Adjustment method and system for polymerizing photovoltaic, energy storage and charging pile resources

Abstract

The invention belongs to the technical field of power distribution automation, and particularly relates to an adjusting method and system for aggregating photovoltaic, energy storage and charging pile resources. A multi-time scale optimization step; a real-time rolling optimization and instruction generation step; and an instruction execution and security check step. According to the invention, roof photovoltaic, user side energy storage and electric vehicle charging piles which are accessed disorderly and are difficult to manage originally are aggregated into a unified virtual scheduling whole, so that the problems of reverse heavy overload, node voltage out-of-limit and the like of a distribution transformer caused by reverse power transmission are effectively solved; and the safe bearing capacity and the operation stability of the power distribution network are obviously improved.

Description

Technical Field

[0001] This invention belongs to the field of power distribution automation technology, specifically relating to a method and system for regulating resources that integrate photovoltaic, energy storage, and charging piles. Background Technology

[0002] Photovoltaic power generation is characterized by significant intermittency and volatility. Its large-scale, high-proportion integration into traditional distribution networks poses a severe challenge to the operation and control of existing power grids. Taking some areas as an example, the installed capacity of distributed photovoltaic power has exceeded 100% of the local load demand, leading to frequent problems such as distribution transformers being severely overloaded due to reverse power flow and voltage exceeding limits at distribution nodes during peak photovoltaic power generation periods. These issues seriously threaten the safe and stable operation of the power grid and power quality. This phenomenon profoundly exposes the limitations of traditional distribution networks in planning, design, and operation control: they lack the ability to effectively aggregate, accurately sense, and coordinate the massive, dispersed, and heterogeneous flexible resources (such as photovoltaics, energy storage, and charging piles) on the low-voltage side of distribution areas.

[0003] Currently, common solutions to the aforementioned problems often focus on the independent regulation of single-type resources or rely on local control strategies with delayed responses and a lack of coordination. For example, simply limiting photovoltaic output will reduce the utilization rate of new energy sources, resulting in "curtailment"; while independently adjusting energy storage or charging piles can alleviate some pressure, it is difficult to achieve system-level global optimization. Microgrid technology is considered a key path to solving resource coordination, its core being the integration of previously independently operating resources such as photovoltaics, energy storage, and charging piles into a controllable whole. However, how to achieve efficient aggregation and multi-timescale coordinated optimization of these heterogeneous resources, especially under the premise of meeting the safety operation constraints of the distribution area (such as transformers not being overloaded and voltage not exceeding limits), while simultaneously improving the local photovoltaic absorption rate and the economic efficiency of user electricity consumption, remains a challenge for current technology. Existing technologies lack a coordinated regulation method that can balance global optimization and real-time response, as well as safety constraints and economic goals. Summary of the Invention

[0004] The purpose of this invention is to provide a method and system for regulating photovoltaic, energy storage and charging pile resources to solve the problems existing in the background art.

[0005] To achieve the above objectives, the present invention provides the following technical solution: a method for regulating resources that integrate photovoltaic, energy storage, and charging piles, comprising the following steps:

[0006] S1: Data sensing and uploading steps: The intelligent gateway controller collects real-time operating data of photovoltaic, energy storage and charging piles, calculates the net power of the transformer area and the state of charge of energy storage, and uploads it to the cloud aggregation and control platform;

[0007] S2: Multi-timescale optimization steps: The cloud-based aggregation and control platform, based on uploaded operating data, ultra-short-term photovoltaic power generation forecasts, and charging load forecasts, takes the safety of the transformer area as the primary constraint and aims to maximize photovoltaic consumption and minimize user electricity costs. It performs optimization calculations, generates a resource reference operation plan, and sends it to the smart gateway controller.

[0008] S3: Real-time rolling optimization and instruction generation steps: Based on the issued reference plan and local real-time data, the intelligent gateway controller uses model predictive control method to perform rolling optimization and generate collaborative control instructions for charging piles, energy storage systems and photovoltaic inverters;

[0009] S4: Instruction execution and security verification steps: After the intelligent gateway controller performs local security verification on the collaborative control instruction, it sends it to the corresponding terminal device for execution.

[0010] Preferably, in the data sensing and uploading step:

[0011] The operational data includes: the output power of the i-th photovoltaic inverter at time t. The charging and discharging power of the j-th energy storage system at time t The real-time power demand of the k-th charging pile at time t

[0012] The net power P of the transformer area total The formula for calculating (t) is:

[0013]

[0014] The energy storage state of charge (SOC) ESS The formula for calculating (t) is: Where E ESS (t) represents the remaining energy of the energy storage system at time t. This refers to the rated capacity of the energy storage system.

[0015] Preferably, the data sensing and uploading step further includes a security status determination:

[0016] When the net power of the transformer area meets the requirements When this occurs, an emergency load reduction strategy is triggered; among which... This refers to the rated power of the distribution transformer;

[0017] When the energy storage state of charge satisfies SOC ESS When (t) < 20%, the discharge operation of the energy storage system is forcibly locked.

[0018] Preferably, the ultra-short-term photovoltaic power generation forecast in the multi-timescale optimization step is a rolling forecast of photovoltaic power generation for the next 15 minutes to 4 hours based on historical data, real-time meteorological information and machine learning algorithms by the cloud-based aggregated control platform.

[0019] Preferably, in the multi-timescale optimization step, when the cloud-based aggregation and control platform establishes an optimization model and solves the resource reference operation plan, the adjustment priority followed is: first, adjust the charging load; second, schedule the energy storage system; and finally, limit the photovoltaic output.

[0020] Preferably, the model predictive control method in the real-time rolling optimization and instruction generation step performs rolling optimization with a cycle of 5-15 minutes, and dynamically adjusts the power setpoints of each resource by solving the optimal control problem in the finite time domain.

[0021] Preferably, in the instruction execution and security verification step, the generated collaborative control instruction includes:

[0022] The power adjustment command or start / stop time suggestion command is generated for the charging pile;

[0023] Generate charge / discharge power commands for the energy storage system;

[0024] A command to limit the maximum output of the photovoltaic inverter.

[0025] Preferably, the method further includes closed-loop feedback and adaptive optimization steps:

[0026] The intelligent gateway controller monitors the actual state changes of the command execution background area and feeds the data back to the cloud aggregation and control platform;

[0027] The cloud-based aggregation and control platform compares the actual operating status with the optimization expectations. If the deviation exceeds the threshold, it triggers the parameter self-tuning of the optimization algorithm or starts a new cycle of optimization calculation.

[0028] Preferably, in the closed-loop feedback and adaptive optimization step, the cloud-based aggregation and control platform uses machine learning algorithms to continuously correct the prediction model and optimization strategy based on data accumulated over a long period of operation.

[0029] This invention also discloses a regulating system for aggregating photovoltaic, energy storage, and charging pile resources to implement the method, employing a cloud-edge-device collaborative hierarchical control architecture. The system includes:

[0030] The cloud-based aggregation and control platform, deployed in a cloud computing center or power data center, is used to perform multi-timescale optimization calculations;

[0031] At least one smart gateway controller is installed as an edge device in the distribution radio area for data acquisition, real-time calculation, rolling optimization and command issuance;

[0032] Multiple terminal devices, including photovoltaic inverters, energy storage converters, and smart charging piles, are connected to the smart gateway controller via a communication network;

[0033] The intelligent gateway controller is connected to the cloud-based aggregation and control platform via a communication network.

[0034] The beneficial effects of this invention are:

[0035] 1) Addressing the urgent need for safe and stable operation of distribution substations with high photovoltaic (PV) integration, this invention achieves precise perception and collaborative control of massive distributed resources. Through a "cloud-edge-device" collaborative architecture, it aggregates previously disordered and difficult-to-manage rooftop PV, user-side energy storage, and electric vehicle charging piles into a unified virtual scheduling system. This effectively solves problems such as reverse overload of distribution transformers and node voltage exceeding limits caused by power backfeeding, significantly improving the safety carrying capacity and operational stability of the distribution network.

[0036] 2) A coordinated regulation method based on multi-timescale optimization and model predictive control (MPC) is proposed to improve the local consumption rate of new energy and the economic efficiency of user electricity consumption. This method dynamically adjusts the charging plan of charging piles and the charging and discharging strategy of energy storage through day-ahead planning, intraday rolling, and real-time correction, and only applies flexible power curtailment to photovoltaic power when necessary, prioritizing load matching and power generation. This ensures that most of the surplus photovoltaic power is consumed locally within the distribution area, effectively reducing the phenomenon of "curtailment" and lowering the overall electricity cost for users.

[0037] 3) A cloud-edge collaborative closed-loop control architecture was constructed, improving the real-time performance and reliability of system control. By placing complex global optimization calculations in the cloud and fast-response real-time control at the edge, this invention utilizes the powerful computing capabilities of the cloud while ensuring autonomous closed-loop control capabilities at the edge in extreme situations such as communication interruptions. This achieves millisecond-level response to changes in the state of the transformer area, ensuring the timeliness of regulation and the robustness of the system.

[0038] 4) A dynamic adjustment priority mechanism with safety as its core is introduced to optimize the utilization efficiency of flexible resources. This invention establishes an adjustment sequence of "adjusting charging load first, then scheduling energy storage, and finally limiting photovoltaic power." This strategy, while ensuring grid security, respects users' electricity consumption intentions to the greatest extent, fully leverages the adjustment potential of flexible resources such as charging piles and energy storage, and achieves optimal safety control at the lowest economic cost.

[0039] 5) A closed-loop feedback-based continuous system performance evaluation and self-learning mechanism was established, enabling online evaluation and early warning of network security and operational efficiency. By continuously comparing optimization targets with actual operating results, the system can evaluate the effectiveness of the current control strategy in real time and use historical data for self-learning to correct the prediction model. When operational indicators deviate from safety thresholds or performance declines, the system can issue timely warnings and automatically adjust strategies, thereby ensuring the long-term efficient and stable operation of the aggregated control system. Attached Figure Description

[0040] Figure 1 This is the data sensing and adjustment logic diagram in this invention. Detailed Implementation

[0041] The present invention will now be described in further detail with reference to specific embodiments.

[0042] To address the refined regulation needs of distribution substations in the context of new power systems, which require high-proportion photovoltaic integration and diversified load growth, "cloud-edge-device" collaborative control with real-time sensing, intelligent decision-making, and collaborative execution capabilities has become crucial. This system demands extremely high security in terms of data real-time performance, control reliability, and policy adaptability. Unlike traditional independent, decentralized, and slow-responding local control methods, this approach requires unified aggregation and dynamic optimization of massive heterogeneous terminals, placing specific demands on low latency, high reliability of communication networks, and the computing power of edge nodes.

[0043] like Figure 1 As shown, this invention is based on a hierarchical control architecture of "cloud-edge-device" collaboration. With the core objective of simultaneously satisfying the safe operation of distribution transformer areas and efficient energy utilization, it proposes a collaborative regulation method combining multi-timescale optimization and model predictive control. It consists of a cloud-based aggregation control platform, intelligent gateway controllers installed in the distribution transformer areas, and widely distributed photovoltaic inverters, energy storage converters, and smart charging piles. Through resource virtualization and collaborative optimization technologies, it aggregates dispersed source-storage-load resources into a controllable and scalable whole, effectively solving the problems of voltage exceeding limits, transformer overload, and photovoltaic power consumption in distribution transformer areas.

[0044] 1. System hardware deployment and establishment of the data perception layer

[0045] Cloud platform deployment: Deploy an aggregated control platform in a cloud computing center or power data center. This platform has high-performance computing capabilities and is used to run complex optimization algorithms and big data analysis.

[0046] Edge gateway deployment: Install a smart gateway controller in the distribution transformer area (low-voltage side of the distribution transformer) where fine-grained control is required. This controller needs to have multiple power data acquisition ports, strong edge computing capabilities, and at least two communication modules (such as 4G / 5G, fiber optic Ethernet, power line carrier communication HPLC, or low-power wireless).

[0047] Terminal equipment upgrade and connection: Connect the photovoltaic inverters, energy storage converters, and smart charging piles in the transformer substation to the smart gateway via the communication network. These terminal devices need to support standard communication protocols (such as Modbus, MQTT, IEC 104, etc.) and be able to receive and execute power control commands from the upper-level system.

[0048] Formula for real-time sensing of operating status:

[0049]

[0050] in:

[0051] P total (t): Dynamically calculates the net power of the distribution area (unit: kW), and determines the grid interaction demand (electricity purchase / sale) in real time.

[0052] The output power (in kW) of the i-th photovoltaic inverter at time t is read through the Modbus register address 0x0004 collected by the smart gateway.

[0053] The charging and discharging power (unit: kW) of the j-th energy storage system is obtained through CAN bus message ID 0x18FFA001, with discharge being positive and charging being negative.

[0054] The real-time power demand (in kW) of the kth charging pile is reported via the MeterValues ​​message in the OCPP protocol.

[0055] E ESS (t): The remaining energy (kWh) of the energy storage system at time t, which is read through the 0x2F04 register of the BMS.

[0056] The rated capacity of an energy storage system (e.g., 100kWh) is an inherent parameter of the equipment.

[0057] The gateway needs to calculate the net power P of the transformer area in real time. total (t) and energy storage SOC are uploaded to the cloud via a communication protocol (Modbus / TCP message format: [device address][function code][data address][CRC check]).

[0058] when (Transformer load rate exceeds 80%), triggering emergency load reduction strategy.

[0059] When SOC ESS When (t) < 20%, the energy storage discharge is forcibly locked to prevent the battery from being damaged by over-discharge.

[0060] 2. Panoramic data acquisition and multi-timescale prediction

[0061] Real-time data acquisition: The smart gateway controller continuously collects and uploads the following key data:

[0062] Distribution area status: transformer load rate, three-phase voltage, current, and power factor.

[0063] Photovoltaic power generation: Real-time output power, operating status, and power generation of each inverter.

[0064] Energy storage system: Current state of charge (SOC), maximum charge / discharge power, and health status.

[0065] Charging load: Real-time power of each charging station, user-defined expected charging completion time, and current battery capacity (SOC) of the vehicle.

[0066] Ultra-short-term forecasting: The cloud platform performs rolling ultra-short-term forecasts of photovoltaic power generation (15 minutes to 4 hours in the future) based on historical data, real-time meteorological information (irradiance, cloud cover) and machine learning algorithms. At the same time, it forecasts charging load based on historical charging behavior data and real-time requests.

[0067] 3. Multi-timescale optimization and model predictive control strategies

[0068] Daily / Intraday Optimization (Cloud-based): Based on forecast data, the aggregated control platform establishes an optimization model with a 24-hour or several-hours-forward cycle and multiple objectives, prioritizing transformer area safety, maximizing photovoltaic (PV) consumption, and minimizing user electricity costs. This model solves for the optimal reference operating plan for each resource, with the core adjustment priority being: first, adjust charging load → second, dispatch energy storage → finally, limit PV output. This plan is then distributed to the transformer area's smart gateway as the execution benchmark.

[0069] Real-time rolling optimization (edge): Based on a reference plan distributed from the cloud and combined with locally collected real-time data, the intelligent gateway controller uses the Model Predictive Control (MPC) method to perform rolling optimization every 5-15 minutes. The MPC controller solves a finite-time optimal control problem and dynamically adjusts the power setpoints of each resource to cope with prediction errors and unexpected situations, achieving precise real-time control.

[0070] 4. Generation and execution of collaborative control instructions

[0071] Command generation: Control commands are classified according to resource characteristics.

[0072] For charging piles: Generate power adjustment instructions (such as reducing the power from 7kW to 3kW) or start / stop time suggestion instructions (to guide charging during peak photovoltaic power generation periods).

[0073] For energy storage systems: Generate charge / discharge power commands (e.g., charge at 20kW for 2 hours).

[0074] For photovoltaic inverters: In extreme cases, generate maximum output limit commands (such as limiting output power to 80% of rated capacity).

[0075] Security Verification and Command Issuance: Before executing a command, the smart gateway performs a local security verification to prevent command conflicts or devices exceeding their limits. After successful verification, the command is issued to each terminal device for execution within milliseconds via a reliable communication channel.

[0076] 5. Closed-loop feedback and adaptive optimization

[0077] Performance monitoring: The smart gateway continuously monitors the actual status changes in the command execution background area, such as whether the transformer load has returned to a safe range, whether the voltage is stable, and the actual photovoltaic absorption rate.

[0078] Feedback and Correction: This real-time performance data is fed back to the cloud platform. The platform compares the actual operating status with the optimization expectations. If there is a significant deviation, it triggers the self-tuning of the optimization algorithm's parameters or starts a new cycle of optimization calculations.

[0079] Adaptive learning: The system accumulates data through long-term operation and uses machine learning algorithms to continuously correct prediction models and optimize strategies, thereby improving adaptability and control accuracy for future scenarios.

[0080] It should be noted that those skilled in the art can make various improvements and modifications without departing from the principles of this invention, and these improvements and modifications should also be considered within the scope of protection of this invention.

Claims

1. A method for regulating the aggregation of photovoltaic, energy storage, and charging pile resources, characterized in that, The method comprises the following steps: S1: data sensing and uploading step: real-time collection of operation data of photovoltaic, energy storage and charging piles by the intelligent gateway controller, calculation of net power of the transformer area and state of charge of the energy storage, and uploading to the cloud aggregation control platform; S2: multi-time scale optimization step: based on the uploaded operation data, ultra-short-term photovoltaic power generation prediction and charging load prediction, the cloud aggregation control platform performs optimization calculation to maximize photovoltaic power consumption and minimize user electricity cost, generates a resource reference operation plan, and sends the plan to the intelligent gateway controller; S3: real-time rolling optimization and instruction generation step: based on the reference plan and local real-time data, the intelligent gateway controller performs rolling optimization by using a model predictive control method to generate collaborative control instructions for charging piles, energy storage systems and photovoltaic inverters; S4: instruction execution and safety checking step: after the intelligent gateway controller performs local safety checking on the collaborative control instructions, the instructions are sent to corresponding terminal devices for execution.

2. The method of claim 1, wherein, In the data sensing and uploading step: The operation data includes: output power of the i-th photovoltaic inverter at t moment charge-discharge power of the j-th energy storage system at t moment and real-time power demand of the k-th charging pile at t moment The net power P of the transformer area total The calculation formula of (t) is: The energy storage state of charge SOC ESS The calculation formula of (t) is: Wherein E ESs (t) is the remaining energy of the energy storage system at time t, is the rated capacity of the energy storage system.

3. The method of claim 2, wherein, In the data sensing and uploading step, a safety state judgment is further included: When the net power of the transformer area satisfies , an emergency load shedding strategy is triggered; wherein is the rated power of the distribution transformer. When the energy storage state of charge satisfies SOC Ess (t) < 20%, the discharge operation of the energy storage system is forcibly locked.

4. The method of claim 1, wherein, The ultra-short-term photovoltaic power generation prediction in the multi-time scale optimization step is a rolling prediction of photovoltaic power generation in the next 15 minutes to 4 hours by the cloud aggregation control platform based on historical data, real-time weather information and machine learning algorithms.

5. The method of claim 1, wherein, In the multi-time scale optimization step, when the cloud aggregation control platform establishes an optimization model and solves the resource reference operation plan, the adjustment priority followed is: adjusting charging load first, then scheduling energy storage systems, and finally limiting photovoltaic output.

6. The method of claim 1, wherein, The model predictive control method in the real-time rolling optimization and instruction generation step performs rolling optimization with a period of 5-15 minutes, dynamically adjusts the power set value of each resource by solving the optimal control problem in a limited time domain.

7. The method of claim 1, wherein, In the instruction execution and safety checking step, the generated collaborative control instructions include: power adjustment instructions or start / stop time suggestion instructions for the charging piles; charge / discharge power instructions for the energy storage systems; maximum output limit instructions for the photovoltaic inverters.

8. The method of claim 1, wherein, The method further comprises a closed-loop feedback and adaptive optimization step: The intelligent gateway controller monitors the actual state changes of the transformer area after instruction execution, and feeds back the data to the cloud aggregation control platform; The cloud aggregation control platform compares the actual operation state with the optimization expectation, and if the deviation exceeds a threshold, triggers parameter self-tuning of the optimization algorithm or starts a new cycle of optimization calculation.

9. The method of claim 8, wherein, In the closed-loop feedback and adaptive optimization step, the cloud aggregation control platform uses machine learning algorithms to continuously correct the prediction model and optimization strategy based on long-term operation accumulated data.

10. An adjusting system for realizing the method of any one of claims 1-7 of the polymeric photovoltaic, energy storage, charging pile resources, adopting a cloud-edge-end collaborative hierarchical control architecture, characterized in that, The system comprises: a cloud aggregation control platform deployed in a cloud computing center or a power data center for performing multi-time scale optimization calculation; at least one intelligent gateway controller installed as an edge device in a power distribution transformer area for data collection, real-time calculation, rolling optimization and instruction sending. A plurality of terminal devices, including photovoltaic inverters, energy storage converters and intelligent charging piles, are connected with the intelligent gateway controller through a communication network. The intelligent gateway controller is connected with the cloud aggregation control platform through a communication network.

Images