control instructions to each execution device in the region through the communication network, and instructs it to perform corresponding control operations.

[0022] Optionally, the voltage regulation scheme performs voltage regulation to ensure that the voltage of each area is stable within a target range, which is expressed as:

[0023] Among them, V i is the voltage of the ith node, V target The target voltage value is achieved by adjusting the settings of the voltage regulating device to achieve the optimal voltage distribution.

[0024] Optionally, the regional instruction generation submodule formulates a regional control strategy based on the regional optimization calculation results and the global control instructions, combined with specific regional characteristics and actual conditions, including:

[0025] Identify load demand and resource allocation in the region, and determine the priority of each load point in combination with global control instructions;

[0026] Based on the optimization calculation results, the specific load transfer and power regulation schemes are determined to ensure that the load distribution and voltage regulation in the area meet the global optimization objectives.

[0027] Optionally, the data training unit trains the optimization model according to the historical data and real-time data in the operation of the distribution network. The data training unit collects multi-source heterogeneous data in the distribution network, including voltage, current, load, environmental parameters, etc. The model adopts a neural network model, and the neural network model is trained using a feedforward neural network (FNN). The optimized neural network model is generated by calculating the input layer, the hidden layer, and the output layer. During the training process, the weights and biases are updated by the back propagation algorithm and the gradient descent method to minimize the loss function;

[0028] The model updating unit dynamically adjusts the optimization model according to the actual operating effects of the global optimization layer and the regional optimization layer. The model updating unit monitors the actual effects of global and regional regulation in real time, collects feedback data, evaluates the performance of the current optimization model, and adjusts the model parameters according to the feedback data to improve the accuracy and adaptability of the regulation strategy.

[0029] Optionally, the model parameter adjustment is performed according to the feedback data, including retraining the neural network, adjusting the learning rate η and updating the training data set, the steps are as follows:

[0030] Model fine-tuning: Fine-tune the neural network through the mini-batch gradient descent method to ensure that the model adapts to the new data:

[0031] Among them, θ represents all model parameters (weights and biases), m is the number of samples in the mini-batch, is the gradient of the loss function with respect to the parameters;

[0032] Retraining: When the model performs poorly, retrain the neural network using new feedback data to ensure the accuracy and adaptability of the model.

[0033] Beneficial effects of the present invention:

[0034] The present invention realizes coordinated management from the whole network to each region through multi-level optimization and control of the global optimization layer and the regional optimization layer. The global optimization layer is responsible for collecting and analyzing the data of the whole network, generating global load balancing and resource optimization configuration schemes, and ensuring the stability and efficiency of the overall operation of the system; the regional optimization layer performs detailed analysis and optimization calculations on the data in each region according to the geographical distribution and load characteristics, and generates load distribution and voltage regulation schemes in the region, thereby improving the power supply stability and efficiency of each region. This multi-level control method can effectively reduce power loss in the region and improve the overall operation efficiency of the power grid.

[0035] In the present invention, the adaptive learning module uses a neural network algorithm to train and optimize according to historical data and real-time data in the operation of the distribution network to form an optimization model that adapts to different control needs. The module includes a data training unit and a model updating unit. The data training unit uses multi-source heterogeneous data to train the model. The model updating unit dynamically adjusts the optimization model according to the actual operation effects of the global and regional optimization layers. The application can adaptively optimize the global and regional control strategies, improve the accuracy and adaptability of the control strategies, ensure that the distribution network can respond to load changes and emergencies in a timely manner, and improve the intelligent management level of the system.

[0036] The present invention establishes an effective coordination mechanism between global control and regional control. The global control instructions provide macro-guidance for regional control, ensuring that the control strategy of each region is consistent with the goal of the entire network. The regional control unit formulates a specific regional control strategy based on the global control instructions and the regional optimization calculation results to ensure that the load distribution and voltage regulation in the region meet the global optimization goals. This coordinated control mechanism not only improves the power supply stability and efficiency in each region, but also enhances the overall coordination and operating efficiency of the system, effectively avoiding resource conflicts and imbalances between regions. BRIEF DESCRIPTION OF THE DRAWINGS

[0037] In order to more clearly illustrate the technical solutions in the present invention or the prior art, the drawings required for use in the embodiments or the description of the prior art will be briefly introduced below. Obviously, the drawings in the following description are only for the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative work.

[0038] Figure 1 Schematic diagram of system function modules according to an embodiment of the present invention. DETAILED DESCRIPTION

[0039] The present invention is described in detail below in conjunction with the accompanying drawings and specific embodiments. At the same time, it is explained here that in order to make the embodiments more detailed, the following embodiments are the best and preferred embodiments, and those skilled in the art may also adopt other alternatives to implement some known technologies; and the accompanying drawings are only for more specific description of the embodiments, and are not intended to specifically limit the present invention.

[0040] It should be noted that the references to "one embodiment", "an embodiment", "an exemplary embodiment", "some embodiments" and the like in the specification indicate that the embodiments described may include specific features, structures or characteristics, but not every embodiment may include the specific features, structures or characteristics. In addition, when a specific feature, structure or characteristic is described in conjunction with an embodiment, it should be within the knowledge of a person skilled in the art to implement such feature, structure or characteristic in conjunction with other embodiments (whether or not explicitly described).

[0041] In general, a term can be understood, at least in part, from its use in context. For example, depending, at least in part, on the context, the term "one or more" as used herein can be used to describe any feature, structure, or characteristic in the singular sense, or can be used to describe a combination of features, structures, or characteristics in the plural sense. Additionally, the term "based on" can be understood as not necessarily intended to convey an exclusive set of factors, but can instead, depending, at least in part, on the context, allow for the presence of other factors that are not necessarily explicitly described.

[0042] like Figure 1 As shown in the figure, the intelligent control system for distribution network for high and low voltage transmission includes the following modules:

[0043] Global optimization layer module: responsible for the global control of the entire distribution network. The global optimization layer module includes a global data acquisition unit, a global data processing unit and a global control unit. The global data acquisition unit is used to collect the operation data of the entire network. The global data processing unit analyzes the collected data and generates a global load balancing and resource optimization configuration plan. The global control unit generates a global control instruction based on the optimization plan, and the global control instruction is sent to the regional optimization layer module to ensure the stability and efficiency of the overall operation of the system.

[0044] Regional optimization layer module: According to the geographical distribution and load characteristics of the distribution network, the regional optimization layer module optimizes and controls the multiple divided regions. The regional optimization layer module includes multiple regional data acquisition units, regional data processing units and regional control units. The regional data acquisition unit is used to collect the operating data of each region. The regional data processing unit analyzes the data in each region and generates the load distribution and voltage regulation plan in the region. The regional control unit generates regional control instructions based on the load distribution and voltage regulation plan in the region and the global control instructions to reduce the power loss in the region and improve the stability of regional power supply.

[0045] Adaptive learning module: Based on the historical data and real-time data of the distribution network operation, the global and regional control strategies are adaptively optimized through machine learning algorithms. The adaptive learning module includes a data training unit and a model updating unit. The data training unit trains the optimization model based on the collected multi-source heterogeneous data. The model updating unit dynamically adjusts the optimization model according to the actual operating results of the global optimization layer and the regional optimization layer to improve the accuracy and adaptability of the control strategy.

[0046] The global data acquisition unit collects the operation data of the entire network. The global data acquisition unit obtains voltage, current, power, load, equipment status and environmental parameter data in real time through sensors, smart meters and monitoring devices arranged at each node of the distribution network, and transmits the data to the global data processing unit;

[0047] The global data processing unit analyzes the collected operation data, removes abnormal values ​​and noise data, and performs load forecasting, trend analysis and status evaluation based on historical data and real-time data. It uses optimization algorithms, combines overall optimization goals and constraints, and generates global load balancing and resource optimization configuration solutions;

[0048] Outlier detection and removal: Use the zscore method to detect outliers: Where x is the data point, μ is the mean of the data, and σ is the standard deviation of the data. When |z|>3, it is considered an outlier.

[0049] Data points x exceeding three times the standard deviation are considered abnormal. After removing these data, the mean and standard deviation are recalculated and repeated until there are no abnormal values.

[0050] Noise filtering: Smooth the data using a weighted moving average filter:

[0051] in, is the smoothed data at time t, N is the window size, w i is the weight coefficient, usually w i Can be set to decreasing weights, such as exponentially weighted average;

[0052] The global control unit generates global control instructions according to the optimization plan. The global control unit includes a global instruction generation submodule and a global instruction sending submodule. The global instruction generation submodule formulates a specific control strategy according to the plan generated by the optimization calculation module. The global instruction sending submodule sends the control instructions to each regional optimization layer module through the communication network, and directs the execution of corresponding control operations to ensure the overall operation stability and efficiency of the distribution network.

[0053] The global data processing unit uses the support vector regression model to extract features from historical data and real-time data, predict future load demand, generate load forecast results, and use the time series decomposition method to decompose the load data into trend, seasonal and residual components, analyze the changing characteristics of each component, and evaluate the load trend.

[0054] The specific plan is as follows:

[0055] Load Forecasting: Load forecasting using Support Vector Regression (SVR) model:

[0056] in, is the predicted load, x i is the training sample, α i is the support vector, K(x i ,x) is the kernel function (such as Gaussian kernel), and b is the bias term.

[0057] By training the SVR model, optimize the parameter α i and b, and make predictions based on real-time data x.

[0058] Trend analysis: Analyze load trends using time series decomposition methods:

[0059] y(t)=T(t)+S(t)+R(t), where y(t) is the load value at time t, T(t) is the trend component, S(t) is the seasonal component, and R(t) is the random component.

[0060] By using the STL method, the time series is decomposed into trend, seasonality and residual parts:

[0061] T(t) = LOESS(y(t));

[0062] S(t)=y(t)-T(t)-R(t);

[0063] R(t)=y(t)-T(t)-S(t).

[0064] The optimization algorithm adopts linear programming method to perform global load balance calculation based on the load demand and power supply capacity of the whole network to ensure the power supply stability of each load point and the optimal allocation of resources. Through genetic algorithm, it comprehensively considers the utilization efficiency and configuration strategy of each resource, dynamically adjusts the resource allocation plan, and realizes the optimal configuration of resources in the distribution network.

[0065] The specific plan is as follows:

[0066] Global load balancing: Use linear programming method for load balancing:

[0067]

[0068] Among them, x i is the variable to be optimized, representing the power supply of each load point, c i is the cost coefficient, which represents the power supply cost of each load point, a ij is the constraint coefficient matrix, which represents the relationship between each power supply point and load point, b j is the constraint condition, which represents the load demand.

[0069] Use the simplext method to solve the linear programming problem and get the optimal power supply solution x i .

[0070] Resource optimization configuration: Use genetic algorithm (GA) for resource optimization:

[0071] Among them, f(x) is the fitness function, which represents the overall benefit of resource allocation, and w i is the weight coefficient, which indicates the weight of each resource, g i (x i ) is the resource allocation utility function, which represents the allocation benefit of resource i.

[0072] The genetic algorithm process includes selection, crossover and mutation, and the optimal resource allocation solution is found through multiple generations of evolution.

[0073] The global instruction generation submodule formulates specific control strategies based on the optimization calculation results. The control strategies include load transfer and power regulation. The strategies are expressed as:

[0074] Strategy = {P ij |i∈ load point, j∈ power supply point}, where P ij represents the power transmitted from the power supply point j to the load point i;

[0075] The instructions sent by the global instruction sending submodule include specific load transfer instructions and power adjustment instructions.

[0076] The regional data acquisition unit collects the operating data of each area. The regional data acquisition unit obtains the voltage, current, power, load, equipment status and environmental parameter data in the area in real time through sensors, smart meters and monitoring devices arranged in each area, and transmits it to the regional data processing unit;

[0077] The regional data processing unit performs regional optimization calculations on the collected data within the region. The regional data processing unit performs load forecasting, trend analysis and status evaluation based on the regional historical data and regional real-time data, and generates load distribution and voltage regulation plans within the region in combination with the regional geographical distribution and load characteristics;

[0078] The regional control unit generates regional control instructions according to the optimization plan. The regional control unit includes a regional instruction generation submodule and a regional instruction execution submodule. The regional instruction generation submodule formulates specific regional control strategies according to the load distribution and voltage regulation plan in the region and the global control instructions issued by the global optimization layer module. The regional instruction execution submodule sends the control instructions to each execution device in the region through the communication network, and instructs them to perform corresponding control operations.

[0079] The above analysis and calculations are basically the same as those in the previous global optimization layer module, including data cleaning, data analysis (such as load forecasting and trend analysis) and optimization calculations (such as load balancing and resource allocation), but the application level is different. The global optimization layer module is for global regulation of the entire distribution network, while the regional optimization layer module analyzes and optimizes data in a specific area. The difference is that the regional control unit needs to formulate a specific regional control strategy based on the optimization plan generated by the regional optimization calculation module and the global control instructions issued by the global optimization layer module.

[0080] The voltage regulation scheme regulates the voltage to ensure that the voltage in each area is stable within the target range, which is expressed as:

[0081] Among them, V i is the voltage of the ith node, V target The target voltage value is achieved by adjusting the settings of the voltage regulating device to achieve the optimal voltage distribution.

[0082] The regional instruction generation submodule formulates regional control strategies based on the regional optimization calculation results and global control instructions, combined with specific regional characteristics and actual conditions, including:

[0083] Identify load demand and resource allocation in the region, and determine the priority of each load point in combination with global control instructions;

[0084] Based on the optimization calculation results, the specific load transfer and power regulation schemes are determined to ensure that the load distribution and voltage regulation in the area meet the global optimization objectives.

[0085] For example, if the global control instruction requires reducing the load in a certain area to relieve the pressure on the entire network, the regional control unit will give priority to reducing the power supply to non-critical loads in the area, while adjusting the power supply of other load points to balance the overall load.

[0086] The data training unit trains the optimization model based on the historical data and real-time data in the distribution network operation. The data training unit collects multi-source heterogeneous data in the distribution network, including voltage, current, load, environmental parameters, etc. The model adopts a neural network model, and the neural network model is trained using a feedforward neural network (FNN). Through the calculation of the input layer, hidden layer and output layer, an optimized neural network model is generated. During the training process, the weights and biases are updated through the back propagation algorithm and the gradient descent method to minimize the loss function; the details are as follows:

[0087] Input layer: accepts input data X, where X = [x1, x2, ..., x n ] represents multiple input characteristics, including voltage, current, power, load, equipment status, and environmental parameters (temperature, humidity, wind speed).

[0088] Hidden layer: nonlinear transformation through weight matrix W and activation function σ:

[0089] Among them, w ij is the weight, b j is the bias, h j is the output of the hidden unit.

[0090] Output layer: Generates the final output through another set of weight matrices V and activation function σ′

[0091] Among them, v jk is the weight, c k is the bias, is the output of the output unit.

[0092] Loss function: Use mean square error as the loss function:

[0093] Where N is the number of samples, y i is the true value, is the predicted value.

[0094] Backpropagation and weight update: Gradients are calculated through the backpropagation algorithm, and weights and biases are updated using gradient descent: Among them, η is the learning rate.

[0095] The model update unit dynamically adjusts the optimization model according to the actual operating results of the global optimization layer and the regional optimization layer. The model update unit monitors the actual effects of global and regional regulation in real time, collects feedback data, evaluates the performance of the current optimization model, and adjusts the model parameters according to the feedback data to improve the accuracy and adaptability of the regulation strategy.

[0096] The model parameter adjustment is performed based on the feedback data, including retraining the neural network, adjusting the learning rate η, and updating the training data set. The steps are as follows:

[0097] Model fine-tuning: Fine-tune the neural network through the mini-batch gradient descent method to ensure that the model adapts to the new data:

[0098] Among them, θ represents all model parameters (weights and biases), m is the number of samples in the mini-batch, is the gradient of the loss function with respect to the parameters;

[0099] Retraining: When the model performs poorly, retrain the neural network using new feedback data to ensure the accuracy and adaptability of the model.

[0100] The present invention covers any substitution, modification, equivalent method and scheme made on the essence and scope of the present invention. In order to make the public have a thorough understanding of the present invention, specific details are described in detail in the following preferred embodiments of the present invention, but those skilled in the art can fully understand the present invention without the description of these details. In addition, in order to avoid unnecessary confusion about the essence of the present invention, well-known methods, processes, procedures, components and circuits are not described in detail.

[0101] The above is only a preferred embodiment of the present invention. It should be pointed out that for ordinary technicians in this technical field, several improvements and modifications can be made without departing from the principle of the present invention. These improvements and modifications should also be regarded as the scope of protection of the present invention.

Claims

1. Intelligent control system for distribution network for high and low voltage power transmission, characterized by: include: Global optimization layer module: responsible for the global regulation of the entire distribution network. The global optimization layer module includes a global data acquisition unit, a global data processing unit and a global control unit. The global data acquisition unit is used to collect the operation data of the entire network. The global data processing unit analyzes the collected data and generates a global load balancing and resource optimization configuration plan. The global control unit generates a global control instruction according to the optimization plan, and the global control instruction is sent to the regional optimization layer module; Regional optimization layer module: according to the geographical distribution and load characteristics of the distribution network, the regional optimization layer module performs optimization and control on the divided multiple regions. The regional optimization layer module includes multiple regional data acquisition units, regional data processing units and regional control units. The regional data acquisition units are used to collect the operation data of each region. The regional data processing units analyze the data in each region and generate the load distribution and voltage regulation scheme in the region. The regional control unit generates the regional control instruction according to the load distribution and voltage regulation scheme in the region and the global control instruction. Adaptive learning module: Based on the historical data and real-time data in the operation of the distribution network, the global and regional control strategies are adaptively optimized through machine learning algorithms. The adaptive learning module includes a data training unit and a model updating unit. The data training unit trains the optimization model based on the collected multi-source heterogeneous data. The model updating unit dynamically adjusts the optimization model according to the actual operating effects of the global optimization layer and the regional optimization layer.

2. The intelligent control system for distribution network for high and low voltage power transmission according to claim 1 is characterized in that: The global data acquisition unit collects the operation data of the whole network. The global data acquisition unit obtains voltage, current, power, load, equipment status and environmental parameter data in real time through sensors, smart meters and monitoring devices arranged at each node of the distribution network. The data is transmitted to the global data processing unit; The global data processing unit analyzes the collected operation data, removes abnormal values ​​and noise data, and performs load forecasting, trend analysis and status evaluation based on historical data and real-time data, adopts optimization algorithms, combines overall optimization objectives and constraints, and generates global load balancing and resource optimization configuration solutions; The global control unit generates global control instructions according to the optimization plan. The global control unit includes a global instruction generation submodule and a global instruction sending submodule. The global instruction generation submodule formulates a specific control strategy according to the plan generated by the optimization calculation module. The global instruction sending submodule sends the control instructions to each regional optimization layer module through the communication network to command the execution of corresponding control operations.

3. The intelligent control system for distribution network for high and low voltage power transmission according to claim 2 is characterized in that: The global data processing unit uses a support vector regression model to extract features from historical data and real-time data to predict future load demand and generate load forecast results. It uses a time series decomposition method to decompose load data into trend, seasonal and residual components, analyzes the change characteristics of each component, and evaluates the load trend.

4. The intelligent control system for distribution network for high and low voltage power transmission according to claim 3 is characterized in that: The optimization algorithm adopts linear programming method to perform global load balance calculation based on the load demand and power supply capacity of the whole network. Through genetic algorithm, it comprehensively considers the utilization efficiency and configuration strategy of each resource, dynamically adjusts the resource allocation plan, and realizes the optimal configuration of resources in the distribution network.

5. The intelligent control system for distribution network for high and low voltage power transmission according to claim 2 is characterized in that: The global instruction generation submodule formulates a specific control strategy based on the optimization calculation results. The control strategy includes load transfer and power regulation. The strategy is expressed as: Strategy = {P ij |i∈ load point, j∈ power supply point}, where P ij represents the power transmitted from the power supply point j to the load point i; The instructions sent by the global instruction sending submodule include specific load transfer instructions and power adjustment instructions.

6. The intelligent control system for distribution network for high and low voltage power transmission according to claim 2 is characterized in that: The regional data acquisition unit collects the operating data of each area. The regional data acquisition unit obtains the voltage, current, power, load, equipment status and environmental parameter data in the area in real time through sensors, smart meters and monitoring devices arranged in each area, and transmits the data to the regional data processing unit; The regional data processing unit performs regional optimization calculation on the collected data within the region. The regional data processing unit performs load forecasting, trend analysis and status evaluation based on regional historical data and regional real-time data, and generates load distribution and voltage regulation solutions within the region in combination with the geographical distribution and load characteristics of the region; The regional control unit generates regional control instructions according to the optimization plan. The regional control unit includes a regional instruction generation submodule and a regional instruction execution submodule. The regional instruction generation submodule formulates a specific regional control strategy according to the load distribution and voltage regulation plan in the region and the global control instructions issued by the global optimization layer module. The regional instruction execution submodule sends the control instructions to each execution device in the region through the communication network, and instructs it to perform corresponding control operations.

7. The intelligent control system for distribution network for high and low voltage power transmission according to claim 6 is characterized in that: The voltage regulation scheme performs voltage regulation to ensure that the voltage of each area is stable within the target range, which is expressed as: Among them, V i is the voltage of the ith node, V target The target voltage value is achieved by adjusting the settings of the voltage regulating device to achieve the optimal voltage distribution.

8. The intelligent control system for distribution network for high and low voltage power transmission according to claim 7 is characterized in that: The regional instruction generation submodule formulates a regional control strategy based on the regional optimization calculation results and the global control instructions, combined with the specific regional characteristics and actual conditions, including: Identify load demand and resource allocation in the region, and determine the priority of each load point in combination with global control instructions; Based on the optimization calculation results, the specific load transfer and power regulation schemes are determined to ensure that the load distribution and voltage regulation in the area meet the global optimization goals.

9. The intelligent control system for distribution network for high and low voltage power transmission according to claim 1 is characterized in that: The data training unit trains the optimization model according to the historical data and real-time data in the operation of the distribution network. The data training unit collects multi-source heterogeneous data in the distribution network, including voltage, current, load, environmental parameters, etc. The model adopts a neural network model. The neural network model is trained using a feedforward neural network. The neural network model is generated by calculating the input layer, the hidden layer, and the output layer. During the training process, the weights and biases are updated by the back propagation algorithm and the gradient descent method to minimize the loss function. The model updating unit dynamically adjusts the optimization model according to the actual operating effects of the global optimization layer and the regional optimization layer. The model updating unit collects feedback data by real-time monitoring of the actual effects of global and regional regulation, evaluates the performance of the current optimization model, and adjusts the model parameters according to the feedback data.

10. The intelligent control system for distribution network for high and low voltage power transmission according to claim 9, characterized in that: The model parameter adjustment is performed according to the feedback data, including retraining the neural network, adjusting the learning rate η and updating the training data set, the steps are as follows: Model fine-tuning: Fine-tune the neural network through the mini-batch gradient descent method to ensure that the model adapts to the new data: Among them, θ represents all model parameters, m is the number of samples in a small batch, is the gradient of the loss function with respect to the parameters; Retraining: Retrain the neural network using new feedback data to ensure the accuracy and adaptability of the model.

Images