Integrated energy station management system based on optical storage and charging micro-grid

Abstract

The invention belongs to the technical field of energy station management, and particularly relates to a comprehensive energy station management system based on a light storage and charging micro-grid, and the system comprises a sensing layer which is used for collecting the real-time operation data of a photovoltaic facility, an energy storage system, a charging pile and an environment; the network layer is used for transmitting data of the sensing layer in real time and establishing two-way communication with a power grid dispatching center at the same time; and the application layer is used for realizing energy optimization scheduling, equipment cooperation, fault early warning and power grid interaction according to the data transmitted by the network layer. The AI dynamic energy scheduling module, the prediction model module, the optimization model module and the solving module are arranged, so that the energy utilization efficiency can be improved, the light abandoning rate can be greatly reduced, the electricity purchasing cost can be remarkably reduced, and economic optimization is realized.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of energy station management, in particular to a comprehensive energy station management system based on a light-storage-charging microgrid. BACKGROUND

[0002] The penetration rate of new energy vehicles continues to rise, and the demand for public and private charging is growing explosively. At the same time, distributed photovoltaics, due to their clean and low-carbon characteristics, have become one of the core energy supply units of comprehensive energy stations, and energy storage systems can effectively smooth out photovoltaic output fluctuations and alleviate the peak-valley difference of charging load. The combination of the three forms a light-storage-charging microgrid comprehensive energy station, which has become an important part of urban energy networks.

[0003] Although the current light-storage-charging microgrid comprehensive energy station can achieve "power generation - power storage - power consumption" integration, the existing management system still has many technical bottlenecks. On the one hand, energy dispatching relies on fixed strategies (such as "photovoltaic priority, surplus power storage"), without combining real-time load fluctuations, energy storage status, and dynamic adjustments of peak and valley electricity prices. On the other hand, photovoltaic inverters, energy storage converters, and charging piles are controlled independently, lacking a coordinated linkage mechanism. In addition, device fault warnings rely on threshold judgments and cannot achieve predictive maintenance, resulting in low system operation efficiency, poor stability, and high operation and maintenance costs. The specific problems are as follows:

[0004] (1) Lack of energy dispatching intelligence, high light rejection rate and high electricity purchase cost

[0005] The existing system mostly uses "static dispatching strategies" and cannot match the dynamic changes of photovoltaic output and charging load in real time. For example: a commercial district light-storage-charging comprehensive energy station is equipped with a 150kW photovoltaic array, an 80kWh energy storage system, and 8 60kW charging piles. During the summer noon 12:00-14:00, the real-time photovoltaic output reaches 140kW, while the charging load is only 30kW (many business vehicles have left), and the energy storage system is fully charged (SOC=100%). The system cannot effectively absorb the excess 110kW of electricity, and can only be disposed of by "light rejection", with a light rejection rate of 78.6%. During the evening peak of 18:00-20:00, the photovoltaic output tends to be 0, and the energy storage SOC drops to 20% (the minimum protection threshold). The system needs to purchase electricity from the grid to meet the 480kW load demand of the 8 charging piles. Due to the peak segment of the grid (electricity price 1.5 yuan / kWh), the daily electricity purchase cost exceeds 1200 yuan, and the energy utilization efficiency and economy are seriously imbalanced.

[0006] (2) Poor multi-device coordination, insufficient microgrid stability

[0007] The lack of a unified and coordinated control mechanism for photovoltaic (PV) systems, energy storage, and charging piles can easily lead to microgrid voltage fluctuations and equipment tripping. For example, a residential community's PV-energy storage-charging station is equipped with 100kW PV, 60kWh energy storage, and six 40kW charging piles. One morning between 8:00 and 9:00 AM (peak charging time for commuters), all five charging piles simultaneously activated fast charging mode (total load 200kW). At this time, the PV output was only 40kW, and the energy storage SOC was 30% (discharge power limit 50kW), resulting in a total system power supply (90kW) far below the load demand. Because the charging piles did not receive power adjustment instructions and continued to operate at full power, the microgrid voltage plummeted from 380V to 340V (below the national standard lower limit of 353V), triggering the charging piles' over / under voltage protection. All five devices tripped simultaneously, leaving over 30 vehicles waiting to be charged unable to receive power, resulting in a very poor user experience.

[0008] (iii) Lack of predictive maintenance and high rate of sudden failures.

[0009] Current systems only monitor equipment status at the level of "real-time data acquisition + threshold alarms," ​​failing to predict equipment lifespan degradation and potential failures. For example, the energy storage system (lithium battery) of a photovoltaic-energy storage-charging station in a highway service area had been operating for three years. The system only monitored whether the battery voltage and current exceeded the range of "0.8Ue~1.2Ue," without analyzing the battery's cycle life or internal resistance changes. On a holiday peak charging day (with 10 charging piles operating at full capacity), the energy storage battery's internal resistance had increased to 2.5 times its initial value (which was not monitored). During discharge, localized overheating occurred, triggering a fire alarm and causing the entire energy station to shut down for four hours. This not only resulted in direct economic losses but also caused significant traffic congestion.

[0010] Based on the above, an integrated energy station management system based on a photovoltaic, energy storage, and charging microgrid is invented. Summary of the Invention

[0011] To solve the above-mentioned technical problems, the present invention provides the following technical solution:

[0012] A comprehensive energy station management system based on a photovoltaic-storage-charging microgrid includes:

[0013] The sensing layer is used to collect real-time operational data from photovoltaic facilities, energy storage systems, charging piles, and the environment.

[0014] The network layer is used for real-time transmission of data from the sensing layer, and at the same time, it establishes two-way communication with the power grid dispatch center.

[0015] The application layer is used to realize energy optimization scheduling, equipment coordination, fault early warning and grid interaction based on the data transmitted from the network layer;

[0016] The application layer includes:

[0017] The AI ​​dynamic energy dispatch module is used to predict photovoltaic output and charging load in the next hour using LSTM neural network, and to establish a multi-objective optimization model with the goal of "curtailment rate ≤5% and minimization of electricity purchase cost" by combining energy storage SOC and grid peak and valley electricity price. At the same time, the optimal dispatch strategy is solved by improving particle swarm algorithm.

[0018] The multi-device collaborative control module is used to dynamically adjust the power of the charging pile, the output power of the photovoltaic inverter and the charging and discharging power of the energy storage converter, and to compensate for reactive power through the energy storage converter, adjust the power factor of the photovoltaic inverter and maintain the microgrid voltage within the range of 380V±3%.

[0019] The predictive maintenance and fault diagnosis module is used to predict the remaining battery life by collecting the internal resistance data of the energy storage battery and combining it with the number of charge and discharge cycles using a random forest algorithm. At the same time, when the battery SOH≤80% or the equipment parameters exceed the safety threshold, it issues a three-level warning and generates maintenance suggestions.

[0020] As a preferred embodiment of the integrated energy station management system based on photovoltaic-storage-charging microgrid described in this invention, the AI ​​dynamic energy scheduling module includes:

[0021] The prediction model module uses an LSTM neural network to combine historical photovoltaic output data, real-time irradiance, and weather forecasts to predict photovoltaic output for the next hour; at the same time, it predicts charging load for the next hour based on historical charging data and user reservation information.

[0022] The optimization model module is used to establish a multi-objective optimization equation with the objectives of "curtailment rate ≤ 5% and minimum electricity purchase cost" as inputs: photovoltaic predicted output, charging predicted load, energy storage SOC, and grid peak-valley electricity price. ,in, , These are the weighting coefficients;

[0023] The solution module is used to solve for the optimal scheduling strategy using an improved particle swarm optimization algorithm, with a scheduling instruction generation cycle of 5 minutes.

[0024] As a preferred embodiment of the integrated energy station management system based on photovoltaic-storage-charging microgrid described in this invention, the multi-device collaborative control module includes:

[0025] The power coordination module is used to send dynamic power commands to the charging pile according to the AI ​​scheduling strategy, and at the same time control the output power of the photovoltaic inverter and the charging and discharging power of the energy storage converter.

[0026] The voltage stabilization module is used to monitor the microgrid voltage in real time. When the voltage deviates from 380V±5%, the energy storage converter automatically compensates for reactive power, and the photovoltaic inverter adjusts the power factor to ensure that the voltage is stable within the national standard range.

[0027] As a preferred embodiment of the integrated energy station management system based on photovoltaic-storage-charging microgrid described in this invention, the predictive maintenance and fault diagnosis module includes:

[0028] The status monitoring module is used to collect internal resistance data through the internal resistance sensor of the energy storage battery and, in combination with the number of charge-discharge cycles, establish a prediction model for the remaining battery life; at the same time, it monitors the temperature of the charging pile module and the status of the photovoltaic inverter IGBT.

[0029] The fault warning module is used to automatically issue a three-level warning and generate maintenance suggestions when the battery SOH≤80% and the charging pile module temperature≥85℃.

[0030] The fault diagnosis module is used to establish a knowledge base for equipment faults and to locate the cause of faults by matching real-time data with the knowledge base.

[0031] As a preferred embodiment of the integrated energy station management system based on photovoltaic-storage-charging microgrid described in this invention, the application layer further includes:

[0032] The user interaction and charging service optimization module supports users in reserving charging time slots and automatically coordinates the reservation time slots when multiple users reserve the same charging station. It also supports charging service value-added services and feedback.

[0033] As a preferred embodiment of the integrated energy station management system based on photovoltaic-storage-charging microgrid described in this invention, the user interaction and charging service optimization module includes:

[0034] The charging reservation and load prediction coordination module provides a user-end APP / mini-program entry point, supports users to reserve charging time slots and select charging power, and synchronizes reservation data to the AI ​​dynamic energy scheduling module in real time.

[0035] The charging conflict coordination and intelligent navigation module is used to automatically coordinate the reservation time based on the charging time priority and scheduling feasibility when multiple users reserve the same charging pile. It also guides users to the available charging pile through the navigation function, reducing on-site waiting time.

[0036] The charging service value-added and feedback closed-loop module is used to support photovoltaic green electricity traceability charging and time-of-use electricity price reminders. At the same time, it receives equipment fault feedback from users and automatically synchronizes it to the predictive maintenance and fault diagnosis module, forming a closed loop of user feedback-fault location-maintenance.

[0037] As a preferred embodiment of the integrated energy station management system based on photovoltaic-storage-charging microgrid described in this invention, the application layer further includes:

[0038] The demand response and grid interaction module is used to control the full-power charging of the energy storage system during off-peak hours and to supply charging by discharging the energy storage during peak hours. At the same time, it works in conjunction with the grid dispatch center to increase the energy storage discharge power and provide frequency regulation services.

[0039] As a preferred embodiment of the integrated energy station management system based on photovoltaic-storage-charging microgrid described in this invention, the demand response and grid interaction module includes:

[0040] The peak-valley arbitrage module is used to receive peak-valley electricity price signals from the power grid, so as to control the energy storage system to charge at full power during the valley period and use the energy storage to discharge and charge during the peak period, thereby reducing the purchase of electricity from the power grid.

[0041] The ancillary service module is used to work in conjunction with the power grid dispatch center. When the power grid issues a peak-shaving demand, it increases the energy storage discharge power and reduces the purchase of electricity. When the power grid issues a frequency regulation demand, the energy storage converter responds quickly and provides primary frequency regulation services to obtain ancillary service revenue.

[0042] Compared with existing technologies:

[0043] 1. By setting up an AI dynamic energy scheduling module, and utilizing the prediction model module, optimization model module, and solution module, not only can energy utilization efficiency be improved, but the curtailment rate of solar power can also be significantly reduced, thereby significantly reducing the cost of electricity purchase and achieving economic optimization.

[0044] 2. By setting up a multi-device collaborative control module, and utilizing the power collaboration module and voltage stabilization module, not only can the stability of microgrid operation be improved and the voltage / frequency maintained within the national standard range, but the risk of equipment tripping can also be reduced, thus ensuring charging continuity.

[0045] 3. By setting up predictive maintenance and fault diagnosis modules, and utilizing condition monitoring, fault early warning, and fault diagnosis modules, not only can equipment failures be predicted and sudden risks reduced, but maintenance response efficiency can also be improved, downtime losses reduced, and equipment lifespan extended. Attached Figure Description

[0046] Fig. 1 This is a schematic diagram of the overall framework of the present invention;

[0047] Fig. 2 This is a schematic diagram of the application layer framework of the present invention. Detailed Implementation

[0048] To make the objectives, technical solutions, and advantages of the present invention clearer, the embodiments of the present invention will be described in further detail below with reference to the accompanying drawings.

[0049] This invention provides an integrated energy station management system based on a photovoltaic-storage-charging microgrid. Please refer to [link / reference]. Figs. 1-2 ,include:

[0050] The sensing layer is used to collect real-time operating data of photovoltaic facilities, energy storage systems, charging piles, and the environment; the sampling frequency is 1 time / second, and the data accuracy reaches 0.5 level; the sensing layer includes photovoltaic monitoring units (current / voltage / irradiance sensors), energy storage monitoring units (battery voltage / current / temperature / internal resistance sensors), charging pile metering units (charging power / power sensors), and environmental monitoring units (temperature / wind speed sensors).

[0051] The network layer is used for real-time transmission of data from the sensing layer (latency ≤ 100ms) and for establishing bidirectional communication with the power grid dispatch center.

[0052] The application layer is used to realize energy optimization scheduling, equipment coordination, fault early warning and grid interaction based on the data transmitted from the network layer;

[0053] The application layer includes:

[0054] The AI ​​dynamic energy dispatch module is used to predict photovoltaic output and charging load in the next hour using LSTM neural network, and to establish a multi-objective optimization model with the goal of "curtailment rate ≤5% and minimization of electricity purchase cost" by combining energy storage SOC and grid peak and valley electricity price. At the same time, the optimal dispatch strategy is solved by improving particle swarm algorithm.

[0055] The multi-device collaborative control module is used to dynamically adjust the power of the charging pile, the output power of the photovoltaic inverter and the charging and discharging power of the energy storage converter, and to compensate for reactive power through the energy storage converter, adjust the power factor of the photovoltaic inverter and maintain the microgrid voltage within the range of 380V±3%.

[0056] The predictive maintenance and fault diagnosis module is used to predict the remaining battery life by collecting the internal resistance data of the energy storage battery and combining it with the number of charge and discharge cycles using a random forest algorithm. At the same time, when the battery SOH≤80% or the equipment parameters exceed the safety threshold, it issues a three-level warning and generates maintenance suggestions.

[0057] The user interaction and charging service optimization module supports users to reserve charging time slots and automatically coordinates the reservation time slots when multiple users reserve the same charging station. It also supports charging service value-added services and feedback.

[0058] The demand response and grid interaction module is used to control the full-power charging of the energy storage system during off-peak hours and to supply charging by discharging the energy storage during peak hours. At the same time, it works in conjunction with the grid dispatch center to increase the energy storage discharge power and provide frequency regulation services.

[0059] The AI ​​dynamic energy scheduling module includes:

[0060] The prediction model module uses an LSTM neural network to combine historical photovoltaic power output data (last year), real-time irradiance, and weather forecasts to predict photovoltaic power output for the next hour (accuracy ≥ 92%); at the same time, it predicts charging load for the next hour based on historical charging data (last 3 months) and user reservation information (accuracy ≥ 88%).

[0061] The optimization model module is used to establish a multi-objective optimization equation with the objectives of "curtailment rate ≤ 5% and minimization of electricity purchase cost". The inputs include predicted photovoltaic output, predicted charging load, energy storage SOC (state of charge), and grid peak-valley electricity price (obtained in real time). ,in, , This is a weighting coefficient, which can be adjusted according to user needs;

[0062] The solution module is used to solve the optimal scheduling strategy (such as "photovoltaic priority charging + surplus power charging and storage" and "energy storage discharge to make up for peak demand + valley power charging and storage") using an improved particle swarm algorithm (introducing a dynamic adjustment mechanism for inertial weights). The scheduling instruction generation cycle is 5 minutes.

[0063] The multi-device collaborative control module includes:

[0064] The power coordination module is used to send dynamic power commands to the charging pile according to the AI ​​scheduling strategy (such as reducing the charging pile power from 60kW to 30kW if the energy storage SOC is ≤20% during peak charging time), while controlling the output power of the photovoltaic inverter (±10% adjustment range) and the charging and discharging power of the energy storage converter (0~100% continuously adjustable).

[0065] The voltage stabilization module is used to monitor the microgrid voltage in real time. When the voltage deviates from 380V±5%, the energy storage converter automatically compensates for reactive power (compensation range -50kVar to +50kVar), and the photovoltaic inverter adjusts the power factor (0.8 lagging to 0.8 leading) to ensure that the voltage is stable within the national standard range.

[0066] The predictive maintenance and fault diagnosis module includes:

[0067] The status monitoring module is used to collect internal resistance data through the internal resistance sensor of the energy storage battery and, in combination with the number of charge-discharge cycles, establish a prediction model for the remaining battery life (SOH) (using the random forest algorithm, with a prediction error ≤5%); at the same time, it monitors the temperature of the charging pile module and the status of the photovoltaic inverter IGBTs.

[0068] The fault warning module is used to automatically issue a three-level warning (SMS + platform alarm) when the battery SOH≤80% (life decay threshold) or the charging pile module temperature≥85℃ (overheating threshold), and generate maintenance suggestions (such as "replace energy storage battery module" and "clean the heat dissipation channel of charging pile").

[0069] The fault diagnosis module is used to establish a knowledge base for equipment faults (containing 100+ typical fault cases). By matching real-time data with the knowledge base, the cause of the fault can be located (such as "charging pile tripping = voltage fluctuation + module overheating"). The diagnostic accuracy rate is ≥90%.

[0070] The user interaction and charging service optimization module includes:

[0071] The charging reservation and load prediction coordination module provides an entry point for users' apps / mini-programs, allowing them to reserve charging time slots (accurate to 15 minutes) and select charging power (30kW / 60kW). The reservation data is synchronized to the AI ​​dynamic energy scheduling module in real time. For example, if a user reserves a 60kW charging time slot from 9:00 to 10:00 the next day, the module automatically converts this request into "60kW additional load during the 9:00-10:00 time slot" and adds it to the load prediction model, improving the prediction accuracy from 88% to over 95%.

[0072] The charging conflict coordination and intelligent navigation module is used to automatically coordinate the reservation time period (e.g., adjusting some users to 8:30-9:30) when multiple users reserve charging piles in the same time period (e.g., the number of reservations exceeds the number of charging piles from 8:00 to 9:00 in the morning) based on the priority of charging time (e.g., emergency charging users <1 hour, regular users 1-2 hours) and scheduling feasibility (combined with the photovoltaic output prediction of module 1). It also guides users to available charging piles through the navigation function, reducing on-site waiting time.

[0073] The charging service value-added and feedback closed-loop module is used to support photovoltaic green electricity traceability charging (showing users the proportion of photovoltaic power in the charging amount, such as "80% of this charging comes from photovoltaic green electricity"), time-of-use electricity price reminders (pushing "High electricity price period is coming soon, it is recommended to adjust the charging time" 10 minutes before the peak period), and receiving user feedback on equipment failures (such as "charging pile is stuck"), which are automatically synchronized to the predictive maintenance and fault diagnosis module to form a user feedback-fault location-maintenance closed loop.

[0074] By setting up user interaction and charging service optimization modules:

[0075] Improve user satisfaction: By coordinating appointments and providing navigation, the pain point of "charging queues" has been resolved, increasing the user charging success rate from 82% to 99%; the green electricity traceability function meets users' low-carbon needs, and the user retention rate has increased by 30%.

[0076] Optimize load forecasting accuracy: Incorporate the "deterministic load" reserved by users into the scheduling model to avoid scheduling deviations caused by sudden loads and reduce curtailment rate (curtailment rate can be controlled within 5%).

[0077] Expanding value-added services: Based on user charging data (such as monthly charging frequency and preferred power), value-added services such as "green electricity packages" and "member discounts" can be launched, generating an additional annual revenue of approximately 20,000 to 30,000 yuan per station.

[0078] The demand response and power grid interaction module includes:

[0079] The peak-valley arbitrage module is used to receive peak and valley electricity price signals from the power grid (such as 1.5 yuan / kWh during peak hours and 0.3 yuan / kWh during valley hours) to control the energy storage system to charge at full power during valley hours (from 20% to 100% SOC), and to prioritize the energy storage system to discharge and charge during peak hours, thereby reducing the amount of electricity purchased from the power grid.

[0080] The ancillary service module is used to link with the power grid dispatch center. When the power grid issues a peak-shaving demand, it increases the energy storage discharge power (maximum 100% of rated power) to reduce electricity purchases. When the power grid issues a frequency regulation demand, the energy storage converter responds quickly (frequency regulation response time ≤200ms) to provide a primary frequency regulation service and obtain ancillary service revenue.

[0081] In practical use, the specific steps are as follows:

[0082] S1: Real-time operational data of photovoltaic facilities, energy storage systems, charging piles, and the environment are collected through the sensing layer; the sampling frequency is 1 time / second, and the data accuracy reaches 0.5 level; the sensing layer includes a photovoltaic monitoring unit (current / voltage / irradiance sensor), an energy storage monitoring unit (battery voltage / current / temperature / internal resistance sensor), a charging pile metering unit (charging power / power sensor), and an environmental monitoring unit (temperature / wind speed sensor).

[0083] S2: Real-time transmission of data from the sensing layer to the network layer (latency ≤ 100ms), while establishing bidirectional communication with the power grid dispatch center;

[0084] S3: Through the application layer, based on the data transmitted from the network layer, energy optimization scheduling, equipment coordination, fault early warning and grid interaction are realized;

[0085] S31: Through the AI ​​dynamic energy dispatch module, the LSTM neural network is used to predict the photovoltaic output and charging load in the next hour. Combined with the energy storage SOC and the grid peak and valley electricity price, a multi-objective optimization model is established with the goal of "curtailment rate ≤5% and minimization of electricity purchase cost". At the same time, the optimal dispatch strategy is solved by improving the particle swarm algorithm.

[0086] S32: The multi-device collaborative control module dynamically adjusts the charging pile power, photovoltaic inverter output power and energy storage converter charging and discharging power, and compensates for reactive power through the energy storage converter, adjusts the power factor of the photovoltaic inverter, and maintains the microgrid voltage within the range of 380V±3%.

[0087] S33: The predictive maintenance and fault diagnosis module uses the collected internal resistance data of the energy storage battery and the number of charge-discharge cycles to predict the remaining battery life using a random forest algorithm; at the same time, when the battery SOH≤80% or the equipment parameters exceed the safety threshold, a three-level warning is issued and maintenance suggestions are generated.

[0088] S34: The user interaction and charging service optimization module supports users to reserve charging time slots, and automatically coordinates the reservation time slots when multiple users reserve the same charging station. It also supports charging service value-added services and feedback.

[0089] S35: Through the demand response and grid interaction module, the energy storage system is controlled to charge at full power during valley hours and to discharge and charge during peak hours. At the same time, it is linked with the grid dispatch center to increase the energy storage discharge power and provide frequency regulation services.

[0090] Although the present invention has been described above with reference to embodiments, various modifications can be made and components can be replaced with equivalents without departing from the scope of the invention. In particular, as long as there is no structural conflict, the features in the disclosed embodiments can be combined with each other in any manner. The lack of an exhaustive description of these combinations in this specification is merely for the sake of brevity and resource conservation. Therefore, the present invention is not limited to the specific embodiments disclosed herein, but includes all technical solutions falling within the scope of the claims.

Claims

1. A comprehensive energy station management system based on a photovoltaic-storage-charging microgrid, characterized in that, include: The sensing layer is used to collect real-time operational data from photovoltaic facilities, energy storage systems, charging piles, and the environment. The network layer is used for real-time transmission of data from the sensing layer, and at the same time, it establishes two-way communication with the power grid dispatch center. The application layer is used to realize energy optimization scheduling, equipment coordination, fault early warning and grid interaction based on the data transmitted from the network layer; The application layer includes: The AI ​​dynamic energy dispatch module is used to predict photovoltaic output and charging load in the next hour using LSTM neural network, and to establish a multi-objective optimization model with the goal of "curtailment rate ≤5% and minimum electricity purchase cost" by combining energy storage SOC and grid peak and valley electricity price. At the same time, the optimal dispatch strategy is solved by improving particle swarm algorithm. The multi-device collaborative control module is used to dynamically adjust the power of the charging pile, the output power of the photovoltaic inverter and the charging and discharging power of the energy storage converter, and to compensate for reactive power through the energy storage converter, adjust the power factor of the photovoltaic inverter, and maintain the microgrid voltage within the range of 380V±5%. The predictive maintenance and fault diagnosis module is used to predict the remaining battery life by collecting the internal resistance data of the energy storage battery and combining it with the number of charge and discharge cycles using a random forest algorithm. At the same time, when the battery SOH≤80% or the equipment parameters exceed the safety threshold, it issues a three-level warning and generates maintenance suggestions.

2. The integrated energy station management system based on a photovoltaic-storage-charging microgrid according to claim 1, characterized in that, The AI ​​dynamic energy scheduling module includes: The prediction model module uses an LSTM neural network to combine historical photovoltaic output data, real-time irradiance, and weather forecasts to predict photovoltaic output for the next hour; at the same time, it predicts charging load for the next hour based on historical charging data and user reservation information. The optimization model module is used to establish a multi-objective optimization equation with the objectives of "curtailment rate ≤ 5% and minimum electricity purchase cost". The inputs include predicted photovoltaic output, predicted charging load, energy storage SOC, and grid peak-valley electricity price. ,in, , These are the weighting coefficients; The solution module is used to solve for the optimal scheduling strategy using an improved particle swarm optimization algorithm, with a scheduling instruction generation cycle of 5 minutes.

3. The integrated energy station management system based on a photovoltaic-storage-charging microgrid according to claim 1, characterized in that, The multi-device collaborative control module includes: The power coordination module is used to send dynamic power commands to the charging pile according to the AI ​​scheduling strategy, and at the same time control the output power of the photovoltaic inverter and the charging and discharging power of the energy storage converter. The voltage stabilization module is used to monitor the microgrid voltage in real time. When the voltage deviates from 380V±5%, the energy storage converter automatically compensates for reactive power, and the photovoltaic inverter adjusts the power factor to ensure that the voltage is stable within the national standard range.

4. The integrated energy station management system based on a photovoltaic-storage-charging microgrid according to claim 1, characterized in that, The predictive maintenance and fault diagnosis module includes: The status monitoring module is used to collect internal resistance data through the internal resistance sensor of the energy storage battery and, in combination with the number of charge-discharge cycles, establish a prediction model for the remaining battery life; at the same time, it monitors the temperature of the charging pile module and the status of the photovoltaic inverter IGBT. The fault warning module is used to automatically issue a three-level warning and generate maintenance suggestions when the battery SOH≤80% and the charging pile module temperature≥85℃. The fault diagnosis module is used to establish a knowledge base for equipment faults and to locate the cause of faults by matching real-time data with the knowledge base.

5. A comprehensive energy station management system based on a photovoltaic-storage-charging microgrid according to claim 1, characterized in that, The application layer also includes: The user interaction and charging service optimization module supports users in reserving charging time slots and automatically coordinates the reservation time slots when multiple users reserve the same charging station. It also supports charging service value-added services and feedback.

6. A comprehensive energy station management system based on a photovoltaic-storage-charging microgrid according to claim 5, characterized in that, The user interaction and charging service optimization module includes: The charging reservation and load prediction coordination module provides a user-end APP / mini-program entry point, supports users to reserve charging time slots and select charging power, and synchronizes the reservation data to the AI ​​dynamic energy scheduling module in real time. The charging conflict coordination and intelligent navigation module is used to automatically coordinate the reservation time based on the charging time priority and scheduling feasibility when multiple users reserve the same charging pile. It also guides users to the available charging pile through the navigation function, reducing on-site waiting time. The charging service value-added and feedback closed-loop module is used to support photovoltaic green electricity traceability charging and time-of-use electricity price reminders. At the same time, it receives equipment fault feedback from users and automatically synchronizes it to the predictive maintenance and fault diagnosis module, forming a user feedback-fault location-maintenance closed loop.

7. A comprehensive energy station management system based on a photovoltaic-storage-charging microgrid according to claim 1, characterized in that, The application layer also includes: The demand response and grid interaction module is used to control the full-power charging of the energy storage system during off-peak hours and to discharge and charge the energy storage system during peak hours. At the same time, it works in conjunction with the grid dispatch center to increase the energy storage discharge power and provide frequency regulation services.

8. A comprehensive energy station management system based on a photovoltaic-storage-charging microgrid according to claim 7, characterized in that, The demand response and power grid interaction module includes: The peak-valley arbitrage module is used to receive peak-valley electricity price signals from the power grid, so as to control the energy storage system to charge at full power during the valley period and use the energy storage to discharge and charge during the peak period, thereby reducing the purchase of electricity from the power grid. The ancillary service module is used to work in conjunction with the power grid dispatch center. When the power grid issues a peak-shaving demand, it increases the energy storage discharge power and reduces the purchase of electricity. When the power grid issues a frequency regulation demand, the energy storage converter responds quickly to provide primary frequency regulation services and obtains ancillary service revenue.

Images