Low-voltage power distribution network operation control system and control method based on cloud side end

Abstract

The invention belongs to the technical field of power distribution network operation management systems, and particularly relates to a low-voltage power distribution network operation control system and method based on a cloud side end. Comprising a cloud service layer, an edge service layer and a terminal equipment layer, the cloud service layer comprises a data analysis processing module, a cloud computing service module and a scheduling control module; the edge control layer is composed of an edge fusion terminal and an edge control terminal; the terminal equipment layer is composed of tail end control equipment and tail end collection equipment; based on a current cloud side end service structure, an edge service optimization mode is provided, the thinking of feedback regulation control is utilized, and the final purpose of power distribution control service is taken as an optimization basis; the low-voltage power distribution network operation control system and method based on the cloud side end are efficient and can effectively reduce the data processing pressure and cost of the cloud server end and the edge server end.

Description

Technical Field

[0001] The present invention belongs to the technical field of distribution network operation management systems, and in particular, relates to a cloud-edge-based low-voltage distribution network operation control system and control method. Background Art

[0002] The edge allocation of computing and analysis of huge resources based on cloud-edge technology can effectively reduce the computing pressure of cloud service centers and avoid the data transmission and storage costs caused by the long-distance transmission of huge basic data and result data. At present, the main difficulties in the cloud-edge processing process of distribution network operation control are concentrated on the overly large basic equipment and various data forms. Although the data format and standard can be unified in the processing process, the efficiency and timeliness of the data path will be relatively reduced. In many cases, the timeliness of data is often crucial for power grid operation control. In order to meet the real-time data processing needs and use data efficiently, it is often difficult to achieve for most existing edge service terminals, or the cost is high. Summary of the invention

[0003] The purpose of the present invention is to provide, based on actual needs, a cloud-edge-based low-voltage distribution network operation control system and a control method thereof that can simultaneously reduce the data processing pressure on the cloud service end and the edge service end while maintaining the real-time data processing capability in the aforementioned scenario, can better realize the real-time processing capability, and fully guarantee and utilize the timeliness of the data.

[0004] To achieve the above purpose, the present invention adopts the following technical solution.

[0005] A low-voltage distribution network operation control system based on cloud-edge, including a cloud service layer, an edge service layer, and a terminal device layer;

[0006] The cloud service layer includes data analysis and processing module, cloud computing service module, and scheduling control module;

[0007] The edge control layer consists of several edge fusion terminals and edge control terminals serving each area;

[0008] The terminal device layer consists of terminal control devices and terminal collection devices.

[0009] The analysis and processing module extracts the regional load forecast results and actual load monitoring data obtained by the edge control layer and performs necessary preprocessing to enable them to be used for cloud computing services;

[0010] The cloud computing service module establishes regional load forecasting models and power grid coordination control models based on data mining and machine learning, and completes the training of the corresponding models;

[0011] The cloud computing service module periodically optimizes the regional load forecasting model and sends the trained regional load forecasting model to the edge fusion terminal in the edge control layer;

[0012] The grid coordination control model determines the control scheme for power distribution parameters in each area;

[0013] The dispatch control module generates dispatch control instructions based on the output of the power grid coordination control model and sends them to the edge control layer.

[0014] The edge fusion terminal is set up near the managed area and serves as an edge computing resource node to provide computing resource support for the cloud service end and perform collaborative computing with other edge computing resource nodes. The edge control terminal completes the coordinated control of the distribution network and optimizes regional operation control.

[0015] The terminal control equipment completes the specific execution of the power distribution status control in the area; the terminal collection equipment obtains and processes the monitoring or data of the distribution station area, smart substation, new energy site and power station operation data, providing a basis for the system to perceive and analyze the power grid status.

[0016] A further improvement or preferred implementation of the aforementioned cloud-edge-based low-voltage distribution network operation control system, the regional load forecasting model specifically aims to minimize the power loss and control cost of the regional distribution network and determine the regional distribution parameter control plan;

[0017] The cloud computing service module periodically trains and optimizes the regional load forecasting model based on the regional load forecasting results and the actual load monitoring data, and sends the trained regional load forecasting model to the edge fusion terminal in the edge control layer;

[0018] The grid coordination control model aims to minimize power loss and maximize energy utilization in the distribution network. It uses the regional load forecast results uploaded by the edge fusion terminal in the edge control layer as input, determines the regional distribution parameter control scheme for each area in the distribution area, and outputs the control parameters of the edge control terminal in the edge control layer.

[0019] The dispatching control module extracts and classifies the control parameter outputs of each edge control terminal output by the power grid coordination control model, and generates corresponding dispatching control instructions, sending direct parameter control instructions to the edge control terminal and interval dynamic parameter control instructions to the edge fusion terminal.

[0020] A further improvement or preferred implementation of the aforementioned cloud-edge-based low-voltage distribution network operation control system, distribution network collaborative control refers to the edge fusion terminal obtaining the trained regional load forecasting model from the cloud computing service module, using the monitoring data collected by the terminal device layer to realize regional load forecasting, determining the preferred control scheme of the edge control terminal in the corresponding area and executing it, uploading the corresponding load forecast results and the load results obtained by actual monitoring to the cloud service layer, and performing periodic tracking and control according to the control instructions issued by the dispatching control module;

[0021] Regional operation control optimization refers to the area controlled by the edge fusion terminal, using the operation control parameters of the same period in the previous decision cycle as the control basis, the actual load and power output of the region in each time period as input, and the goal of minimizing the power loss and control cost of the distribution network in the region, to determine the optimal control scheme for the edge control terminal.

[0022] A further improvement or preferred implementation of the aforementioned cloud-edge-based low-voltage distribution network operation and control system, the terminal control equipment includes static VAR compensators, shunt capacitors, on-load tap-changing transformers, and low-voltage voltage regulators arranged in each area; the terminal collection equipment includes distribution switches, voltage regulating equipment, inverters, sensors, and data acquisition terminals arranged in each area.

[0023] A further improvement or preferred implementation of the aforementioned cloud-edge-based low-voltage distribution network operation control system, wherein the regional load forecasting model is created based on the following method:

[0024] A. Determine the load forecasting model control target Goal = λ1Goal1 + λ2Goal2, where:

[0025] Regional power loss minimization goal

[0026] The regional control cost minimization goal Goal2 = minC0 = C1 + C2;

[0027] λ1 refers to the regional power loss decision weight, λ2 refers to the regional control cost decision weight; T refers to the data collection period, I refers to the number of regional nodes, J i It refers to the set consisting of node j of branch ij where node i is located, Δt is the control period; C0 refers to the regional control cost, C1 refers to the control execution cost, and C2 refers to the power dispatch cost;

[0028] B. Determine the constraints of the load forecasting model

[0029]

[0030] Where D is the total number of node power-related devices, It refers to the active power injection of node device d at the i-th node at time t; It refers to the reactive power output of node device d at the i-th node at time t; It refers to the voltage amplitude of the nodes at both ends of the line in, A in , B in Refers to the real and imaginary parts of the line in node admittance matrix, It refers to the voltage phase difference between the two ends of line in at time t.

[0031] Further improvement or preferred implementation of the aforementioned cloud-edge-based low-voltage distribution network operation control system,

[0032] Where R refers to the number of control device types, K r is the total number of devices in category r, c r It refers to the cost of each level adjustment control of the r-type equipment; It refers to the device control level in the tth time period, It refers to the control level of the control device in the t+1th time period, c r It refers to the control adjustment cost of each level of the rth type of equipment.

[0033]

[0034] Where S refers to the total number of transmission and distribution nodes in the region. st It refers to the amount of electricity input through the sth node in the tth cycle, out st It refers to the total amount of electricity output through the sth node in the tth cycle, P t It refers to the electricity price in the tth period.

[0035] The present application also provides a low-voltage distribution network operation control method based on cloud edge, comprising the following steps:

[0036] Stp1, obtain regional power grid operation status data based on the terminal acquisition equipment in each region and upload it to the edge fusion terminal;

[0037] Stp2, the edge fusion terminal obtains the regional load forecasting model from the cloud extreme service module, and uses the collected regional power grid operation status data to predict the regional power grid load; the edge fusion terminal runs the regional operation control plan to optimize according to the current load forecast results, determines the best operation control plan under the current load forecast results and sends it to the edge control terminal for optimization control;

[0038] The edge fusion terminal periodically tracks the grid load based on the data collected by the regional terminal acquisition device, and uploads the load forecast results and real-time load monitoring values ​​to the cloud computing service module;

[0039] Stp3, the edge control terminal determines and executes the specific control plan according to the regional operation control plan instructions and cloud instructions;

[0040] Stp4, the cloud computing service module establishes a regional load forecasting model, optimizes and trains the load forecasting model based on the load forecast value calculated by the edge fusion terminal and the actual load monitoring value, and sends the trained load model to the edge fusion terminal;

[0041] The cloud computing service module collects the load forecast results uploaded by each region, inputs the regional power distribution parameter control scheme of each region in the power distribution area determined by the power grid coordination control model, and outputs the control parameters of the edge control terminal in the edge control layer;

[0042] The cloud computing service module periodically collects the load forecast results uploaded by each region, inputs the regional power distribution parameter control scheme of each region in the power distribution area determined by the power grid coordination control model, and periodically outputs the control parameters of the edge control terminal in the edge control layer;

[0043] Stp5, the scheduling control module extracts the output structure of the cloud computing service module, generates static control instructions and dynamic control instructions according to the control parameter type, and sends them to the edge control terminal and the edge fusion terminal respectively;

[0044] Stp6, the edge fusion terminal obtains periodic dynamic control instructions, and generates regional control instructions according to the regional power grid status and sends them to the edge control terminal. The edge control terminal obtains the regional control instructions from the edge fusion terminal and the static instructions issued by the dispatching control module, and generates the final control instruction set on the premise of preferentially satisfying the static instructions and secondly satisfying the regional control instructions;

[0045] Stp7, the terminal control device obtains the control instruction set to complete the control of the terminal device and returns to Stp1 for continuous execution.

[0046] A further improvement or preferred implementation of the aforementioned cloud-edge-based low-voltage distribution network operation control method, wherein the static control instruction refers to an instruction corresponding to a control parameter that can be arbitrarily changed within a determined value or a static range; the dynamic instruction refers to an instruction corresponding to a control parameter that needs to rely on other parameters to determine the parameter value or parameter value range.

[0047] A further improvement or preferred implementation of the aforementioned cloud-edge-based low-voltage distribution network operation control method, the periodic data collection and instruction generation are based on a period that is determined based on the power supply and distribution cycle time of the distribution network, and the period refers to a day, a week, etc.

[0048] Its beneficial effects are:

[0049] Based on the current cloud-edge service structure, this application proposes to optimize the edge service mode and utilize the thinking of feedback regulation and control. Taking the ultimate goal of distribution control service as the basis for optimization, this application establishes an efficient low-voltage distribution network operation control system and method based on the cloud-edge, which can effectively reduce the data processing pressure and cost of the cloud service end and the edge service end. BRIEF DESCRIPTION OF THE DRAWINGS

[0050] Figure 1 This is the system structure diagram of the low-voltage distribution network operation control system based on the cloud-edge terminal;

[0051] Figure 2 It is a basic flow chart of the low-voltage distribution network operation control method based on the cloud-edge. DETAILED DESCRIPTION

[0052] The present invention is described in detail below in conjunction with specific embodiments.

[0053] The cloud-edge-based low-voltage distribution network operation and control system of the present application is mainly used to solve the real-time data processing of large amounts of data faced by the distribution network and the instant processing needs based on real-time data. Through the feedback processing process based on the ultimate goal of the control service, the long-distance transmission and storage of huge basic data are avoided. The computing resources of the cloud service layer are used to build a regional control model, which is sent to the cloud-edge server that can directly obtain real-time data. The cloud service layer optimizes and trains multiple regional control models in the region according to the ultimate goal, and does not directly process and analyze real-time data, effectively avoiding data delays; the edge control layer only needs to directly use real-time data to implement real-time control of the regional power grid, thereby reducing control delays. The edge server reduces the pressure on its own cloud computing by directly calling the control model of the cloud service layer. The terminal device layer ensures the efficient collection of data required for system operation through various control structures and data collection structures.

[0054] The basic structural principle of the low-voltage distribution network operation control system based on the cloud-edge terminal of the present invention is as follows Figure 1 As shown, it mainly includes cloud service layer, edge service layer and terminal device layer;

[0055] The cloud service layer includes data analysis and processing modules, cloud computing service modules, and scheduling control modules; the edge control layer is composed of several edge fusion terminals and edge control terminals serving various regions; the terminal device layer is composed of terminal control devices and terminal collection devices;

[0056] The cloud service layer is used to establish and continuously improve the service control model required by the edge server according to the final demand of the distribution network operation control. It does not directly run and use the service control model, but can use the feedback data of each control result in the region to continuously train and optimize each control model itself, and connect the model prediction output of multiple regions in series to analyze the overall state of the distribution network, and then perform early control. The cloud service layer only needs to use historical management and control data and the prediction results and actual results uploaded by the edge control layer to train and analyze the control model, without the huge real-time data generated in the current system, which effectively avoids a large number of real-time data transmission delays and the corresponding data transmission and storage costs over long distances.

[0057] Specifically, the analysis and processing module extracts the regional load forecast results and actual load monitoring data obtained by the edge control layer and performs necessary preprocessing to make them available for cloud computing services;

[0058] The cloud computing service module establishes regional load forecasting models and power grid coordination control models based on data mining and machine learning, and completes the training of the corresponding models; the cloud computing service module periodically optimizes the regional load forecasting model, and sends the trained regional load forecasting model to the edge fusion terminal in the edge control layer;

[0059] The grid coordination control model determines the control scheme for power distribution parameters in each area;

[0060] The dispatch control module generates dispatch control instructions based on the output of the power grid coordination control model and sends them to the edge control layer; the regional load forecasting model specifically determines the regional distribution parameter control plan with the goal of minimizing the power loss and control cost of the regional distribution network;

[0061] The cloud computing service module periodically trains and optimizes the regional load forecasting model based on the regional load forecasting results and the actual load monitoring data, and sends the trained regional load forecasting model to the edge fusion terminal in the edge control layer;

[0062] The grid coordination control model aims to minimize power loss and maximize energy utilization in the distribution network. It uses the regional load forecast results uploaded by the edge fusion terminal in the edge control layer as input, determines the regional distribution parameter control scheme for each area in the distribution area, and outputs the control parameters of the edge control terminal in the edge control layer.

[0063] The dispatching control module extracts and classifies the control parameter outputs of each edge control terminal output by the power grid coordination control model, and generates corresponding dispatching control instructions, sending direct parameter control instructions to the edge control terminal and interval dynamic parameter control instructions to the edge fusion terminal.

[0064] The edge fusion terminal is set up near the managed area and serves as an edge computing resource node to provide computing resource support for the cloud service end and perform collaborative computing with other edge computing resource nodes. The edge control terminal completes the coordinated control of the distribution network and optimizes regional operation control.

[0065] Distribution network collaborative control means that the edge fusion terminal obtains the trained regional load forecasting model from the cloud computing service module, uses the monitoring data collected by the terminal device layer to realize regional load forecasting, determines the optimal control scheme of the edge control terminal in the corresponding area and executes it, uploads the corresponding load forecast results and the load results obtained by actual monitoring to the cloud service layer, and performs periodic tracking and control according to the control instructions issued by the dispatching control module;

[0066] Regional operation control optimization refers to the area controlled by the edge fusion terminal, using the operation control parameters of the same period in the previous decision cycle as the control basis, the actual load and power output of the region in each time period as input, and the goal of minimizing the power loss and control cost of the distribution network in the region, to determine the optimal control scheme for the edge control terminal.

[0067] The terminal control equipment completes the specific execution of the power distribution status control in the area; the terminal collection equipment obtains and processes the monitoring or data of the distribution station area, smart substation, new energy site and power station operation data, providing a basis for the system to perceive and analyze the power grid status.

[0068] In this embodiment, the terminal control equipment includes static VAR compensators, shunt capacitors, on-load tap-changing transformers, and low-voltage regulators installed in each area; the terminal collection equipment includes distribution switches, voltage regulating equipment, inverters, sensors, and data acquisition terminals installed in each area.

[0069] As a specific implementation, the present invention also provides a regional load forecasting model, which is created based on the following method:

[0070] A. Determine the load forecasting model control target Goal = λ1Goal1 + λ2Goal2, where:

[0071] Regional power loss minimization goal

[0072] The regional control cost minimization goal Goal2 = minC0 = C1 + C2;

[0073] λ1 refers to the regional power loss decision weight, λ2 refers to the regional control cost decision weight; T refers to the data collection period, I refers to the number of regional nodes, J iIt refers to the set consisting of node j of branch ij where node i is located, Δt is the control period; C0 refers to the regional control cost, C1 refers to the control execution cost, and C2 refers to the power dispatch cost;

[0074] B. Determine the constraints of the load forecasting model

[0075]

[0076] Where D is the total number of node power-related devices, It refers to the active power injection of node device d at the i-th node at time t; It refers to the reactive power output of node device d at the i-th node at time t; It refers to the voltage amplitude of the nodes at both ends of the line in, A in , B in Refers to the real and imaginary parts of the line in node admittance matrix, It refers to the voltage phase difference between the two ends of line in at time t.

[0077] Further improvement or preferred implementation of the aforementioned cloud-edge-based low-voltage distribution network operation control system,

[0078] Where R refers to the number of control device types, K r is the total number of devices in category r, c r It refers to the cost of each level adjustment control of the r-type equipment; It refers to the device control level in the tth time period, It refers to the control level of the control device in the t+1th time period, c r It refers to the control adjustment cost of each level of the rth type of equipment.

[0079]

[0080] Where S refers to the total number of transmission and distribution nodes in the region. st It refers to the amount of electricity input through the sth node in the tth cycle, out st It refers to the total amount of electricity output through the sth node in the tth cycle, P t It refers to the electricity price in the tth period.

[0081] Based on the aforementioned control system structure, the present application also provides a low-voltage distribution network operation control method based on the cloud edge, such as Figure 2 As shown, the basic steps include:

[0082] Stp1, obtain regional power grid operation status data based on the terminal acquisition equipment in each region and upload it to the edge fusion terminal;

[0083] Stp2, the edge fusion terminal obtains the regional load forecasting model from the cloud extreme service module, and uses the collected regional power grid operation status data to predict the regional power grid load; the edge fusion terminal runs the regional operation control plan to optimize according to the current load forecast results, determines the best operation control plan under the current load forecast results and sends it to the edge control terminal for optimization control;

[0084] The edge fusion terminal periodically tracks the grid load based on the data collected by the regional terminal acquisition device, and uploads the load forecast results and real-time load monitoring values ​​to the cloud computing service module;

[0085] In specific implementation, the periodic data collection and instruction generation are based on the cycle time of the power distribution network, which refers to a cycle of one day, one week, etc. This cycle is mainly specified based on the actual cycle of regional power distribution control, and the model prediction results and real-time control results of each cycle are uploaded as feedback data, so that the cloud service layer can optimize the control model of the next cycle and control the power distribution status of the entire distribution network.

[0086] Stp3, the edge control terminal determines and executes the specific control plan according to the regional operation control plan instructions and cloud instructions;

[0087] Stp4, the cloud computing service module establishes a regional load forecasting model, optimizes and trains the load forecasting model based on the load forecast value calculated by the edge fusion terminal and the actual load monitoring value, and sends the trained load model to the edge fusion terminal;

[0088] The cloud computing service module collects the load forecast results uploaded by each region, inputs the regional power distribution parameter control scheme of each region in the power distribution area determined by the power grid coordination control model, and outputs the control parameters of the edge control terminal in the edge control layer;

[0089] The cloud computing service module periodically collects the load forecast results uploaded by each region, inputs the regional power distribution parameter control scheme of each region in the power distribution area determined by the power grid coordination control model, and periodically outputs the control parameters of the edge control terminal in the edge control layer;

[0090] Stp5, the scheduling control module extracts the output structure of the cloud computing service module, generates static control instructions and dynamic control instructions according to the control parameter type, and sends them to the edge control terminal and the edge fusion terminal respectively;

[0091] Stp6, the edge fusion terminal obtains periodic dynamic control instructions, and generates regional control instructions according to the regional power grid status and sends them to the edge control terminal. The edge control terminal obtains the regional control instructions from the edge fusion terminal and the static instructions issued by the dispatching control module, and generates the final control instruction set on the premise of preferentially satisfying the static instructions and secondly satisfying the regional control instructions;

[0092] Among them, the static control instruction refers to an instruction corresponding to a control parameter that has a certain value or can be changed arbitrarily within a static range; the dynamic instruction refers to an instruction corresponding to a control parameter that needs to rely on other parameters to determine the parameter value or parameter value range.

[0093] Stp7, the terminal control device obtains the control instruction set to complete the control of the terminal device and returns to Stp1 for continuous execution.

[0094] Finally, it should be noted that the above embodiments are only used to illustrate the technical solution of the present invention, rather than to limit the scope of protection of the present invention. Although the present invention has been described in detail with reference to the preferred embodiments, those skilled in the art should understand that the technical solution of the present invention can be modified or replaced by equivalents without departing from the essence and scope of the technical solution of the present invention.

Claims

1. A low-voltage distribution network operation control system based on cloud edge, characterized in that: Includes cloud service layer, edge service layer, and terminal device layer; The cloud service layer includes data analysis and processing modules, cloud computing service modules, and scheduling control modules; the edge control layer is composed of several edge fusion terminals and edge control terminals serving various regions; the terminal device layer is composed of terminal control devices and terminal collection devices; The analysis and processing module extracts the regional load forecast results and actual load monitoring data obtained by the edge control layer and performs necessary preprocessing to enable them to be used for cloud computing services; The cloud computing service module establishes regional load forecasting models and power grid coordination control models based on data mining and machine learning, and completes the training of the corresponding models; The cloud computing service module periodically optimizes the regional load forecasting model and sends the trained regional load forecasting model to the edge fusion terminal in the edge control layer; The grid coordination control model determines the control scheme for power distribution parameters in each area; The dispatch control module generates dispatch control instructions based on the output of the power grid coordination control model and sends them to the edge control layer; The edge fusion terminal is set up near the managed area and serves as an edge computing resource node to provide computing resource support for the cloud service end and perform collaborative computing with other edge computing resource nodes. The edge control terminal completes the coordinated control of the distribution network and optimizes regional operation control. The terminal control equipment completes the specific execution of the power distribution status control in the area; the terminal collection equipment obtains and processes the monitoring or data of the distribution station area, smart substation, new energy site and power station operation data, providing a basis for the system to perceive and analyze the power grid status.

2. According to the cloud-edge-end based low-voltage distribution network operation control system of claim 1, it is characterized in that: The regional load forecasting model specifically aims to minimize the power loss and control cost of the regional distribution network and determine the regional distribution parameter control plan; The cloud computing service module periodically trains and optimizes the regional load forecasting model based on the regional load forecasting results and the actual load monitoring data, and sends the trained regional load forecasting model to the edge fusion terminal in the edge control layer; The grid coordination control model aims to minimize power loss and maximize energy utilization in the distribution network. It uses the regional load forecast results uploaded by the edge fusion terminal in the edge control layer as input, determines the regional distribution parameter control scheme for each area in the distribution area, and outputs the control parameters of the edge control terminal in the edge control layer. The dispatching control module extracts and classifies the control parameter outputs of each edge control terminal output by the power grid coordination control model, and generates corresponding dispatching control instructions, sending direct parameter control instructions to the edge control terminal and interval dynamic parameter control instructions to the edge fusion terminal.

3. The cloud-edge-based low-voltage distribution network operation control system according to claim 2 is characterized in that: Distribution network collaborative control means that the edge fusion terminal obtains the trained regional load forecasting model from the cloud computing service module, uses the monitoring data collected by the terminal device layer to realize regional load forecasting, determines the optimal control scheme of the edge control terminal in the corresponding area and executes it, uploads the corresponding load forecast results and the load results obtained by actual monitoring to the cloud service layer, and performs periodic tracking and control according to the control instructions issued by the dispatching control module; Regional operation control optimization refers to the area controlled by the edge fusion terminal, using the operation control parameters of the same period in the previous decision cycle as the control basis, the actual load and power output of the region in each time period as input, and the goal of minimizing the power loss and control cost of the distribution network in the region, to determine the optimal control scheme for the edge control terminal.

4. The cloud-edge-based low-voltage distribution network operation control system according to claim 1 is characterized in that: The terminal control equipment includes static VAR compensators, shunt capacitors, on-load tap-changing transformers, and low-voltage regulators installed in each area; the terminal collection equipment includes distribution switches, voltage regulating equipment, inverters, sensors, and data acquisition terminals installed in each area.

5. The cloud-edge-based low-voltage distribution network operation control system according to claim 1 is characterized in that: The regional load forecasting model is created based on the following method: A. Determine the load forecasting model control target Goal = λ1Goal1 + λ2Goal2, where: Regional power loss minimization goal The regional control cost minimization goal Goal2 = minC0 = C1 + C2; λ1 refers to the regional power loss decision weight, λ2 refers to the regional control cost decision weight; T refers to the data collection period, I refers to the number of regional nodes, J i It refers to the set consisting of node j of branch ij where node i is located, Δt is the control period; C0 refers to the regional control cost, C1 refers to the control execution cost, and C2 refers to the power dispatch cost; B. Determine the constraints of the load forecasting model Where D is the total number of node power-related devices, It refers to the active power injection of node device d at the i-th node at time t; It refers to the reactive power output of node device d at the i-th node at time t; It refers to the voltage amplitude of the nodes at both ends of the line in, A in , B in Refers to the real and imaginary parts of the line in node admittance matrix, It refers to the voltage phase difference between the two ends of line in at time t.

6. The cloud-edge-based low-voltage distribution network operation control system according to claim 5 is characterized in that: Where R refers to the number of control device types, K r is the total number of devices in category r, c r It refers to the cost of each level adjustment control of the r-type equipment; It refers to the device control level in the tth time period, It refers to the control level of the control device in the t+1th time period, c r It refers to the control adjustment cost of each level of equipment in the rth category; Where S refers to the total number of transmission and distribution nodes in the region. st It refers to the amount of electricity input through the sth node in the tth cycle, out st It refers to the total amount of electricity output through the sth node in the tth cycle, P t It refers to the electricity price in the tth period.

7. The cloud-edge-based low-voltage distribution network operation control method according to claim 1 is characterized in that: The steps include: Stp1, obtain regional power grid operation status data based on the terminal acquisition equipment in each region and upload it to the edge fusion terminal; Stp2, the edge fusion terminal obtains the regional load prediction model from the cloud extreme service module and uses the collected regional power grid operation status data to predict the regional power grid load; The edge fusion terminal optimizes the operation control scheme of the operation area according to the current load forecast results, determines the best operation control scheme under the current load forecast results and sends it to the edge control terminal for optimization control; The edge fusion terminal periodically tracks the grid load based on the data collected by the regional terminal acquisition device, and uploads the load forecast results and real-time load monitoring values ​​to the cloud computing service module; Stp3, the edge control terminal determines and executes the specific control plan according to the regional operation control plan instructions and cloud instructions; Stp4, the cloud computing service module establishes a regional load forecasting model, optimizes and trains the load forecasting model based on the load forecast value calculated by the edge fusion terminal and the actual load monitoring value, and sends the trained load model to the edge fusion terminal; The cloud computing service module collects the load forecast results uploaded by each region, inputs the regional power distribution parameter control scheme of each region in the power distribution area determined by the power grid coordination control model, and outputs the control parameters of the edge control terminal in the edge control layer; The cloud computing service module periodically collects the load forecast results uploaded by each region, inputs the regional power distribution parameter control scheme of each region in the power distribution area determined by the power grid coordination control model, and periodically outputs the control parameters of the edge control terminal in the edge control layer; Stp5, the scheduling control module extracts the output structure of the cloud computing service module, generates static control instructions and dynamic control instructions according to the control parameter type, and sends them to the edge control terminal and the edge fusion terminal respectively; Stp6, the edge fusion terminal obtains periodic dynamic control instructions, and generates regional control instructions according to the regional power grid status and sends them to the edge control terminal. The edge control terminal obtains the regional control instructions from the edge fusion terminal and the static instructions issued by the dispatching control module, and generates the final control instruction set on the premise of preferentially satisfying the static instructions and secondly satisfying the regional control instructions; Stp7, the terminal control device obtains the control instruction set to complete the control of the terminal device and returns to Stp1 for continuous execution.

8. The cloud-edge-based low-voltage distribution network operation control method according to claim 7 is characterized in that: The static control instruction refers to an instruction corresponding to a control parameter that can be arbitrarily changed within a fixed value or static range; the dynamic instruction refers to an instruction corresponding to a control parameter that needs to rely on other parameters to determine the parameter value or parameter value range.

9. The cloud-edge-based low-voltage distribution network operation control method according to claim 7 is characterized in that: The periodic data collection and instruction generation are based on a cycle determined based on the power supply and distribution cycle time of the power distribution network, and the cycle refers to a cycle of one day, one week, etc.

Images