Optical storage and charging integrated intelligent scheduling system based on digital twinning

Abstract

The application discloses a kind of light storage and charging integrated intelligent scheduling systems based on digital twinning. Including photovoltaic power generation module, energy storage module, charging module, grid interaction module, intelligent control module, digital twin module and communication module;Intelligent control module: integrated model predictive control algorithm and deep reinforcement learning algorithm, for receiving the real-time data transmitted by each module, data processing, prediction analysis and dispatching strategy generation;Digital twin module: for integrating the real-time operation data and historical data of each module, realizing the visual monitoring of system operation state, fault early warning and simulation.This application introduces model predictive control and deep reinforcement learning algorithm, clear algorithm parameters and control logic, realize the accurate prediction of photovoltaic power generation power, energy storage remaining power and charging load, dynamically generate optimal scheduling strategy, solve the problem of non-intelligent system scheduling and low energy utilization, improve the intelligent scheduling level of system.

Description

Technical Field

[0001] This invention relates to an intelligent scheduling system for integrated photovoltaic, energy storage, and charging based on digital twins. Background Technology

[0002] With the rapid development of the new energy industry, integrated photovoltaic-storage-charging systems, as a fusion of photovoltaic power generation, energy storage technology, and charging technology, have become an important carrier for solving the problems of new energy consumption, grid load regulation, and electric vehicle charging needs. Currently, existing integrated photovoltaic-storage-charging systems mostly use fixed logic control for scheduling, which has the following technical pain points: First, the level of scheduling intelligence is low, making it impossible to accurately predict the dynamic changes in photovoltaic power generation, remaining energy storage capacity, and charging load, resulting in unreasonable energy allocation and low energy utilization. Second, the coordination among photovoltaic, energy storage, and charging is poor, failing to fully integrate peak-valley electricity price signals to optimize energy flow, making it difficult to achieve the triple goals of peak shaving and valley filling, new energy consumption, and optimal economic benefits. Third, the system has insufficient compatibility, lacking standardized interfaces and protocols, making seamless integration with the distribution network and the source-grid-load-storage system impossible, and resulting in weak multi-energy complementarity. Fourth, operation and maintenance management is rudimentary, lacking real-time monitoring and fault warning mechanisms for system operation status, making it prone to equipment failures leading to system downtime and affecting operational stability. Summary of the Invention

[0003] To address the shortcomings of the existing technologies, the present invention aims to provide a digital twin-based integrated intelligent scheduling system for photovoltaic, energy storage, and charging, which features accurate prediction, intelligent scheduling, multi-system collaboration, and intelligent operation and maintenance. This system aims to solve the problems of unintelligent scheduling, poor collaboration, insufficient compatibility, and low reliability in existing systems.

[0004] To overcome the above-mentioned shortcomings, the present invention provides a photovoltaic-storage-charging integrated intelligent dispatching system based on digital twins, comprising a photovoltaic power generation module, an energy storage module, a charging module, a grid interaction module, an intelligent control module, a digital twin module, and a communication module. Each module achieves data interaction and command transmission through the communication module. Photovoltaic power generation module: used to collect solar energy and convert it into electrical energy, while collecting the operating parameters of photovoltaic modules (including light intensity, power generation, and module temperature) in real time and transmitting them to the intelligent control module; Energy storage module: Used to store excess electrical energy generated by photovoltaic power generation module and electrical energy during off-peak hours of the power grid. At the same time, it releases electrical energy when photovoltaic power generation is insufficient or charging load is high, and collects the operating parameters of the energy storage battery pack in real time (including remaining power, charging and discharging power, battery temperature, and health status). Charging module: Used to provide charging services for electric vehicles, collect the operating parameters of each charging pile (including charging power, charging current, charging time, number of vehicles to be charged), and receive instructions from the intelligent control module to adjust the charging power and realize the dynamic allocation of charging load. Grid Interaction Module: Used to realize power interaction and data communication between the system and the distribution network, including grid interface, power metering unit and grid connection control unit; power metering unit is used to measure the amount of power exchange between the system and the grid (including power to the grid and power off the grid), and grid connection control unit is used to control the grid connection / off-grid switching between the system and the grid, and at the same time collect the real-time parameters of the grid (including grid voltage, frequency, peak and valley electricity price signals) and transmit them to the intelligent control module; Intelligent control module: Integrates model predictive control (MPC) algorithm and deep reinforcement learning algorithm, used to receive real-time data transmitted from each module, perform data processing, predictive analysis and generate scheduling strategies; Digital twin module: Used to integrate real-time and historical data from various modules to achieve visualized monitoring of system operation status, fault early warning, and simulation.

[0005] Furthermore, the intelligent control module includes: Data processing unit: performs noise reduction, filtering and integration of operating parameters transmitted by photovoltaic power generation module, energy storage module, charging module and grid interaction module to form standardized data; noise reduction adopts Kalman filtering algorithm (Q=1e-6, R=1e-3, initial variance P0=1), and the data sampling frequency is 10Hz; Predictive Analysis Unit: Based on a deep reinforcement learning algorithm (using the DQN algorithm, with a neural network structure of 12 neurons in the input layer, 2 hidden layers (64 neurons / layer), and 3 neurons in the output layer, a learning rate of 0.001, a discount factor of 0.95, and an experience replay buffer size of 10000), combined with 12 months of historical operating data, real-time solar irradiance data, grid peak-valley electricity price signals, and charging load data, it accurately predicts photovoltaic power generation, remaining energy storage capacity, and charging load demand for the next 1-24 hours, with a prediction error ≤5%. The scheduling strategy generation unit is based on the Model Predictive Control (MPC) algorithm (24h prediction time domain, 1h control time domain, constraints: energy storage charging and discharging power ≤200kW (configurable), total power of charging modules ≤820kW (configurable), objective function: maximizing photovoltaic absorption rate + minimizing grid interaction cost). Combining prediction data and system operation objectives (prioritizing photovoltaic absorption, balancing energy storage lifespan, and optimizing grid interaction cost), it dynamically adjusts the energy flow path, generates optimal scheduling instructions, and controls the operating status of each module. Coordinated control unit: Linked to peak and valley electricity price signals, it controls the charging of energy storage modules during off-peak hours and the discharging of energy storage modules during peak hours to achieve "peak shaving and valley filling"; at the same time, it coordinates the operation of photovoltaic power generation modules, energy storage modules and charging modules, giving priority to using photovoltaic power to charge electric vehicles, storing excess power in energy storage modules, and supplementing power supply from energy storage modules or the grid when photovoltaic power is insufficient, achieving deep coordination of "photovoltaic priority, storage regulation and grid complementarity".

[0006] Furthermore, the aforementioned visualization monitoring uses a digital twin model to display the operating parameters, energy flow paths, and overall system operating status of each module in real time, making it easy for staff to intuitively grasp the system's operating status; the data refresh rate is 1 second / time, and it supports multi-view switching.

[0007] Furthermore, the aforementioned fault warning: based on real-time operating data and fault diagnosis algorithm (using BP neural network algorithm, with the input being the operating parameters of each module and the output being the fault type and level), it identifies potential faults in each module (such as photovoltaic module degradation, thermal runaway of energy storage battery, and charging pile faults), promptly issues warning signals and pushes fault handling suggestions, with a warning response time ≤10s; Furthermore, the simulation is based on a digital twin model to simulate the system's operating state under different lighting conditions, charging loads, and grid electricity prices, with a simulation step size of 1 minute, providing data support for scheduling strategy optimization.

[0008] Furthermore, the communication module adopts a 5G industrial module to realize data interaction and command transmission between modules, with a transmission delay of ≤100ms.

[0009] Furthermore, the energy storage module also includes a thermal runaway protection unit, which adopts supramolecular full immersion energy storage technology and combines real-time battery temperature monitoring data. When the battery temperature exceeds a preset threshold, it automatically activates cooling protection to suppress battery thermal runaway and improve system operational safety.

[0010] Furthermore, the present invention includes several AC charging piles, DC charging piles, and a charging control unit; The charging control unit supports multi-mode charging (including constant current charging, constant voltage charging, and intelligent fast charging). It can automatically switch charging modes according to the battery type and charging needs of the electric vehicle, combined with the scheduling instructions of the intelligent control module. The constant current charging current range is 0-80A, the constant voltage charging voltage range is 200-750V, and the charging efficiency in intelligent fast charging mode is ≥95%, improving charging efficiency and battery life.

[0011] Furthermore, the intelligent control module also includes a parameter optimization unit, which can adjust the parameters of the model predictive control algorithm and the deep reinforcement learning algorithm (such as the prediction time domain of MPC and the learning rate of DQN) based on the actual project operation experience, optimize the adaptability of the scheduling strategy, and improve the system operating efficiency.

[0012] Compared with the prior art, the present invention has the following advantages: 1. High level of intelligence: By introducing model predictive control (MPC) and deep reinforcement learning algorithms, the algorithm parameters and control logic are clearly defined, enabling accurate prediction of photovoltaic power generation, remaining energy storage capacity and charging load (prediction error ≤5%), dynamically generating the optimal scheduling strategy, solving the problems of unintelligent scheduling and low energy utilization in the existing system, and improving the level of intelligent scheduling of the system; 2. Strong synergy: It achieves deep synergy between photovoltaics, energy storage, charging and grid, and optimizes energy distribution by linking peak and valley electricity price signals. It not only prioritizes the consumption of photovoltaic new energy, but also realizes the grid's "peak shaving and valley filling", thereby improving energy utilization efficiency and grid operation stability. 3. Good compatibility: The addition of standardized interfaces and protocols (compliant with relevant national standards) enables seamless connection to distribution networks and power generation, grid, load and energy storage systems, enhancing multi-energy complementarity capabilities and adapting to the deployment of integrated photovoltaic, energy storage and charging projects of different scales and scenarios; 4. High reliability: The system integrates a digital twin module and a thermal runaway protection unit, clearly defining the equipment model and protection parameters. This enables real-time monitoring of system operation status, fault early warning (response time ≤10s), and safety protection, reducing equipment failure rate and ensuring long-term stable system operation. 5. High practicality: The model and algorithm parameters of each module are clearly defined. Based on the actual operation experience of photovoltaic, energy storage and charging projects, the algorithm parameters and scheduling strategies can be optimized to adapt to the needs of multiple scenarios such as communities, highway service areas and parks, which facilitates implementation and reduces system operation and maintenance costs and grid interaction costs. Attached Figure Description

[0013] Figure 1 System architecture diagram; Figure 2 A schematic diagram of the intelligent control module's workflow; Figure 3 A functional diagram of the digital twin module. Detailed Implementation

[0014] The embodiments of the present invention will now be described in detail with reference to the accompanying drawings.

[0015] In the description of this invention, it should be understood that the terms "center", "upper", "lower", "front", "rear", "left", "right", "vertical", "horizontal", "top", "bottom", "inner", "outer", etc., indicate the orientation or positional relationship based on the orientation or positional relationship shown in the accompanying drawings. They are only for the convenience of describing this invention and simplifying the description, and do not indicate or imply that the device or element referred to must have a specific orientation, or be constructed and operated in a specific orientation. Therefore, they should not be construed as limitations on this invention.

[0016] The terms "first" and "second" are used for descriptive purposes only and should not be construed as indicating or implying relative importance or implicitly specifying the number of technical features indicated. Thus, a feature defined as "first" or "second" may explicitly or implicitly include one or more of that feature. In the description of this invention, unless otherwise stated, "a plurality of" means two or more.

[0017] In the description of this invention, it should be noted that, unless otherwise explicitly specified and limited, the terms "installation," "connection," and "joining" should be interpreted broadly. For example, they can refer to a fixed connection, a detachable connection, or an integral connection; they can refer to a direct connection or an indirect connection through an intermediate medium; and they can refer to the internal communication between two components. Those skilled in the art can understand the specific meaning of the above terms in this invention based on the specific circumstances.

[0018] like Figures 1 to 3 As shown, this invention provides an integrated intelligent scheduling system for photovoltaic, energy storage, and charging based on digital twins, including a photovoltaic power generation module, an energy storage module, a charging module, a grid interaction module, an intelligent control module, a digital twin module, and a communication module. Each module achieves data interaction and command transmission through the communication module. The specific structure is as follows: 1. Photovoltaic power generation module: Used to collect solar energy and convert it into electrical energy, including several photovoltaic modules and an inverter. The photovoltaic modules are used to receive solar energy and generate direct current (DC), and the inverter is used to convert DC to alternating current (AC). At the same time, the inverter collects the operating parameters of the photovoltaic modules (including light intensity, power generation, and module temperature) in real time and transmits them to the intelligent control module.

[0019] 2. Energy Storage Module: This module stores excess electrical energy generated by the photovoltaic (PV) power generation module and electrical energy during off-peak hours. It also releases energy when PV power generation is insufficient or during peak charging load periods. The module includes an energy storage battery pack, a battery management system (BMS), and a bidirectional converter. The BMS collects real-time operating parameters of the energy storage battery pack (including remaining capacity, charging / discharging power, battery temperature, and health status). The bidirectional converter enables energy conversion between the energy storage battery pack, the grid, and the charging module, and receives commands from the intelligent control module to adjust the charging / discharging direction and power.

[0020] 3. Charging Module: This module provides charging services for electric vehicles and includes several AC charging piles, DC charging piles, and a charging control unit. The charging control unit collects the operating parameters of each charging pile (including charging power, charging current, charging time, and the number of vehicles waiting to be charged), receives instructions from the intelligent control module to adjust the charging power, and achieves dynamic allocation of the charging load.

[0021] 4. Grid Interaction Module: This module enables power exchange and data communication between the system and the distribution network. It includes a grid interface, a power metering unit, and a grid connection control unit. The power metering unit measures the amount of power exchanged between the system and the grid (including grid-connected and off-grid power). The grid connection control unit controls the grid connection / off-grid switching between the system and the grid, and simultaneously collects real-time grid parameters (including grid voltage, frequency, and peak / valley electricity price signals) and transmits them to the intelligent control module.

[0022] 5. Intelligent Control Module: As the core control unit of the system, it integrates Model Predictive Control (MPC) algorithm and deep reinforcement learning algorithm. It receives real-time data transmitted from various modules, performs data processing, predictive analysis, and generates scheduling strategies. Specific algorithm parameters and functions include: (1) Data processing: The operating parameters transmitted by the photovoltaic power generation module, energy storage module, charging module and grid interaction module are denoised, filtered and integrated to form standardized data; the denoising adopts the Kalman filter algorithm (Q=1e-6, R=1e-3, initial variance P0=1) and the data sampling frequency is 10Hz; (2) Predictive analysis: Based on the deep reinforcement learning algorithm (using the DQN algorithm, the neural network structure is 12 neurons in the input layer, 2 hidden layers (64 neurons / layer), and 3 neurons in the output layer, with a learning rate of 0.001, a discount factor of 0.95, and an experience replay buffer size of 10000), combined with the historical 12 months of operating data, real-time light data, grid peak and valley electricity price signals, and charging load data, the photovoltaic power generation, remaining energy storage capacity, and charging load demand for the next 1-24 hours are accurately predicted, with a prediction error of ≤5%; (3) Scheduling strategy generation: Based on the model predictive control (MPC) algorithm (prediction time domain 24h, control time domain 1h, constraints are energy storage charging and discharging power ≤200kW (configurable), total power of charging modules ≤820kW (configurable), objective function is to maximize photovoltaic absorption rate + minimize grid interaction cost), combined with prediction data and system operation objectives (prioritize photovoltaic absorption, balance energy storage life, optimize grid interaction cost), the energy flow path is dynamically adjusted to generate the optimal scheduling command and control the operation status of each module; (4) Coordinated control: Link peak and valley electricity price signals, control the charging of energy storage modules during the low period of the power grid and control the discharging of energy storage modules during the peak period of the power grid to achieve "peak shaving and valley filling"; at the same time, coordinate the operation of photovoltaic power generation modules, energy storage modules and charging modules, give priority to using photovoltaic power to charge electric vehicles, store excess power in energy storage modules, and supplement power supply from energy storage modules or the power grid when photovoltaic power is insufficient, to achieve deep coordination of "photovoltaic priority, storage regulation and grid complementarity".

[0023] 6. Digital Twin Module: Utilizing technologies such as Unity3D, a digital twin model of the integrated photovoltaic, energy storage, and charging system is constructed. Equipped with an industrial computer, it integrates real-time and historical operational data from each module to achieve visualized monitoring of system operation status, fault early warning, and simulation. Specific functions include: (1) Visual monitoring: The operating parameters, energy flow path and overall system operating status of each module are displayed in real time through digital twin model, which makes it easy for staff to intuitively grasp the system operation; the data refresh frequency is 1 second / time, and it supports multi-view switching; (2) Fault warning: Based on real-time operation data and fault diagnosis algorithm (using BP neural network algorithm, the input is the operation parameters of each module, and the output is the fault type and level), identify potential faults of each module (such as photovoltaic module degradation, thermal runaway of energy storage battery, and charging pile fault), issue early warning signals in a timely manner and push fault handling suggestions, with an early warning response time of ≤10s; (3) Simulation: Based on the digital twin model, the system operation status under different lighting conditions, charging load and grid electricity price is simulated. The simulation step size is 1 minute, which provides data support for scheduling strategy optimization.

[0024] 7. Communication Module: Adopting 5G industrial modules (supporting 5G SA / NSA dual-mode, downlink speed ≥1Gbps) + Internet of Things (IoT) communication technology, it realizes data interaction and command transmission between modules with a transmission latency ≤100ms; at the same time, it adds standardized interfaces and protocols (complying to GB / T 36278-2018 "Technical Specification for Electric Vehicle Charging and Swapping Facilities Access to Distribution Network" and T / CITS260—2025 "Technical Specification for Integrated Renewable Energy Power Generation of Source-Grid-Load-Storage") to achieve seamless connection between the system and the distribution network and source-grid-load-storage system, and improve the multi-energy complementarity capability.

[0025] The energy storage module also includes a thermal runaway protection unit, which adopts supramolecular full immersion energy storage technology and combines real-time battery temperature monitoring data. When the battery temperature exceeds a preset threshold, it automatically activates cooling protection to suppress battery thermal runaway and improve system operational safety.

[0026] The charging module's charging control unit supports multiple charging modes (including constant current charging, constant voltage charging, and intelligent fast charging). It can automatically switch charging modes according to the electric vehicle's battery type and charging needs, combined with the scheduling instructions of the intelligent control module. The constant current charging current range is 0-80A, the constant voltage charging voltage range is 200-750V, and the charging efficiency in intelligent fast charging mode is ≥95%, improving charging efficiency and battery life.

[0027] The intelligent control module also includes a parameter optimization unit, which can adjust the parameters of the model predictive control (MPC) algorithm and the deep reinforcement learning algorithm (such as the prediction time domain of MPC and the learning rate of DQN) based on the actual project operation experience, optimize the adaptability of the scheduling strategy, and improve the system operating efficiency.

[0028] Example: This embodiment provides an integrated intelligent scheduling system for photovoltaic, energy storage, and charging based on digital twins, applied to a photovoltaic, energy storage, and charging project in an office park. The specific structure is as follows: The photovoltaic power generation module uses 800 LONGi LHP06-72H high-efficiency monocrystalline silicon photovoltaic modules (peak power of 590W per module), with a total installed capacity of 472kW (actual installed capacity of 100kW, with 18% redundancy reserved). The inverter uses 4 Huawei SUN2000-100KTL-H1 centralized inverters (rated power of 100kW, conversion efficiency of 98.7%). The system collects real-time data on the irradiance, power generation, and module temperature of the photovoltaic modules and transmits it to the intelligent control module. The data sampling frequency is 10Hz.

[0029] The energy storage module uses 400 CATL-50Ah lithium iron phosphate batteries (3.2V per cell, connected in series to form 100 strings and in parallel to form 4 groups), with a total capacity of 204.8kWh (200kWh actually used). The battery management system is BYD BMS-1000 (sampling accuracy ±1%, supports multiple strings and parallel connections). The bidirectional converter uses one Sungrow SG350HX bidirectional converter (rated power 200kW, conversion efficiency 97.9%), which supports continuously adjustable charging and discharging power from 0-200kW. The thermal runaway protection unit uses Shanghai Electric supramolecular immersion fluid (boiling point 150℃, insulation class Class F). When the battery temperature exceeds 55℃, it automatically starts cooling protection at a cooling rate of 6℃ / min to suppress battery thermal runaway.

[0030] The charging module is equipped with 10 TELD TCU-60kW DC charging piles (output voltage 200-750V, output current 0-80A). The charging control unit uses the Infineon XC2785 charging controller to collect the charging power, charging current and number of vehicles to be charged of each charging pile in real time. It supports switching between three modes: constant current, constant voltage and intelligent fast charging. The charging efficiency is 95.5% in intelligent fast charging mode.

[0031] The grid interaction module uses one Schneider 10kV high-voltage interface (rated current 630A, insulation class IP65), and the power metering unit uses one Huali DSSD536 high-precision smart meter (metering accuracy 0.2S level, supports bidirectional metering), which can measure the grid-connected and grid-off power of the system. The grid connection control unit uses one Xuji WGB-871 grid connection controller (supports automatic grid connection / off-grid switching, response time 45ms), which collects grid voltage (10kV±5%), frequency (50Hz±0.5Hz) and peak-valley electricity price signals in real time (off-peak electricity price 0.35 yuan / kWh, peak electricity price 0.85 yuan / kWh).

[0032] The intelligent control module employs a single Siemens S7-1500 industrial-grade PLC controller (CPU model 1516-3PN / DP, 1MB memory, processing speed 0.1μs / instruction), integrating Model Predictive Control (MPC) and Deep Reinforcement Learning (DQN) algorithms. The data processing unit uses a Kalman filter algorithm (Q=1e-6, R=1e-3, initial variance P0=1) to reduce noise in the parameters transmitted by each module. The predictive analysis unit, based on 12 months of historical operating data, predicts the photovoltaic power generation, remaining energy storage capacity, and charging load demand for the next 24 hours, with a prediction error of 4.8%. The strategy generation unit generates the optimal scheduling strategy based on the predicted data. 1. During periods of ample sunshine (7:00-17:00): The photovoltaic power generation modules generate electricity at full capacity, prioritizing the charging of electric vehicles. Excess electricity is stored in the energy storage module. If there is still surplus photovoltaic electricity, it is connected to the distribution network through the grid interaction module. 2. Off-peak hours (23:00-7:00 the next day): Control the charging of energy storage modules to store low-priced electricity from the grid; 3. Peak grid hours (8:00-11:00, 17:00-21:00): Control the discharge of energy storage modules to supplement charging load demand, reduce the amount of electricity discharged from the grid, and reduce electricity costs; 4. During periods of insufficient or no sunlight (17:00-7:00 the next day, excluding off-peak hours): Energy storage modules will be used first to charge electric vehicles. If the stored energy is insufficient, the power grid will supplement the supply.

[0033] The digital twin module uses Unity3D to build a 3D digital model of the system and is equipped with an Advantech IPC-610L industrial computer (CPU i7-12700, memory 32GB, hard drive 1TB). It visualizes the operating parameters and energy flow paths of each module, with a data refresh rate of 1 second. The fault early warning unit monitors the operating parameters of each module in real time. When the photovoltaic module's power generation is abnormally reduced (below 80% of normal power), the energy storage battery temperature exceeds 50°C, or the charging pile malfunctions, it promptly issues an early warning signal and pushes handling suggestions. The early warning response time is 8 seconds. The simulation unit can simulate the system's operating state under different light intensities and charging loads, with a simulation step size of 1 minute, and optimizes the scheduling strategy.

[0034] The communication module adopts Huawei's 5G industrial module (model ME909s-821, supporting 5G SA / NSA dual-mode, downlink speed of 1.2Gbps) + Internet of Things (IoT) communication technology, with a transmission latency of 85ms. It adds standardized interfaces and protocols (compliant with GB / T36278-2018) to achieve seamless connection with the industrial park's source-grid-load-storage system, share operating data, and enhance multi-energy complementarity capabilities.

[0035] The present invention has been described in detail above with reference to the accompanying drawings. However, the present invention is not limited to the embodiments described above. Within the scope of knowledge possessed by those skilled in the art, various changes can be made without departing from the spirit of the present invention. Many other changes and modifications made without departing from the concept and scope of the present invention should be considered within the scope of protection of the present invention.

[0036] In the description of this specification, specific features, structures, materials, or characteristics may be combined in any suitable manner in one or more embodiments or examples.

[0037] The above description is merely a specific embodiment of the present invention, but the scope of protection of the present invention is not limited thereto. Any variations or substitutions that can be easily conceived by those skilled in the art within the technical scope disclosed in the present invention should be included within the scope of protection of the present invention. Therefore, the scope of protection of the present invention should be determined by the scope of the claims.

Claims

1. A digital twin-based integrated intelligent scheduling system for photovoltaic, energy storage, and charging, characterized in that: It includes photovoltaic power generation modules, energy storage modules, charging modules, grid interaction modules, intelligent control modules, digital twin modules, and communication modules. Each module interacts with the others and transmits commands through the communication module. Photovoltaic power generation module: used to collect solar energy and convert it into electrical energy, while simultaneously collecting the operating parameters of the photovoltaic modules in real time and transmitting them to the intelligent control module; Energy storage module: Used to store excess electrical energy generated by photovoltaic power generation modules and electrical energy during off-peak hours of the power grid. It also releases electrical energy when photovoltaic power generation is insufficient or charging load is high, and collects the operating parameters of the energy storage battery pack in real time. Charging module: Used to provide charging services for electric vehicles, collect the operating parameters of each charging pile, and receive instructions from the intelligent control module to adjust the charging power and realize the dynamic distribution of charging load; Grid Interaction Module: Used to realize the power interaction and data communication between the system and the distribution network, including the grid interface, power metering unit and grid connection control unit; the power metering unit is used to measure the amount of power exchange between the system and the grid, and the grid connection control unit is used to control the grid connection / off-grid switching between the system and the grid, and at the same time collect the real-time parameters of the grid and transmit them to the intelligent control module; Intelligent control module: integrates model predictive control algorithm and deep reinforcement learning algorithm, used to receive real-time data transmitted from each module, perform data processing, predictive analysis and generate scheduling strategies; Digital twin module: Used to integrate real-time and historical data from various modules to achieve visualized monitoring of system operation status, fault early warning, and simulation.

2. The integrated intelligent scheduling system for photovoltaic storage and charging based on digital twins according to claim 1, characterized in that, The intelligent control module includes: Data processing unit: performs noise reduction, filtering and integration of operating parameters transmitted by photovoltaic power generation module, energy storage module, charging module and grid interaction module to form standardized data; noise reduction adopts Kalman filter algorithm, and the data sampling frequency is 10Hz; Predictive Analysis Unit: Based on deep reinforcement learning algorithms, combined with 12 months of historical operating data, real-time solar irradiance data, grid peak-valley electricity price signals, and charging load data, it accurately predicts photovoltaic power generation, remaining energy storage capacity, and charging load demand for the next 1-24 hours, with a prediction error of ≤5%. Scheduling strategy generation unit: Based on model predictive control algorithm, combined with prediction data and system operation objectives, dynamically adjusts energy flow path, generates optimal scheduling instructions, and controls the operation status of each module; Coordinated control unit: Linked to peak and valley electricity price signals, it controls the charging of energy storage modules during off-peak hours and the discharging of energy storage modules during peak hours to achieve "peak shaving and valley filling"; at the same time, it coordinates the operation of photovoltaic power generation modules, energy storage modules and charging modules, giving priority to using photovoltaic power to charge electric vehicles, storing excess power in energy storage modules, and supplementing power from energy storage modules or the grid when photovoltaic power is insufficient, achieving deep coordination of "photovoltaic priority, storage regulation and grid complementarity".

3. The integrated intelligent scheduling system for photovoltaic storage and charging based on digital twins according to claim 1, characterized in that, The aforementioned visualization monitoring uses a digital twin model to display the operating parameters of each module, energy flow paths, and overall system operating status in real time, allowing staff to intuitively grasp the system's operating status; the data refresh rate is 1 second / time, and it supports multi-view switching.

4. The integrated intelligent scheduling system for photovoltaic, energy storage, and charging based on digital twins according to claim 1, characterized in that, The aforementioned fault warning: Based on real-time operating data and fault diagnosis algorithms, it identifies potential faults in each module, issues warning signals in a timely manner, and pushes fault handling suggestions, with a warning response time of ≤10s.

5. The integrated intelligent scheduling system for photovoltaic, energy storage, and charging based on digital twins according to claim 1, characterized in that, The simulation is based on a digital twin model to simulate the system operation under different lighting conditions, charging loads, and grid electricity prices. The simulation step size is 1 minute, which provides data support for scheduling strategy optimization.

6. The integrated intelligent scheduling system for photovoltaic storage and charging based on digital twins according to claim 1, characterized in that, The communication module uses a 5G industrial module to realize data interaction and command transmission between modules, with a transmission delay of ≤100ms.

7. The integrated intelligent scheduling system for photovoltaic storage and charging based on digital twins according to claim 1, characterized in that, The energy storage module also includes a thermal runaway protection unit, which adopts supramolecular full immersion energy storage technology and combines real-time battery temperature monitoring data. When the battery temperature exceeds a preset threshold, it automatically activates cooling protection to suppress battery thermal runaway and improve system operational safety.

8. The integrated intelligent scheduling system for photovoltaic, energy storage, and charging based on digital twins according to claim 1, characterized in that, The system includes several AC charging piles, DC charging piles, and a charging control unit. The charging control unit supports multi-mode charging and can automatically switch charging modes according to the battery type and charging needs of the electric vehicle, combined with the scheduling instructions of the intelligent control module. The constant current charging current range is 0-80A, the constant voltage charging voltage range is 200-750V, and the charging efficiency in intelligent fast charging mode is ≥95%, improving charging efficiency and battery life.

9. The integrated intelligent scheduling system for photovoltaic storage and charging based on digital twins according to claim 1, characterized in that, The intelligent control module also includes a parameter optimization unit, which can adjust the parameters of the model predictive control algorithm and the deep reinforcement learning algorithm based on the actual project operation experience, optimize the adaptability of the scheduling strategy, and improve the system operating efficiency.

Images