Optical storage and charging integrated power station monitoring system and control method

Abstract

The invention discloses an optical storage and charging integrated power station monitoring system, and belongs to the technical field of energy management and electric power automation. The system adopts a layered architecture and comprises a physical layer, a data acquisition layer, a cloud edge collaboration platform layer and an application layer. The core of the method is that an edge computing node with a built-in digital twin model and a dynamic reconstruction decision engine is deployed in the cloud edge collaboration platform layer. The digital twin model is used for performing real-time mapping and state prediction on photovoltaic power generation, energy storage, charging piles and power distribution units; and the dynamic reconstruction decision engine generates a cooperative control instruction through a multi-target optimization algorithm based on the prediction result, the real-time electricity price and the operation target, realizes intelligent scheduling and distribution of power station energy, and executes topology dynamic reconstruction to guarantee power supply reliability when the system fails. According to the method, the problems of information island, poor collaboration, weak power grid supporting capability and the like of the existing system are solved, and the economical efficiency, the self-adaptive capability and the operation toughness of the light storage and charging integrated power station are remarkably improved.

Description

Technical Field

[0001] This invention relates to the field of energy management and power automation technology, and in particular to a monitoring system and control method for an integrated photovoltaic, energy storage and charging power station. Background Technology

[0002] With the advancement of the "dual-carbon" strategy, integrated photovoltaic-storage-charging power stations have rapidly developed as an important component of the new power system. These power stations typically include photovoltaic power generation units, energy storage battery systems, multiple electric vehicle charging piles, and grid-connected / off-grid switching devices. However, existing monitoring systems suffer from the following significant drawbacks: Information silos and poor collaboration: Photovoltaic, energy storage, charging, and power distribution systems often come from different suppliers, each with its own independent monitoring platform. Inconsistent data formats and communication protocols make efficient data exchange and collaborative control difficult between systems. This hinders globally optimal energy dispatching based on real-time electricity prices, photovoltaic output, load demand, and battery status.

[0003] Rigid operating mode: Existing systems mostly adopt preset and fixed operating strategies (such as "self-generation and self-consumption, surplus power storage"), lacking the ability to adapt to and predict multi-dimensional dynamic factors (such as sudden weather changes, drastic fluctuations in electricity prices, and sudden high-power charging demand), resulting in poor economic efficiency or unstable system operation.

[0004] Weak support capacity for the power grid: Most systems only consider purchasing electricity from or selling electricity to the power grid, lacking the ability to actively participate in the power grid's demand-side response and provide ancillary services such as peak shaving and frequency regulation, thus failing to fully leverage the supporting role of distributed energy resources for the power grid.

[0005] Insufficient fault diagnosis and recovery capabilities: When a part of the system fails, the existing monitoring system can usually only provide alarms and isolation, lacking intelligent reconfiguration and power restoration strategies based on the global state of the system, which reduces the reliability and resilience of the entire power station.

[0006] Therefore, there is an urgent need in this field for an integrated photovoltaic, energy storage, and charging power plant monitoring system that can break down information silos, achieve intelligent collaboration, possess adaptive capabilities, and have grid-friendly interactive characteristics. Summary of the Invention

[0007] To address the shortcomings of existing technologies, this invention aims to provide a photovoltaic-storage-charging integrated power station monitoring system based on digital twin and dynamic reconfiguration technologies, and also provides a control method for the photovoltaic-storage-charging integrated power station monitoring system.

[0008] To achieve the above objectives, the present invention adopts the following technical solution: A photovoltaic, energy storage and charging integrated power station monitoring system includes a physical layer, a data acquisition layer, a cloud-edge collaborative platform layer and an application layer connected in sequence; The physical layer includes a photovoltaic power generation unit, an energy storage unit, a charging pile unit, and an in-station power distribution unit; The data acquisition layer is used to collect real-time operating data of each unit of the physical layer; The cloud-edge collaborative platform layer includes edge computing nodes deployed locally at the power plant and a cloud platform deployed in the cloud; the edge computing nodes have a built-in digital twin model of the power plant and a dynamic reconfiguration decision engine; The application layer is used to provide human-computer interaction and execute control commands; The digital twin model is used to perform synchronous mapping and state prediction of the physical layer based on the real-time operating data; the dynamic reconfiguration decision engine is used to generate a collaborative control instruction set based on the output of the digital twin model, the electricity price signal and the operating target, and to generate a topology dynamic reconfiguration scheme when the system fails.

[0009] Further preferably, the collaborative control instruction set generated by the dynamic reconfiguration decision engine includes control instructions for the output setting value of the photovoltaic inverter, the charging and discharging power setting value of the energy storage converter, the power allocation value of each charging pile, and the grid-connected / off-grid switching switch.

[0010] Further preferably, the operational objectives include at least one of the following: lowest operating cost, smooth power grid interaction, minimum battery life loss, and responsiveness to grid dispatch instructions. The dynamic reconfiguration decision engine employs a multi-objective optimization algorithm to solve for the optimal set of control instructions that satisfy the constraints.

[0011] In a further preferred embodiment, the cloud-edge collaborative platform layer also includes a virtual power plant interface module, which is used to aggregate the local station into a virtual power generation unit and receive and respond to power control commands issued by the upper-level management platform; the power control commands serve as a constraint condition for the optimization calculation of the dynamic reconfiguration decision engine.

[0012] More preferably, the digital twin model includes: A photovoltaic output prediction sub-model is used to predict photovoltaic power generation in future periods based on meteorological data; The load forecasting sub-model is used to predict the charging load for future periods based on historical data and real-time status. A battery health status assessment sub-model is used to assess and predict the health status of energy storage batteries. The power flow calculation sub-model is used for electrical calculations and safety verification based on the substation topology. In a further preferred embodiment, the dynamic reconfiguration decision engine adopts a priority-based and flexible power allocation mechanism when generating charging pile power allocation values. Specifically, it calculates the theoretical power requirement of each charging pile based on the user-set charging demand time and the current state of the vehicle battery, and allocates power according to priority order when the total power is limited.

[0013] Further preferably, the cloud platform is used to store historical operating data, periodically train and update the parameters of the prediction sub-model in the digital twin model, and distribute the updated model to the edge computing node.

[0014] Further preferably, the workflow of the dynamic reconfiguration decision engine in the event of a system failure includes: Receive fault information and locate the faulty device; The new system topology after disconnecting the faulty device is simulated in the digital twin model; Safety verification is performed using the aforementioned power flow calculation sub-model. Generate a system reconfiguration scheme that includes switch opening and closing instructions to isolate faults and restore power supply to non-faulty areas.

[0015] Further preferably, the application layer includes a monitoring interface for maintenance personnel and a vehicle owner service interface for electric vehicle users; the vehicle owner service interface provides a charging pile reservation function and transmits the user's expected charging completion time to the dynamic reconfiguration decision engine as an input parameter for charging power allocation. A control method for a monitoring system of an integrated photovoltaic-storage-charging power station includes the following steps: The data acquisition layer continuously collects real-time operational data from each unit of the physical layer; The digital twin model of the edge computing node is used to perform state synchronization and short-term prediction based on the real-time operating data; By dynamically reconstructing the decision engine, optimization problems are solved based on prediction results, electricity prices, and operational objectives, and a set of collaborative control instructions is generated. The collaborative control instruction set is sent to the corresponding devices in the physical layer for execution. When a fault is detected, a digital twin model is used to simulate the fault and verify the safety, and a dynamic reconfiguration command is generated to restore system operation.

[0016] The beneficial effects of this invention are: This invention breaks down information barriers between photovoltaics, energy storage, and charging by constructing a unified cloud-edge collaborative platform and digital twin model, achieving true global optimization and collaborative control. Utilizing the predictive capabilities of the digital twin model, combined with a dynamic reconfiguration decision engine, the system can proactively adjust its operating strategies to adapt to changes in the external environment (weather, electricity prices) and internal demand (charging load), significantly improving economic efficiency and stability. Through a virtual power plant interface, power plants can flexibly participate in the electricity market, providing ancillary services such as peak shaving and demand response, transforming from simple energy consumers into "prosumers," thus improving the project's return on investment and its grid-friendliness. The dynamic reconfiguration function enables the system to quickly and safely reconfigure power supply paths in the event of a partial fault, maximizing the continuous power supply to critical loads (such as some charging piles) and improving system robustness. By considering battery lifespan degradation in the optimization objectives and optimizing charging and discharging strategies, overcharging, over-discharging, and frequent high-power charging and discharging are avoided, effectively extending the lifespan of the energy storage system. Attached Figure Description

[0017] Figure 1 This is a schematic diagram of the overall architecture of a photovoltaic, energy storage and charging integrated power station monitoring system according to the present invention; Figure 2 This is a flowchart of the dynamic reconfiguration decision engine in a photovoltaic-storage-charging integrated power station monitoring system of the present invention.

[0018] The names corresponding to each mark in the diagram: 10. Physical layer; 11. Photovoltaic power generation unit; 12. Energy storage unit; 13. Charging pile unit; 14. Station power distribution unit; 20. Data Acquisition Layer; 30. Cloud-edge collaboration platform layer; 31. Edge computing nodes; 311, Digital Twin Model; 312, Dynamically reconstruct the decision engine; 313, Virtual Power Plant Interface Module; 32. Cloud platform; 40. Application layer; 41. Interface for operations and maintenance personnel; 42. Car owner service interface. Detailed Implementation

[0019] 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. All other embodiments obtained by those skilled in the art based on the embodiments of the present invention are within the scope of protection of the present invention.

[0020] refer to Figure 1 This embodiment provides a photovoltaic-storage-charging integrated power station monitoring system, whose overall architecture includes a physical layer 10, a data acquisition layer 20, a cloud-edge collaborative platform layer 30, and an application layer 40.

[0021] Physical layer 10 is a collection of physical equipment for the power plant, including: Photovoltaic power generation unit 11: Composed of photovoltaic modules, combiner box, inverter, etc., it converts solar energy into electrical energy.

[0022] Energy storage unit 12: It consists of an energy storage battery pack (such as a lithium-ion battery), a battery management system and an energy storage converter, which realizes the storage and release of electrical energy.

[0023] Charging station unit 13: includes multiple DC fast charging stations and AC slow charging stations to provide charging services for electric vehicles.

[0024] Station power distribution unit 14: includes transformers, switchgear, protection devices, grid-connected / off-grid switching switches, etc., responsible for power distribution, protection and grid-connected / off-grid mode switching.

[0025] The data acquisition layer 20 consists of sensors, smart meters, device controllers, and communication networks (such as industrial Ethernet, 4G / 5G, LoRa, etc.) distributed throughout the units. It is responsible for collecting: Data from photovoltaic power generation unit 11 includes: DC side voltage / current, AC side power, inverter operating status, and power generation.

[0026] Data from energy storage unit 12 includes: battery pack SOC, SOH, voltage, current, temperature, energy storage converter charging and discharging power, status, etc.

[0027] Data from charging pile unit 13 includes: real-time power, charging status, user ID, reservation information (such as expected pick-up time), and vehicle battery SOC for each charging pile.

[0028] Data from the substation's distribution unit 14 includes: power, voltage, frequency, status of each branch switch, and power quality data at the grid connection point.

[0029] All collected data is uploaded to the cloud-edge collaboration platform layer 30 via a unified communication protocol (such as MQTT, OPC UA).

[0030] The cloud-edge collaboration platform layer 30 adopts a "cloud-edge collaboration" architecture.

[0031] Edge computing node 31: Deployed at the power plant site, responsible for processing high-frequency, real-time data and providing rapid response. Its core includes: Digital Twin Model 311: This is a high-fidelity virtual model that corresponds one-to-one with Physical Layer 10. It achieves dynamic synchronization of physical entities by receiving real-time data from Data Acquisition Layer 20. This model integrates multiple sub-models: Photovoltaic output prediction sub-model: Based on historical data and real-time meteorological information (irradiance, temperature), machine learning algorithms (such as LSTM networks) are used to predict the photovoltaic power generation in the next 0-24 hours.

[0032] Load forecasting sub-model: Based on historical charging data, date type (weekday / holiday), and real-time queuing status, predict the charging load curve for the next 0-24 hours.

[0033] Battery health status assessment sub-model: Based on historical battery operating data (number of cycles, depth of charge and discharge, temperature history), it adopts a combination of empirical models and data-driven methods to evaluate and predict the SOH of energy storage batteries in real time.

[0034] Power Flow Calculation Sub-model: Based on the substation topology, it calculates the voltage, current and power distribution of each node in real time under different operating conditions for safety verification.

[0035] The dynamic reconfiguration decision engine 312 is the system's decision-making layer. Its workflow is as follows: Figure 2 As shown.

[0036] Step S1: The engine continuously receives real-time status and prediction data (such as the photovoltaic power output P_pv and load prediction P_load for the next 15 minutes) from the digital twin model 311.

[0037] Step S2: Receive external input, including real-time electricity price information, power regulation command P_vpp (if any) from virtual power plant interface module 313, and operation targets set by operation and maintenance personnel (such as "cost priority" or "smooth grid connection").

[0038] Step S3: Establish a multi-objective optimization model. Taking "lowest operating cost" and "minimum battery life loss" as examples, the objective function can be expressed as: Min [Cost_grid * P_grid * Δt + α * (Battery_Degradation) ] Where Cost_grid is the real-time electricity price, P_grid is the power interaction with the grid (positive for purchasing electricity, negative for selling electricity), Δt is the optimization time step, α is the weighting coefficient, and Battery_Degradation is the quantified value of battery loss based on the SOH model.

[0039] Step S4: Constraints include: power balance constraint P_pv + P_bess + P_grid = P_load + P_loss; upper and lower limits of energy storage SOC; energy storage charging and discharging power constraint; upper and lower limits of charging pile power constraint; grid-connected power limit constraint, etc.

[0040] Step S5: Solve the model using an optimization algorithm (such as mixed integer linear programming or particle swarm optimization) to generate a set of control instructions for a future optimization cycle (such as 15 minutes), including: the charging and discharging power setting value P_bess_ref of the energy storage unit, the power allocation value P_charge_i_ref of each charging pile, and a suggestion on whether to sell electricity to the grid.

[0041] Step S6 (Fault Reconfiguration Mode): When the data acquisition layer reports a fault in a branch (such as a short circuit in a charging pile circuit), the engine immediately initiates the fault handling process. The digital twin model 311 simulates the system's operating state under the new topology after the faulty branch switch is disconnected, performing power flow calculations and safety verifications. Based on the verification results, the dynamic reconfiguration decision engine 312 generates an optimal reconfiguration scheme, such as: closing the tie switch, allowing the energy storage system to supply power to another set of charging piles unaffected by the fault, while adjusting the output of photovoltaic and energy storage to ensure stable system operation, and reporting alarms and confirming execution to the application layer.

[0042] Virtual power plant interface module 313: This module reports the overall operating status of the station (adjustable capacity, adjustable power) to the superior virtual power plant management platform. When it receives a power command from the VPP (such as requiring a reduction of 50kW in the net power of the grid connection point within 30 minutes), this command will serve as a new constraint for the dynamic reconfiguration decision engine 312. The engine will then recalculate and optimize the calculations to meet the VPP's requirements by adjusting the energy storage charging power or appropriately limiting the charging pile power.

[0043] Cloud platform 32: Responsible for storing massive amounts of historical data, big data analysis, and model training. It can periodically use historical data to retrain and optimize the prediction model on the edge side (such as the photovoltaic power output prediction sub-model), and send the updated model parameters to the edge computing node 31 to achieve continuous model evolution.

[0044] Application layer 40 provides visual interfaces and operation entry points for different users.

[0045] Operations and maintenance personnel interface 41: Displays the overall status of the power plant (single-line diagram), real-time / historical data curves, alarm information, and performance reports in a graphical manner. Provides functions such as switching operating modes, remote equipment control (e.g., manually starting and stopping energy storage), and setting optimization targets.

[0046] Owner Service Interface 42: Through a mobile app or mini-program, users can access services such as checking charging station availability, scheduling charging, online payment, and viewing charging progress. Users can enter their desired completion time when scheduling, which will serve as an important input for the flexible allocation of charging station power.

[0047] Detailed Explanation of the Flexible Power Allocation Strategy for Charging Stations: Assume the current total power limit of the system is P_total, and there are three charging stations (C1, C2, C3) charging.

[0048] Step a: The dynamic reconfiguration decision engine 312 obtains the user settings information for each charging station, such as C1 user requiring a full charge in 1 hour, C2 user having no requirements (default minimum power), and C3 user requiring a full charge in 30 minutes.

[0049] Step b: The engine calculates the theoretical power demand P_req_i for each vehicle based on its current SOC, battery capacity, and expected time.

[0050] Step c: Compare the total power demand ΣP_req_i with the total limit P_total.

[0051] If ΣP_req_i ≤ P_total, then allocate on demand.

[0052] If ΣP_req_i > P_total, then the elastic allocation algorithm is activated: Priority sorting: Sort by "urgency" (higher priority the closer the expected time) and / or "user level". Assume the sorting is C3 > C1 > C2.

[0053] Flexible allocation: Prioritize the needs of high-priority piles. Allocate remaining power to low-priority piles. Specifically: First, we will do our best to meet the power requirement P_req_3 of C3.

[0054] Then, the remaining power (P_total - P_req_3) is allocated to C1 and C2. Since C1 also has time requirements, it can be re-allocated proportionally or according to the urgency of its needs, while C2 receives the remaining power or the minimum guaranteed power.

[0055] Step d: The engine sends the calculated P_charge_1_ref, P_charge_2_ref, and P_charge_3_ref to each charging station for execution.

[0056] This strategy achieves both meeting users' diverse needs and ensuring that the total system power does not exceed the limit under limited power resources.

[0057] Example of system workflow (sunny midday scenario): The digital twin model predicts that solar power generation will surge at midday, exceeding the charging load at the station.

[0058] The dynamic reconfiguration decision engine performs optimization calculations based on the goal of "lowest operating cost" and in conjunction with peak electricity price information.

[0059] Generate instructions: a) Prioritize photovoltaic power supply to charging loads; b) Use remaining photovoltaic power to charge energy storage batteries; c) If the energy storage is fully charged, control and switch the off-grid transfer switch to feed power back to the grid to earn revenue from electricity sales.

[0060] At the same time, the VPP platform issued an instruction requiring the station to reduce its grid-connected power by 50kW to support the stability of the local power grid.

[0061] The virtual power plant interface module receives the instruction, and the dynamic reconfiguration decision engine immediately re-optimizes it. The instruction is adjusted to: moderately reduce the output of the photovoltaic inverter (if necessary and permissible), while increasing the charging power of the energy storage system to ensure that the grid connection power is reduced by 50kW, perfectly responding to VPP dispatch.

[0062] In summary, this invention, through the deep integration of digital twin and dynamic reconstruction technologies, constructs an intelligent monitoring system that integrates perception, decision-making, and control, effectively solving many pain points in existing technologies and greatly improving the overall performance of photovoltaic-storage-charging integrated power stations.

Claims

1. A monitoring system for a light storage and charging integrated power station, characterized in that, The physical layer includes a photovoltaic power generation unit, an energy storage unit, a charging pile unit, and a station power distribution unit. The data acquisition layer is used to collect real-time operation data of each unit of the physical layer. The cloud edge collaborative platform layer includes an edge computing node deployed locally in the power station and a cloud platform deployed in the cloud. The application layer is used to provide human-computer interaction and execute control instructions. The digital twin model is used to synchronize mapping and state prediction of the physical layer based on the real-time operation data. The dynamic reconstruction decision engine generates a set of collaborative control instructions based on the output of the digital twin model, the electricity price signal, and the operation target, and generates a topology dynamic reconstruction scheme when the system fails. 2.The monitoring system of the optical storage and charging integrated power station according to claim 1, characterized in that, The dynamic reconstruction decision engine generates a set of collaborative control instructions including control instructions for photovoltaic inverter output set value, energy storage converter charge and discharge power set value, each charging pile power distribution value, and on-off grid switch. 3.The monitoring system of the optical storage and charging integrated power station according to claim 2, characterized in that, The operation target includes at least one of the lowest operation cost, the smoothest grid interaction power, the smallest battery life loss, and the response to grid dispatching instructions. 4.The monitoring system of the optical storage and charging integrated power station according to claim 3, characterized in that, The cloud edge collaborative platform layer also includes a virtual power plant interface module for aggregating the station as a virtual power generation unit, receiving and responding to power control instructions issued by the superior management platform. 5.The monitoring system of the optical storage and charging integrated power station according to claim 4, characterized in that, The digital twin model includes: A photovoltaic output prediction sub-model for predicting photovoltaic power generation power in future periods based on weather data. A load prediction sub-model for predicting future charging load based on historical data and real-time state. A battery health state evaluation sub-model for evaluating and predicting the health state of energy storage batteries. A power grid flow calculation sub-model for electrical calculation and safety verification based on station topology. 6.The monitoring system of the optical storage and charging integrated power station according to claim 5, characterized in that, The dynamic reconstruction decision engine uses a priority-based and flexible power allocation mechanism when generating charging pile power distribution values. 7.The monitoring system of the optical storage and charging integrated power station according to claim 6, characterized in that, The cloud platform is used to store historical operation data and periodically train and update parameters of prediction sub-models in the digital twin model, and issue updated models to the edge computing node. 8.The monitoring system of the optical storage and charging integrated power station according to claim 7, characterized in that, The workflow of the dynamic reconstruction decision engine when the system fails includes: Receive fault information and locate fault equipment. Simulate the new system topology after disconnecting the fault equipment in the digital twin model. Perform safety verification through the power grid flow calculation sub-model. Generate a system reconstruction scheme including switch opening and closing instructions to isolate faults and restore power supply in non-fault areas. 9.The monitoring system of the optical storage and charging integrated power station according to claim 8, characterized in that, The application layer comprises a monitoring interface for operation and maintenance personnel and a vehicle owner service interface for electric vehicle users; the vehicle owner service interface provides a charging pile reservation function and transmits a desired charging completion time input by a user to the dynamic reconstruction decision engine as an input parameter for charging power allocation.

10. A control method for the monitoring system of the integrated optical storage and charging power station according to any one of claims 1-9, characterized in that, The method comprises the following steps: Real-time operation data of each unit of the physical layer is continuously collected through the data collection layer; State synchronization and short-term prediction are performed according to the real-time operation data through a digital twin model of the edge computing node; Based on the prediction result, the electricity price and the operation target, an optimization problem is solved through the dynamic reconstruction decision engine to generate a set of coordinated control instructions; The set of coordinated control instructions is issued to the corresponding devices of the physical layer for execution; When a fault is detected, fault simulation and safety checking are performed through the digital twin model, and a dynamic reconstruction instruction is generated to restore system operation.

Images