Electric vehicle group ordered charging system based on dynamic regulation of regional power grid load
The orderly charging system for electric vehicle fleets, which utilizes multi-level distributed communication and cloud-edge collaborative computing, solves the problems of charging status monitoring and uneven power distribution when large-scale electric vehicles are connected, and achieves fast and accurate charging management and grid load optimization.
Patent Information
- Application Number
- CN202511967642.3
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2025-12-24
- Publication Date
- 2026-02-24
AI Technical Summary
When existing electric vehicle charging systems are connected on a large scale, it is difficult to achieve fast and accurate monitoring of charging status and power allocation, resulting in network latency, chaotic charging queues, and uneven power distribution.
It employs a multi-level distributed communication module, a cloud-edge collaborative computing and scheduling module, a load and demand forecasting module, and a personalized orderly charging strategy module. Combined with the dynamic adjustment of regional power grid load, it achieves real-time data transmission and optimized charging strategies, ensuring timely and accurate control commands, and performs differentiated management based on user needs and power grid constraints.
By leveraging edge-layer autonomy and cloud-based collaborative computing, communication bandwidth requirements are reduced, latency and computational complexity are decreased, the reliable execution of critical instructions is ensured, charging sequence and power allocation are optimized, and the problems of chaotic charging queues and uneven power distribution are solved.
Smart Images
Figure CN121566532A_ABST
Abstract
Description
Technical Field
[0001] This invention relates to the field of electric vehicle charging technology, specifically to an orderly charging system for electric vehicle groups based on dynamic adjustment of regional power grid load. Background Technology
[0002] With the promotion and technological advancement of electric vehicles, developing electric vehicles is an important means of energy conservation and emission reduction, absorbing renewable energy, and developing a low-carbon economy. To meet the increasing demand for electric vehicle travel, the construction of electric vehicle charging facilities will see long-term development in the future, and these facilities will need to be paired with corresponding orderly charging systems for electric vehicle groups during their use.
[0003] However, existing charging systems still have the following technical problems in use:
[0004] After a large number of electric vehicles are connected, the system needs to monitor the charging status, remaining power, and grid load changes of each vehicle in real time. The transmission of massive amounts of data places extremely high demands on the bandwidth of the communication network. Under the traditional centralized control architecture, network latency can easily lead to delays in control commands.
[0005] Dynamic adjustment requires real-time optimization of charging strategies in conjunction with grid load fluctuations, and each strategy must take into account the individual needs of hundreds or thousands of vehicles and regional grid constraints. When the density of charging vehicles increases significantly, existing computing units struggle to quickly complete complex scheduling calculations, easily leading to problems such as chaotic charging queues and uneven power distribution.
[0006] Therefore, an orderly charging system for electric vehicle fleets based on dynamic adjustment of regional power grid load is proposed to address the aforementioned problems. Summary of the Invention
[0007] The purpose of this invention is to provide an orderly charging system for electric vehicle fleets based on dynamic adjustment of regional power grid load, so as to solve the problems mentioned in the background art.
[0008] To achieve the above objectives, the present invention provides the following technical solution: an orderly charging system for electric vehicle groups based on dynamic adjustment of regional power grid load, comprising a multi-level distributed communication module, a cloud-edge collaborative computing and scheduling module, a load and demand forecasting module, a personalized orderly charging strategy module, and a charging pile cluster management and control module;
[0009] The multi-level distributed communication module is responsible for bidirectional data transmission between the vehicle, charging pile and the system, and collects key parameters such as charging status, remaining power, and grid load in real time to ensure that control commands are accurately and timely delivered to the charging equipment.
[0010] The cloud-edge collaborative computing and scheduling module analyzes charging demand and grid constraints in real time, generates the optimal charging strategy, dynamically adjusts the power allocation of each charging pile, achieves peak shaving and valley filling, and handles users' personalized charging needs.
[0011] The load and demand forecasting module predicts the future load curve of the regional power grid and estimates the charging demand distribution of electric vehicle fleets, providing a forward-looking basis for scheduling decisions.
[0012] The personalized and orderly charging strategy module analyzes user charging needs, formulates differentiated charging strategies, balances user needs with grid constraints, manages charging queues, and optimizes charging order.
[0013] The charging pile cluster management and control module monitors the operating status of the charging piles in real time, controls the start and stop of the charging piles and adjusts the power, aggregates the charging pile load information in the area, and feeds it back to the upper-level system.
[0014] Preferably, the multi-level distributed communication module includes an edge communication terminal cluster, an on-board unit (OBU), a charging pile communication submodule, a communication management server, and a distributed cache unit;
[0015] The edge communication terminal cluster is deployed in the community charging station and is responsible for the collection of charging pile data and instruction distribution in the area. It has a built-in 5G communication module and wired communication interface.
[0016] The on-board communication unit (OBU) is integrated into the vehicle and connected to the BMS via the CAN bus to collect key parameters such as SOC and charging status in real time.
[0017] The charging pile communication submodule has built-in RS485 and Ethernet interfaces to realize bidirectional data interaction between the charging pile and the edge terminal.
[0018] The communication management server is deployed in the cloud and is responsible for managing communication links, data routing, and priority control.
[0019] The distributed cache unit is deployed on edge nodes to store critical instructions and configuration information, preventing control failure due to communication interruptions.
[0020] Preferably, the cloud-edge collaborative computing scheduling module includes edge computing nodes, cloud computing center, distributed database and intelligent scheduling algorithm library;
[0021] The edge computing nodes are deployed on local servers in the community and charging station, and are configured with lightweight CPU+GPU to be responsible for vehicle scheduling and power allocation within the area; the cloud computing center adopts a distributed cluster architecture and is responsible for global load prediction, cross-regional coordination and complex optimization calculations; the distributed database stores real-time status data on the edge side and historical data and strategy models in the cloud.
[0022] The intelligent scheduling algorithm library includes heuristic algorithms and precise optimization algorithms.
[0023] Preferably, the load and demand forecasting module includes a power grid load acquisition unit, a vehicle demand acquisition unit, a forecasting model training module, a multi-dimensional feature extractor, and a forecasting result evaluation unit.
[0024] The power grid load acquisition unit connects to smart meters and PMUs to monitor the regional power grid load in real time; the vehicle demand acquisition unit obtains user charging reservation times and personalized demand for target SOCs through OBUs; the prediction model training module trains LSTM / CNN deep learning models based on historical data; the multi-dimensional feature extractor integrates weather, weekday, holiday, and electricity price influencing factors to improve prediction accuracy; and the prediction result evaluation unit uses MAE / RMSE indicators to evaluate prediction accuracy and automatically triggers model updates.
[0025] Preferably, the personalized orderly charging strategy module includes a user demand parser, a charging strategy generator, a charging queue manager, and a strategy execution monitor;
[0026] The user demand parser analyzes the user's reservation time, target SOC, and priority information to generate a demand vector; the charging strategy generator generates differentiated charging strategies based on the demand vector and grid constraints; the charging queue manager maintains a charging priority queue, optimizes the charging order, and reduces waiting time; and the strategy execution monitor tracks the strategy execution in real time and makes dynamic adjustments when necessary.
[0027] Preferably, the charging pile cluster management module includes a charging pile intelligent controller, a cluster coordinator, a status monitoring unit, a security protection unit, and a local autonomy module;
[0028] The intelligent controller for charging piles has a built-in microprocessor responsible for single-pile status monitoring, power control, and protection; the cluster coordinator manages multiple charging piles within the area, enabling power aggregation and unified scheduling; the status monitoring unit collects charging pile voltage, current, temperature, and charging progress parameters in real time; the safety protection unit detects abnormal situations, immediately cuts off the power supply, and reports the issue; and the local autonomous module executes a preset safety strategy to maintain basic functions when communication is interrupted.
[0029] Compared with the prior art, the beneficial effects of the present invention are:
[0030] 1) This application filters most of the redundant data at each level of charging pile, edge, and cloud, and only uploads core information, so that 80% of the data processing is completed at the edge and only key decision information is uploaded to the cloud, which greatly reduces the bandwidth requirements. Communication priority is divided according to the order of control commands, device status, and charging data to ensure that key commands are transmitted first with a latency of <100ms, thus solving the problem of extremely high communication bandwidth requirements for massive data transmission.
[0031] 2) By having local autonomy at the edge layer, it can make independent decisions when communication is interrupted, with a response time of <100ms. It can generate strategies in advance based on prediction results and adjust them before load fluctuations, reducing the need for real-time response. Both the edge and the charging pile are equipped with command caches to ensure reliable execution of commands and prevent communication jitter from affecting the system. This solves the problem of network latency causing control command delays in traditional centralized architectures.
[0032] 3) The edge layer is responsible for scheduling within the region, while the cloud focuses on global optimization. The processing capacity expands linearly, and multiple edge nodes work simultaneously without interfering with each other. The total system processing capacity increases with the number of edge nodes. The edge layer uses lightweight heuristic algorithms for fast decision-making, while complex optimizations are handled asynchronously by the cloud, solving the problem that the computing unit is unable to complete complex scheduling calculations when the density of charging vehicles increases.
[0033] 4) Based on factors such as the urgency of user needs and reservation time, different priorities are assigned to vehicles to ensure emergency vehicle use. Priority is given to allocating the nearest available charging pile to reduce vehicle travel distance. Based on predicted load and user reservations, the optimal charging time period is allocated, and the charging power for non-emergency use is reduced to 50-80% of the rated power to ensure grid safety and solve the problems of chaotic charging queues and uneven power distribution. Attached Figure Description
[0034] Figure 1 This is a diagram showing the module composition of this system;
[0035] Figure 2 This is a diagram showing the unit composition of a multi-level distributed communication module.
[0036] Figure 3 This is a diagram showing the unit composition of the cloud-edge collaborative computing scheduling module;
[0037] Figure 4 This is a diagram showing the unit composition of the load and demand forecasting module.
[0038] Figure 5 A diagram showing the unit composition of a personalized orderly charging strategy module;
[0039] Figure 6 This is a diagram showing the unit composition of the charging pile cluster management and control module. Detailed Implementation
[0040] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.
[0041] Example:
[0042] Please see Figure 1-6 The present invention provides a technical solution:
[0043] The electric vehicle fleet orderly charging system based on regional power grid load dynamic adjustment includes a multi-level distributed communication module, a cloud-edge collaborative computing and scheduling module, a load and demand forecasting module, a personalized orderly charging strategy module, and a charging pile cluster management and control module.
[0044] The multi-level distributed communication module is responsible for bidirectional data transmission between the vehicle, charging pile and the system, and collects key parameters such as charging status, remaining power, and grid load in real time to ensure that control commands are accurately and timely delivered to the charging equipment.
[0045] The cloud-edge collaborative computing and scheduling module analyzes charging demand and grid constraints in real time, generates the optimal charging strategy, dynamically adjusts the power allocation of each charging pile, achieves peak shaving and valley filling, and handles users' personalized charging needs.
[0046] The load and demand forecasting module predicts the future load curve of the regional power grid and estimates the charging demand distribution of electric vehicle fleets, providing a forward-looking basis for scheduling decisions.
[0047] The personalized and orderly charging strategy module analyzes user charging needs, formulates differentiated charging strategies, balances user needs with grid constraints, manages charging queues, and optimizes charging order.
[0048] The charging pile cluster management and control module monitors the operating status of the charging piles in real time, controls the start and stop of the charging piles and adjusts the power, aggregates the charging pile load information in the area, and feeds it back to the upper-level system.
[0049] The multi-level distributed communication module includes an edge communication terminal cluster, an on-board communication unit (OBU), a charging pile communication sub-module, a communication management server, and a distributed cache unit.
[0050] The edge communication terminal cluster is deployed in community charging stations, responsible for data aggregation and instruction distribution of charging piles within the area, and has a built-in 5G communication module and wired communication interface; the on-board unit (OBU) is integrated into the vehicle and connected to the BMS via CAN bus to collect key parameters such as SOC and charging status in real time; the charging pile communication submodule has built-in RS485 and Ethernet interfaces to realize bidirectional data interaction between the charging pile and the edge terminal; the communication management server is deployed in the cloud and is responsible for managing communication links, data routing, and priority control; the distributed cache unit is deployed on the edge node to store key instructions and configuration information to prevent control failure due to communication interruption.
[0051] Edge terminals first filter invalid data and only upload key parameters (SOC, fault information), reducing backbone network traffic by 80%. A four-level communication link is adopted, consisting of charging piles, edge terminals, regional coordinators, and the cloud, to avoid single points of failure. Control commands are assigned the highest priority to ensure transmission latency <100ms; monitoring data is secondary, with a latency <500ms. When 5G communication is interrupted, it automatically switches to the wired communication link between adjacent charging piles to ensure basic control.
[0052] The cloud-edge collaborative computing scheduling module includes edge computing nodes, cloud computing centers, distributed databases, and intelligent scheduling algorithm libraries;
[0053] The edge computing nodes are deployed on local servers in the community and charging station, and are configured with lightweight CPU+GPU to be responsible for vehicle scheduling and power allocation within the area; the cloud computing center adopts a distributed cluster architecture and is responsible for global load prediction, cross-regional coordination and complex optimization calculations; the distributed database stores real-time status data on the edge side and historical data and strategy models in the cloud.
[0054] The intelligent scheduling algorithm library includes heuristic algorithms for rapid decision-making at the edge and precise optimization algorithms for deep optimization in the cloud.
[0055] Multiple edge nodes can process their respective regional tasks simultaneously. The computing power scales linearly with the number of nodes. Only vehicles with changing states are recalculated, avoiding global re-optimization and saving 70% of computing resources. High-priority vehicles are allocated computing resources first to ensure service quality.
[0056] The load and demand forecasting module includes a power grid load acquisition unit, a vehicle demand acquisition unit, a forecasting model training module, a multi-dimensional feature extractor, and a forecasting result evaluation unit.
[0057] The power grid load acquisition unit connects to smart meters and PMUs to monitor the regional power grid load in real time; the vehicle demand acquisition unit obtains user charging reservation times and personalized demand for target SOCs through OBUs; the prediction model training module trains LSTM / CNN deep learning models based on historical data; the multi-dimensional feature extractor integrates weather, weekday, holiday, and electricity price influencing factors to improve prediction accuracy; and the prediction result evaluation unit uses MAE / RMSE indicators to evaluate prediction accuracy and automatically triggers model updates.
[0058] Two hours before the load peak is predicted, the charging plan is pre-adjusted at the edge layer to reduce the frequency of real-time adjustments and push charging suggestions to users during low-load periods (such as when electricity prices are low) to disperse charging demand. This can reduce peak load by 30%. Based on the prediction results, computing resources and communication bandwidth are allocated in advance to avoid temporary resource competition. By adopting the forgetting factor mechanism, the model's response time to new load patterns is shortened to 15 minutes.
[0059] Prediction model output formula:
[0060]
[0061] in:
[0062] : The predicted total load of the regional power grid at time t (kW);
[0063] The input feature vector at time tk;
[0064] n: Length of the feature time window;
[0065] LSTM model parameters;
[0066] If the charging demand forecast for electric vehicle fleets is to be generated, it can be replaced with:
[0067]
[0068] in:
[0069] The charging demand of the i-th vehicle at time t;
[0070] : Initial remaining battery power (%) of vehicle i;
[0071] : Scheduled charging time (h) for vehicle i;
[0072] Battery capacity (kWh) of vehicle i;
[0073] : CNN model training parameters.
[0074] The personalized orderly charging strategy module includes a user demand parser, a charging strategy generator, a charging queue manager, and a strategy execution monitor.
[0075] The user demand parser analyzes the user's reservation time, target SOC, and priority information to generate a demand vector; the charging strategy generator generates differentiated charging strategies based on the demand vector and grid constraints; the charging queue manager maintains a charging priority queue, optimizes the charging order, and reduces waiting time; and the strategy execution monitor tracks the strategy execution in real time and makes dynamic adjustments when necessary.
[0076] Prioritize the allocation of the nearest available charging station to reduce vehicle travel distance. Match the time window and allocate the best charging period according to the user's reservation time and predicted load. When SOC>80%, automatically reduce the charging power to extend battery life. When the grid load is close to the threshold, reduce non-emergency charging power in batches (reduction ≤30%) to avoid overload.
[0077] The charging pile cluster management and control module includes a charging pile intelligent controller, a cluster coordinator, a status monitoring unit, a security protection unit, and a local autonomy module;
[0078] The intelligent controller for charging piles has a built-in microprocessor responsible for single-pile status monitoring, power control, and protection; the cluster coordinator manages multiple charging piles within the area, enabling power aggregation and unified scheduling; the status monitoring unit collects charging pile voltage, current, temperature, and charging progress parameters in real time; the safety protection unit detects abnormal situations, immediately cuts off the power supply, and reports the issue; and the local autonomous module executes a preset safety strategy to maintain basic functions when communication is interrupted.
[0079] Instead of controlling each charging station individually, a unified power adjustment command is sent to the entire cluster, reducing the number of commands by 90%. The status of charging stations is uploaded in real time, reducing the frequency of system queries, lowering the communication burden, accurately calculating the total charging load in the area, and providing accurate data for upper-level decision-making. In the event of a communication interruption, the system operates according to a preset strategy (such as maintaining the current power or reducing to the minimum safe power) to avoid system crashes.
[0080] The foregoing has shown and described the basic principles, main features, and advantages of the present invention. It will be apparent to those skilled in the art that the present invention is not limited to the details of the exemplary embodiments described above, and that the invention can be implemented in other specific forms without departing from its spirit or basic characteristics. Therefore, the embodiments should be considered exemplary and non-limiting in all respects. The scope of the invention is defined by the appended claims rather than the foregoing description. Therefore, all variations falling within the meaning and scope of equivalents of the claims are intended to be included within the present invention, and no reference numerals in the claims should be construed as limiting the scope of the claims.
[0081] Although embodiments of the invention have been shown and described, it will be understood by those skilled in the art that various changes, modifications, substitutions and alterations can be made to these embodiments without departing from the principles and spirit of the invention, the scope of which is defined by the appended claims and their equivalents.
Claims
1. An orderly charging system for electric vehicle fleets based on dynamic adjustment of regional power grid load, characterized in that, It includes a multi-level distributed communication module, a cloud-edge collaborative computing and scheduling module, a load and demand forecasting module, a personalized orderly charging strategy module, and a charging pile cluster management and control module; The multi-level distributed communication module is responsible for bidirectional data transmission between the vehicle, charging pile and the system, and collects key parameters such as charging status, remaining power, and grid load in real time to ensure that control commands are accurately and timely delivered to the charging equipment. The cloud-edge collaborative computing and scheduling module analyzes charging demand and grid constraints in real time, generates the optimal charging strategy, dynamically adjusts the power allocation of each charging pile, achieves peak shaving and valley filling, and handles users' personalized charging needs. The load and demand forecasting module predicts the future load curve of the regional power grid and estimates the charging demand distribution of electric vehicle fleets, providing a forward-looking basis for scheduling decisions. The personalized and orderly charging strategy module analyzes user charging needs, formulates differentiated charging strategies, balances user needs with grid constraints, manages charging queues, and optimizes charging order. The charging pile cluster management and control module monitors the operating status of the charging piles in real time, controls the start and stop of the charging piles and adjusts the power, aggregates the charging pile load information in the area, and feeds it back to the upper-level system.
2. The electric vehicle fleet orderly charging system based on dynamic adjustment of regional power grid load according to claim 1, characterized in that: The multi-level distributed communication module includes an edge communication terminal cluster, an on-board communication unit (OBU), a charging pile communication sub-module, a communication management server, and a distributed cache unit. The edge communication terminal cluster is deployed in the community charging station and is responsible for the collection of charging pile data and instruction distribution in the area. It has a built-in 5G communication module and wired communication interface. The on-board communication unit (OBU) is integrated into the vehicle and connected to the BMS via the CAN bus to collect key parameters such as SOC and charging status in real time. The charging pile communication submodule has built-in RS485 and Ethernet interfaces to realize bidirectional data interaction between the charging pile and the edge terminal. The communication management server is deployed in the cloud and is responsible for managing communication links, data routing, and priority control. The distributed cache unit is deployed on edge nodes to store critical instructions and configuration information, preventing control failure due to communication interruptions.
3. The electric vehicle fleet orderly charging system based on regional power grid load dynamic adjustment according to claim 1, characterized in that: The cloud-edge collaborative computing scheduling module includes edge computing nodes, cloud computing centers, distributed databases, and intelligent scheduling algorithm libraries; The edge computing nodes are deployed on local servers in the community and charging station, and are configured with lightweight CPU+GPU to be responsible for vehicle scheduling and power allocation within the area; the cloud computing center adopts a distributed cluster architecture and is responsible for global load prediction, cross-regional coordination and complex optimization calculations; the distributed database stores real-time status data on the edge side and historical data and strategy models in the cloud. The intelligent scheduling algorithm library includes heuristic algorithms and precise optimization algorithms.
4. The electric vehicle fleet orderly charging system based on regional power grid load dynamic adjustment according to claim 1, characterized in that: The load and demand forecasting module includes a power grid load acquisition unit, a vehicle demand acquisition unit, a forecasting model training module, a multi-dimensional feature extractor, and a forecasting result evaluation unit. The power grid load acquisition unit is connected to a smart meter and a PMU to monitor the regional power grid load in real time. The vehicle demand collection unit obtains the user's charging reservation time and the target SOC personalized demand through the OBU. The prediction model training module trains an LSTM / CNN deep learning model based on historical data; the multi-dimensional feature extractor integrates weather, weekday, holiday, and electricity price influencing factors to improve prediction accuracy; the prediction result evaluation unit uses MAE / RMSE indicators to evaluate prediction accuracy and automatically triggers model updates.
5. The electric vehicle fleet orderly charging system based on regional power grid load dynamic adjustment according to claim 1, characterized in that: The personalized orderly charging strategy module includes a user demand parser, a charging strategy generator, a charging queue manager, and a strategy execution monitor. The user demand parser analyzes the user's reservation time, target SOC, and priority information to generate a demand vector; the charging strategy generator generates differentiated charging strategies based on the demand vector and grid constraints; and the charging queue manager maintains a charging priority queue, optimizes the charging order, and reduces waiting time. The strategy execution monitor tracks the strategy execution status in real time and makes dynamic adjustments as needed.
6. The electric vehicle fleet orderly charging system based on regional power grid load dynamic adjustment according to claim 1, characterized in that: The charging pile cluster management and control module includes a charging pile intelligent controller, a cluster coordinator, a status monitoring unit, a security protection unit, and a local autonomy module; The intelligent controller for charging piles has a built-in microprocessor responsible for single-pile status monitoring, power control, and protection; the cluster coordinator manages multiple charging piles within the area, enabling power aggregation and unified scheduling; the status monitoring unit collects charging pile voltage, current, temperature, and charging progress parameters in real time; the safety protection unit detects abnormal situations, immediately cuts off the power supply, and reports the issue; and the local autonomous module executes a preset safety strategy to maintain basic functions when communication is interrupted.