Optimized operation method for improving carrying capacity of distribution network with integrated coordination of light storage and charging

Abstract

The application discloses a power distribution network carrying capacity improvement optimal operation method of integrated photovoltaic energy storage and charging collaborative regulation, relates to the technical field of power distribution network operation control and distributed energy consumption, and comprises the following steps: collecting data, establishing a probabilistic power flow model, calculating voltage and thermal carrying capacity margin, determining a dynamic carrying capacity threshold value; constructing a hierarchical collaborative regulation framework; in the day-ahead and day-ahead stage, rolling optimization is carried out with the maximum carrying capacity margin expectation as the target, and preventive regulation is triggered when the margin is lower than the threshold value; in real-time operation, when the margin is lower than the safety threshold value, emergency collaborative control is executed according to the three-level ladder sequence of charging pile power active response, energy storage dynamic virtual capacity expansion, photovoltaic reactive power and active power reduction; the state quantity after regulation is fed back and optimized, and a closed-loop self-optimizing operation is formed. The power distribution network carrying capacity improvement optimal operation method of integrated photovoltaic energy storage and charging collaborative regulation realizes dynamic quantitative evaluation and active improvement of the carrying capacity of the power distribution network, and significantly improves the consumption capacity and operation safety of the photovoltaic energy storage and charging system.

Description

Technical Field

[0001] This invention relates to the fields of distribution network operation control and distributed energy consumption technology, specifically a method for improving and optimizing the operation of distribution networks by integrating photovoltaic, energy storage and charging technologies for coordinated regulation. Background Technology

[0002] With the advancement of the "dual carbon" goals, the integrated development of distributed photovoltaic, energy storage, and electric vehicle charging stations is progressing rapidly. While the large-scale integration of photovoltaic, energy storage, and charging systems into the power grid has promoted the local consumption of new energy to some extent, it has also brought severe challenges such as voltage exceeding limits, transformer overload, and three-phase imbalance. Existing technologies have the following main shortcomings in addressing these issues:

[0003] Traditional distribution network capacity assessments are mostly based on the N-1 safety criterion or steady-state power flow sections, lacking dynamic quantification of uncertainties such as photovoltaic output fluctuations and the randomness of charging loads. This leads to overly conservative or overly aggressive assessment results, failing to reflect the grid's true absorption capacity in real time. Existing photovoltaic-storage-charging synergy often employs fixed power factor control or simple peak-shaving and valley-filling strategies, lacking closed-loop feedback on key capacity indicators such as distribution network node voltage sensitivity and line thermal stability limits. Each device operates independently, failing to fully leverage the flexible adjustment capabilities of energy storage and charging piles to proactively improve grid capacity. Traditional day-ahead optimization and real-time control are independent of each other. When photovoltaic output or charging load changes suddenly and drastically, the intraday adjustment capability is insufficient, easily leading to control lag, frequent equipment operation, or exceeding safety constraints. Summary of the Invention

[0004] The purpose of this invention is to provide a method for optimizing the operation of distribution network carrying capacity through integrated photovoltaic, energy storage and charging control, in order to solve the technical problems mentioned in the background art, such as inaccurate carrying capacity assessment, lack of proactive improvement capability in coordinated control, and multi-timescale response mismatch.

[0005] To achieve the above objectives, the present invention provides the following technical solution: a method for optimizing the operation of a distribution network by integrating photovoltaic, energy storage, and charging technologies for coordinated control, comprising the following steps:

[0006] Step S1: Dynamic bearing capacity assessment: Collect distribution network operation data and meteorological forecast data, establish a probabilistic power flow model, calculate voltage bearing capacity margin and thermal bearing capacity margin, and determine the dynamic bearing capacity threshold through Monte Carlo simulation;

[0007] Step S2: Construct a layered architecture: Establish a collaborative control architecture that includes a station domain layer and an equipment layer. The station domain layer issues control commands to the equipment layer based on the dynamic carrying capacity threshold obtained in step S1.

[0008] Step S3: Long-term rolling optimization: In the day-ahead phase, optimize the energy storage charging and discharging plan and the charging pile power limit with the goal of maximizing the overall carrying capacity margin expectation. In the intraday phase, roll the plan based on ultra-short-term forecasts and trigger preventive regulation when the current carrying capacity margin is lower than the dynamic carrying capacity threshold.

[0009] Step S4: Short-time scale emergency coordinated control: In real-time operation, when the load capacity margin is lower than the safety threshold, coordinated control is executed in a three-level ladder order of active power response of charging pile, dynamic virtual capacity expansion of energy storage, and reactive power of photovoltaic inverter, until the load capacity margin rises back to above the safety threshold.

[0010] Step S5: Closed-loop feedback: Feedback the actual distribution network state variables after adjustment in step S4 to the probabilistic power flow model in step S1, update the dynamic carrying capacity threshold, and re-input it to steps S2 and S3 to form a closed-loop self-optimizing operation of evaluation, decision-making, control, and feedback.

[0011] Furthermore, the voltage carrying capacity margin in step S1 and thermal load margin Calculated using the following formulas respectively: ;

[0012] in, For nodes Current voltage; branch road Current current; , , These are the upper voltage limit, lower voltage limit, and rated current, respectively. A set of nodes; For the set of branches.

[0013] Furthermore, in step S1, the bearing capacity margin probability distribution under various typical scenarios is generated through Monte Carlo simulation, and the 5th percentile is taken as the dynamic bearing capacity threshold.

[0014] Furthermore, the objective function for the day-ahead stage in step S3 is:

[0015]

[0016] in, Optimize the total number of time periods for the current day; and Time periods Voltage bearing capacity margin and thermal bearing capacity margin; These are the weighting coefficients; For system operating costs; This represents the cost penalty coefficient.

[0017] Furthermore, the triggering condition for preventive regulation in step S3 is: the current bearing capacity margin is lower than the dynamic bearing capacity threshold, and the ultra-short-term forecast shows that the bearing capacity margin will continue to decline in the next two control cycles.

[0018] Furthermore, the derating range of the charging pile's active power response in step S4 is determined by the following formula:

[0019] in: The current charging power of the charging station; The rated power of the charging pile; This is the control coefficient; This represents the thermal load-bearing capacity margin for the current period.

[0020] Furthermore, the power regulation law for the dynamic virtual capacity expansion of energy storage in step S4 is as follows:

[0021] ;in, This is the active power command value for the energy storage converter; This is the proportional control coefficient; For target bearing capacity margin; This represents the current real-time bearing capacity margin; when < When, the energy storage absorbs power during charging; when > Furthermore, when there is a risk of voltage exceeding the upper limit, energy storage discharge provides support.

[0022] Furthermore, the specific execution sequence of reactive and active power reduction of the photovoltaic inverter in step S4 is as follows: first, the power factor of the photovoltaic inverter is adjusted from 1 to 0.95 within the capacitive range; if the load-bearing capacity margin has not recovered, the power factor is further adjusted to 0.9; and finally, automatic power generation control is put into operation to reduce active power.

[0023] Furthermore, the actual distribution network status quantities fed back to step S1 in step S5 include at least: the real-time voltage amplitude of each node, the real-time current amplitude of each branch, the actual charging and discharging power of the energy storage converter, the actual charging power of the charging pile group, and the actual active and reactive power output of the photovoltaic inverter.

[0024] Furthermore, the closed-loop self-optimization operation cycle is 5 to 15 minutes; within each cycle, steps S1 to S5 are executed sequentially once, and the intraday rolling correction in step S3 and the real-time emergency collaborative control in step S4 are both calculated in parallel based on the latest feedback state variables, wherein the time scale of the intraday rolling correction is 15 minutes, and the time scale of the real-time emergency collaborative control is from the second level to the minute level.

[0025] Compared with the prior art, the beneficial effects of the present invention are:

[0026] This integrated photovoltaic, energy storage, and charging (PV-ESG) coordinated regulation method for improving the carrying capacity of distribution networks achieves a fully closed-loop self-optimization operation from assessment, decision-making, control, and feedback by combining dynamic probabilistic carrying capacity quantification, hierarchical collaborative architecture, multi-timescale rolling optimization, and three-level tiered emergency control. It can proactively improve the distribution network's ability to accept high-penetration PV-ESG while taking into account economy, reliability, and response speed, providing a solution for the efficient operation of distribution networks under the new power system.

[0027] 1. Dynamic probabilistic carrying capacity assessment improves the accuracy and adaptability of the assessment: This invention uses a probabilistic power flow model and Monte Carlo simulation to quantify the impact of the uncertainty of photovoltaic output and charging load on the safety boundary of the distribution network. The 5th percentile of the probability distribution of carrying capacity margin is used as a dynamic threshold, which overcomes the shortcomings of traditional deterministic thresholds that are too conservative or overly aggressive, and makes the assessment results more in line with the actual operating conditions.

[0028] 2. Layered collaboration and multi-timescale rolling optimization to achieve global coordination and forward-looking prevention of regulation: Through the layered architecture of station area layer and equipment layer, day-ahead economic optimization, intraday rolling correction and real-time rapid response are organically linked. In the intraday stage, preventive regulation is triggered in advance based on ultra-short-term forecasts to avoid control lag in extreme scenarios and greatly improve the collaborative operation capability of photovoltaic energy storage and charging system and distribution network.

[0029] 3. Three-tiered emergency coordinated control, taking into account economy, safety and control flexibility: The priority order of charging pile devaluation, energy storage virtual capacity expansion and photovoltaic reduction is executed. Low-cost flexible loads are called first, followed by energy storage for rapid response, and photovoltaic reduction is used last. Under the premise of ensuring rapid recovery of load capacity, the curtailment loss and equipment operation cost are minimized.

[0030] 4. Closed-loop feedback self-optimization mechanism to continuously improve the adaptive capability of the operation strategy: The actual state quantities after real-time adjustment, such as voltage, current, and power, are fed back to the probabilistic power flow model, dynamically update the load capacity threshold and re-input it to the optimization and control module, so that the system can continuously self-correct and evolve with changes in operating conditions, significantly enhancing its robustness to changes in long-term operating scenarios. Attached Figure Description

[0031] Figure 1 This is an overall flowchart of the method of the present invention;

[0032] Figure 2 This is a schematic diagram of the probability distribution of the dynamic bearing capacity margin of the present invention;

[0033] Figure 3 This is a schematic diagram of the hierarchical collaborative control architecture of the present invention;

[0034] Figure 4 This is a three-level tiered emergency control logic diagram of the present invention. Detailed Implementation

[0035] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0036] This invention provides a method for optimizing the operation of a distribution network to enhance its carrying capacity through integrated photovoltaic, energy storage, and charging control. The core of this method involves collecting distribution network operation data and meteorological forecast data, establishing a probabilistic power flow model, calculating voltage and thermal carrying capacity margins, and determining dynamic carrying capacity thresholds through Monte Carlo simulation. A hierarchical collaborative control architecture, including station and equipment layers, is constructed. During the day-ahead and intraday phases, rolling optimization is performed with the goal of maximizing the expected carrying capacity margin, and preventative control is triggered when the margin falls below the threshold. In real-time operation, when the margin falls below the safety threshold, emergency collaborative control is executed in a three-tiered sequence: active power response of charging piles, dynamic virtual capacity expansion of energy storage, and reactive / active power reduction of photovoltaics. The regulated state variables are fed back to the probabilistic power flow model, the thresholds are updated, and the data is re-input into the hierarchical architecture and optimization model, forming a closed-loop self-optimizing operation. This achieves dynamic quantitative assessment and proactive improvement of the distribution network's carrying capacity, enhancing the absorption capacity and operational safety of the photovoltaic, energy storage, and charging system.

[0037] Example 1: To better understand the above technical solution, the following will provide a detailed description of the technical solution in conjunction with the accompanying drawings and specific implementation methods. (Refer to...) Figure 1 As shown in the figure, this is a schematic diagram of the method for improving and optimizing the carrying capacity of the distribution network through integrated photovoltaic, energy storage, and charging regulation. The method for improving and optimizing the carrying capacity of the distribution network through integrated photovoltaic, energy storage, and charging regulation includes the following steps:

[0038] Step S1: Dynamic Bearing Capacity Assessment: Collect distribution network operation data and meteorological forecast data. Specifically: Distribution network operation data includes, but is not limited to, voltage amplitude at each node. Current amplitude of each branch Transformer load rate State of charge of energy storage system Real-time power of charging piles Photovoltaic inverter active power output and no effort put in Meteorological forecast data includes solar irradiance and ambient temperature for the next 4–24 hours, with a time resolution of 15 minutes, sourced from numerical weather prediction or local short-term forecast models. Data update frequency: operational data is acquired at the second level, and meteorological forecast data is updated every 15 minutes. A probabilistic power flow model is established, using the three-point estimation method or Latin hypercube sampling to model the uncertainties of photovoltaic output and charging load. Photovoltaic output follows a Beta distribution, and charging load follows a normal distribution or a nonparametric kernel density distribution based on historical data. Deterministic power flow calculations are performed for each sampling scenario to obtain the probability distributions of voltage and branch current at each node. Voltage bearing capacity margin and thermal bearing capacity margin are calculated. Specifically, the voltage bearing capacity margin... : Reflects the degree to which the node voltage approaches its upper limit, The larger the value, the further the voltage is from the upper limit, and the more abundant the carrying capacity; when When =0, the node voltage has reached its upper limit; when A value less than 0 indicates that the voltage is exceeding the limit; thermal tolerance margin. : Reflects the degree to which the branch current approaches the rated value. The larger the value, the further the current is from the rated value, and the more abundant the carrying capacity; when When =0, the branch current has reached the rated value; when A value less than 0 indicates overload; the overall bearing capacity margin is the minimum value among all nodes and branches, i.e. Represents the current bottleneck capacity of the entire distribution network; voltage bearing margin and thermal load margin Calculated using the following formulas respectively: ;in, For nodes Current voltage; branch road Current current; , , These are the upper voltage limit, lower voltage limit, and rated current, respectively. A set of nodes; The set of branch roads is defined; and the dynamic bearing capacity threshold is determined through Monte Carlo simulation. The Monte Carlo simulation generates bearing capacity margin probability distributions under various typical scenarios, and the 5th percentile is taken as the dynamic bearing capacity threshold. ,like Figure 2 As shown.

[0039] Step S1 is the foundational layer for perception and quantification in the entire method. Its core function lies in: a transition from static to dynamic: traditional methods use fixed safety thresholds, while step S1, through probabilistic power flow and Monte Carlo simulations, incorporates photovoltaic power output fluctuations and the randomness of charging loads, outputting a dynamic threshold that changes over time and across different scenarios, enabling subsequent control strategies to adapt to uncertainty; and providing a quantitative benchmark for subsequent steps: the dynamic carrying capacity threshold output by step S1. and real-time load capacity margin This is the trigger judgment basis for the preventive control in step S3, and also the execution target of the three-level step control in step S4; the quantitative target for improving the bearing capacity is defined: the traditional method uses the constraint of voltage not exceeding the limit and current not exceeding the limit, while step S1 of the present invention defines the bearing capacity as a continuous margin index, making the improvement of the bearing capacity an optimizable and traceable continuous variable target.

[0040] Step S2: Construct a layered architecture: Establish a collaborative control architecture that includes a site domain layer and a device layer, such as... Figure 3 As shown, the station layer issues control commands to the equipment layer based on the dynamic carrying capacity threshold obtained in step S1. Specifically, the equipment layer, at the bottom layer, consists of various physical devices within the integrated photovoltaic-storage-charging station, including: photovoltaic inverters: capable of continuous active power adjustment, 0-100% rated power and reactive power compensation, with an adjustable power factor range of 0.8 leading to 0.8 lagging; energy storage converters: supporting four-quadrant operation, capable of independently controlling active and reactive power, equipped with a virtual synchronous machine (VSG) operation mode, and a response time ≤100ms; charging pile clusters: each charging pile has continuously adjustable power capability, ranging from 10% to 100% of rated power, supports remote power limiting commands, and has a response time ≤1s; and intelligent terminal units. Each device is equipped with an intelligent terminal to collect operational status data and receive instructions from the upper layer. The station domain layer acts as the central decision-making unit, integrating a dynamic carrying capacity calculation module and a multi-objective optimization algorithm. It analyzes the operational data uploaded by the device layer in real time, combines grid dispatch instructions with local load forecast results, and generates a coordinated control strategy that balances economy, safety, and response speed. This refined power command is then distributed to each intelligent terminal via a high-speed communication network to ensure the continuous and stable operation of the photovoltaic-storage-charging system under complex operating conditions. Specifically, to coordinate the coordinated control among multiple station domain layers, each station domain layer edge computing terminal is pre-configured with a carrying capacity contribution coefficient. This coefficient reflects the sensitivity of the site to the overall carrying capacity bottleneck of the distribution network, and is determined by the following factors: the nodal voltage sensitivity of the site's access point relative to the root node; the line thermal stability margin of the feeder where the site is located; and the ratio of the site's energy storage capacity to the charging pile power. The larger the value, the greater the adjustment responsibility that the station's regional layer bears in increasing carrying capacity. In the three-level tiered control of step S4, priority scheduling is given to... Larger station domain layers participate in regulation. The hierarchical architecture operates as follows: the station domain layer operates based on the dynamic carrying capacity threshold output in step S1. and the real-time carrying capacity margin of this site Automatically switches between the following three operating modes: Normal mode: Triggering conditions Control behavior: The station domain layer only performs data acquisition and uploading; the device layer operates freely according to the daily plan in step S3; Early warning mode: Triggering conditions Control behavior: The station layer activates edge computing and initiates preventative regulation, such as limiting the upper limit of charging power, but does not execute emergency control for the time being; Emergency mode: Triggering conditions The station domain layer immediately executes the three-level tiered emergency collaborative control in step S4 and actively sends collaborative requests to adjacent station domain layers.

[0041] Step S2 is the core architecture layer of this invention, characterized by hierarchical collaboration. Its core function is to decouple global optimization from local control: In traditional methods, the distribution network master station needs to directly control every device, resulting in heavy communication burden and large response delay. Step S2 introduces edge computing terminals at the station domain layer, decomposing the global optimization problem into multiple local sub-problems. Each station domain layer independently calculates its local adjustment, and the master station only needs to coordinate the coupling constraints between station domain layers, such as total power balance, significantly reducing communication bandwidth requirements and control complexity. Step S2 retains the global optimization capability of Step S3, with day-ahead and intraday plans uniformly formulated by the master station, while also implementing Step S4. Local rapid response and real-time emergency control are independently decided by the station domain layer, balancing global optimization and local real-time performance. While carrying capacity margin is a network-wide indicator, bottlenecks often concentrate at a few weak nodes. Step S2 decomposes the network-wide carrying capacity indicator to each station domain layer by defining a carrying capacity contribution coefficient, clarifying the responsibility ratio of each station in carrying capacity improvement, and enabling control commands to be precisely applied to weak areas. When adding a new optical storage and charging station, only an edge computing terminal needs to be configured and added to the communication network; no modification to the main station's control logic is required. When the communication of the edge computing terminal at a station domain layer is interrupted, that station domain layer can rely on locally stored data... It operates independently from historical data, maintaining basic carrying capacity and support capabilities.

[0042] Step S3: Long-term rolling optimization: In the day-ahead phase, optimize the energy storage charging and discharging plan and the charging pile power limit with the goal of maximizing the overall carrying capacity margin. Specifically, the objective function in the day-ahead phase is:

[0043]

[0044] in, Optimize the total number of time periods for the current day; and Time periods Voltage bearing capacity margin and thermal bearing capacity margin; These are the weighting coefficients; For system operating costs; This is a cost penalty coefficient; during the intraday phase, a rolling adjustment plan is implemented based on ultra-short-term forecasts, and preventative control is triggered when the current bearing capacity margin is lower than the dynamic bearing capacity threshold. Specifically, the triggering conditions for preventative control are: the current bearing capacity margin is lower than the dynamic bearing capacity threshold, and ultra-short-term forecasts indicate that the bearing capacity margin will continue to decline over the next two control periods; the triggering conditions are further refined as follows:

[0045]

[0046] in, and These are the predicted carrying capacity margin values ​​for the first and second control cycles (15 minutes / cycle) based on ultra-short-term forecasts.

[0047] Specific actions for preventative regulation: Energy storage: Advance the originally planned discharge period or increase the depth of discharge in the current period to release energy storage capacity in advance to cope with the upcoming capacity shortage; Charging piles: Reduce the upper limit of charging power in advance, such as from 100% to 70%, and notify the user's app to push the power limit warning to reduce user anxiety; Photovoltaics: If the forecast shows a risk of voltage exceeding the upper limit, adjust the power factor of the photovoltaic inverter from 1 to 0.98 capacitive in advance to reserve reactive power adjustment space; Regulation range limit: The adjustment range of preventative regulation shall not exceed 30% of the original planned value to avoid excessive regulation leading to excessive rebound of capacity in subsequent periods or depletion of energy storage SOC.

[0048] Step S3 is the core of the decision-making and optimization process of this invention. Its core function is to achieve a global forward-looking plan for improving carrying capacity. Traditional photovoltaic-storage-charging control methods often adopt setpoint modes, such as fixed energy storage charging and discharging periods or greedy modes, such as determining charging and discharging based solely on the current SOC. These methods lack foresight regarding future photovoltaic output and load fluctuations. Step S3, through two-stage rolling optimization at the day-ahead and intraday levels, incorporates the uncertainties of the next 24 hours or even 4 hours into the decision-making model. Its direct goal is to maximize the expected carrying capacity margin, reserving sufficient safety buffers for the distribution network from a time perspective, and avoiding a passive situation of remediation after a fault. A multi-objective trade-off mechanism between cost and carrying capacity: Increased carrying capacity often comes with increased costs. For example, early discharge of energy storage increases cycle losses, and charging pile limiting reduces electricity sales revenue. The objective function of step S3 includes both a carrying capacity gain term and a cost penalty term. By adjusting the weighting coefficients, flexible configuration based on operator preferences is achieved. The weighting coefficients are increased in scenarios prioritizing economic efficiency and decreased in scenarios prioritizing safety. This multi-objective trade-off capability is a significant innovation of this invention, distinguishing it from purely safety-constrained methods. Achieving a leap from passive emergency response to proactive prevention: The preventative control module in step S3 uses ultra-short-term forecasting to predict the next two... The cyclical load-bearing capacity trend, acting only when the load-bearing capacity just enters the warning zone rather than waiting until it exceeds the limit, significantly reduces the frequency and intensity of emergency regulation through this feedforward and feedback control structure. Actual tests show that introducing preventative regulation reduces the trigger rate of the three-level emergency control in step S4, decreases the number of equipment actions, and effectively extends the lifespan of energy storage and charging piles. It also provides a precise starting point for regulation in step S4: the regulation amplitude of the three-level stepped control in emergency mode is not without a baseline. The energy storage charging and discharging plan and the charging pile power limit given by step S3 through intraday rolling optimization constitute the baseline for regulation in step S4. When emergency control requires derating charging piles, the adjustment is made downwards from the upper limit set in step S3. When energy storage needs to operate in virtual synchronous machine mode, the charging and discharging plan in step S3 is used as the initial operating point. This design ensures the consistency between emergency control and optimization plan, avoiding command conflicts. A closed-loop evolution mechanism is formed with steps S1 and S5: the optimization model parameters in step S3, such as weight coefficients and cost coefficients, are not fixed, but are periodically fine-tuned based on the historical control effects fed back from step S5. This adaptive parameter adjustment enables the optimization model to continuously evolve with changes in operating conditions, enhancing the long-term adaptability of the method.

[0049] • Step S4: Short-timescale emergency coordinated control: During real-time operation, when the load-bearing capacity margin is lower than the safety threshold, the charging pile power actively responds. Specifically, the derating range of the charging pile power active response is determined by the following formula: ;in: The current charging power of the charging station; The rated power of the charging pile; This is the control coefficient; This represents the thermal carrying capacity margin for the current period; the dynamic virtual capacity expansion of energy storage, specifically, the power regulation law for dynamic virtual capacity expansion of energy storage is: ;in, This is the active power command value for the energy storage converter; This is the proportional control coefficient; For target bearing capacity margin; This represents the current real-time bearing capacity margin; when < When, the energy storage absorbs power during charging; when > Furthermore, when there is a risk of voltage exceeding the upper limit, energy storage discharge provides support; Regarding the reactive power of the photovoltaic inverter, specifically, the specific execution sequence for reactive and active power reduction of the photovoltaic inverter is as follows: First, adjust the power factor of the photovoltaic inverter from 1 to 0.95 within the capacitive range. If the load capacity margin still has not recovered, further adjust the power factor to 0.9. Finally, engage automatic generation control to reduce active power; this three-tiered coordinated control is executed sequentially until the load capacity margin recovers to above the safety threshold. Figure 4 As shown, the specific linkage and exit mechanism of the three-tier system is as follows: Linkage logic: After each level is executed, a time window is waited: charging pile: 2 seconds; energy storage: 200ms; photovoltaic: 3 seconds, then recalculated. If it is still lower If it does not pass, it will automatically proceed to the next level; Exit condition: when ≥ When the power factor reaches +0.02, the power factor should be restored in reverse order: first, restore the photovoltaic active power, then restore the power factor to 1, then deactivate the virtual capacity expansion of energy storage, and finally restore the charging pile power. The deactivation speed should be slower than the intervention speed to avoid oscillation; safety protection: if any device is continuously adjusted more than 10 times / hour, it will be locked in its current state and an alarm will be triggered to prevent the device from overheating or being damaged.

[0050] Coordination with other steps: Input from step S1 and Step S2: Station-level operation mode, emergency mode, and carrying capacity contribution coefficient. The real-time status after the preventive control in step S3 fails; the control action record output to step S5 is used to provide feedback for updating the threshold and optimizing parameters.

[0051] Step S4 is the rapid response and fallback execution layer of this invention. Its core functions can be summarized in the following three points: Fallback safety: When the preventive control in step S3 is insufficient to prevent the carrying capacity from exceeding the limit, step S4 intervenes at a speed of milliseconds to seconds, forcibly pulling the carrying capacity margin back to the safe range through three-level tiered control, serving as the last line of defense for the safe operation of the distribution network; Flexible control prioritizing economy: Execution is carried out in the order of priority: charging pile devaluation, low-cost, virtual capacity expansion of energy storage, medium-cost, photovoltaic reduction, and high-cost, minimizing curtailment losses and energy storage lifespan losses while ensuring safety, reflecting the optimal trade-off between economy and safety; Execution hub in the closed loop: Receiving the real-time margin from step S1, the mode command from step S2, and the baseline plan from step S3, and outputting the actual action quantity to step S5 after execution, enabling the entire closed loop of assessment, decision-making, control, and feedback to be truly implemented, rather than remaining at the simulation level.

[0052] Step S5: Closed-loop feedback: The actual distribution network state variables after adjustment in step S4 are fed back to the probabilistic power flow model in step S1 to update the dynamic carrying capacity threshold, and then re-input to steps S2 and S3 to form a closed-loop self-optimizing operation of evaluation, decision-making, control, and feedback. Specifically, the actual distribution network state variables include at least: the real-time voltage amplitude of each node, the real-time current amplitude of each branch, the actual charging and discharging power of the energy storage converter, the actual charging power of the charging pile group, and the actual active and reactive power output of the photovoltaic inverter; the cycle of the closed-loop self-optimizing operation is 5-15 minutes; within each cycle, steps S1 to S5 are executed sequentially once, and the intraday rolling correction in step S3 is consistent with that in step S4. The real-time emergency coordinated control in step 4 is based on parallel calculation of the latest feedback state variables. The time scale for intraday rolling correction is 15 minutes, and the time scale for real-time emergency coordinated control is from seconds to minutes. Specifically: the dynamic bearing capacity threshold update mechanism: update frequency: executed every 5 to 15 minutes, with a default of 15 minutes, synchronized with the intraday rolling correction cycle in step S3; update algorithm: the actual voltage and current feedback from the previous cycle are used as new samples and added to the historical dataset of step S1; Monte Carlo simulation is rerun, and the number of scenarios can be reduced to 500 to speed up the calculation, updating the probability distribution of bearing capacity margin; the 5th percentile of the new distribution is taken as the dynamic bearing capacity threshold for the next cycle. Adaptive adjustment: if the actual margin is lower than the threshold for three consecutive cycles, it indicates that the threshold is too optimistic, and the quantile is automatically increased from 5% to 10% to make the threshold more conservative; conversely, if emergency control is never triggered, the quantile is gradually decreased to 3% to reduce unnecessary preventive regulation. Closed-loop operation timing coordination: Cycle duration: 15 minutes by default. Within each cycle, steps S1 to S5 are executed sequentially once, forming a complete closed-loop iteration; Parallel calculation mechanism: The intraday rolling correction of step S3 is performed every 15 minutes, in conjunction with the real-time emergency coordinated control of step S4. These actions are triggered at the second level and run independently in parallel without blocking each other; Step S3 reads the status feedback from the previous cycle as the initial value and performs rolling optimization for the next 4 hours; Step S4 monitors the current carrying capacity margin in real time and independently triggers emergency control, unaffected by the calculation time of step S3; Data sharing: After each action, step S4 immediately writes the action amount, derating rate, energy storage power, etc., into shared memory for step S3 to read in the next rolling cycle; After each optimization, step S3 writes the new plan into shared memory for reference by steps S2 and S4.

[0053] Step S5 is the closed-loop adhesive and continuous evolution engine of this invention. Its core functions can be summarized in the following three points: Forming a complete closed loop of evaluation, decision-making, control, and feedback: Without step S5, steps S1 to S4 are merely one-time open-loop control, unable to adapt to changes in operating conditions. Step S5 feeds back the actual regulated state variables to the probabilistic power flow model of step S1, enabling the next evaluation to correct the model based on real data, achieving increasingly accurate positive evolution over time; Coordinating parallel operation across multiple time scales: The 15-minute rolling optimization of step S3 and the second-level emergency control of step S4 naturally have different time scales. Step S5 ensures that the two do not conflict, block, or lose data through data alignment, shared memory, and parallel scheduling mechanisms, allowing long-term optimization and short-term control to coexist harmoniously; Driving online adaptive model parameters: By dynamically adjusting the threshold selection quantile of step S1, the optimization weight coefficient of step S3, and the regulation coefficient of step S4 based on the actual feedback deviation, the entire method possesses self-learning and environmental adaptability, requiring no frequent manual intervention.

[0054] The contents not described in detail in this specification are existing technologies known to those skilled in the art.

[0055] Although the present invention has been described in detail with reference to the foregoing embodiments, those skilled in the art can still modify the technical solutions described in the foregoing embodiments or make equivalent substitutions for some of the technical features. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of the present invention should be included within the protection scope of the present invention.

Claims

1. A method for improving and optimizing the carrying capacity of distribution networks through integrated photovoltaic, energy storage, and charging regulation, characterized in that: Includes the following steps: Step S1: Dynamic bearing capacity assessment: Collect distribution network operation data and meteorological forecast data, establish a probabilistic power flow model, calculate voltage bearing capacity margin and thermal bearing capacity margin, and determine the dynamic bearing capacity threshold through Monte Carlo simulation; Step S2: Construct a layered architecture: Establish a collaborative control architecture that includes a station domain layer and an equipment layer. The station domain layer issues control commands to the equipment layer based on the dynamic carrying capacity threshold obtained in step S1. Step S3: Long-term rolling optimization: In the day-ahead phase, optimize the energy storage charging and discharging plan and the charging pile power limit with the goal of maximizing the overall carrying capacity margin expectation. In the intraday phase, roll the plan based on ultra-short-term forecasts and trigger preventive regulation when the current carrying capacity margin is lower than the dynamic carrying capacity threshold. Step S4: Short-time-scale emergency coordinated control: In real-time operation, when the load capacity margin is lower than the safety threshold, coordinated control is executed in a three-tiered order: active power response of charging piles, dynamic virtual capacity expansion of energy storage, and reactive power of photovoltaic inverters, until the load capacity margin recovers to above the safety threshold. Step S5: Closed-loop feedback: Feedback the actual distribution network state variables after adjustment in step S4 to the probabilistic power flow model in step S1, update the dynamic carrying capacity threshold, and re-input it to steps S2 and S3 to form a closed-loop self-optimizing operation of evaluation, decision-making, control, and feedback.

2. The method for optimizing the operation of distribution network carrying capacity through integrated photovoltaic, energy storage, and charging control as described in claim 1, characterized in that: Voltage bearing margin in step S1 and thermal load margin Calculated using the following formulas respectively: in, For nodes Current voltage; branch road Current current; , , These are the upper voltage limit, lower voltage limit, and rated current, respectively. A set of nodes; For the set of branches.

3. The method for optimizing the operation of distribution network carrying capacity through integrated photovoltaic, energy storage, and charging control as described in claim 1, characterized in that: In step S1, the bearing capacity margin probability distribution under various typical scenarios is generated by Monte Carlo simulation, and the 5th percentile is taken as the dynamic bearing capacity threshold.

4. The method for optimizing the operation of distribution network carrying capacity through integrated photovoltaic, energy storage, and charging control as described in claim 1, characterized in that: The objective function for the day-ahead stage in step S3 is: in, Optimize the total number of time periods for the current day; and Time periods Voltage carrying capacity margin and thermal carrying capacity margin; These are the weighting coefficients; For system operating costs; This represents the cost penalty coefficient.

5. The method for optimizing the operation of distribution network carrying capacity through integrated photovoltaic, energy storage, and charging control as described in claim 1, characterized in that: The triggering condition for preventive regulation in step S3 is: the current bearing capacity margin is lower than the dynamic bearing capacity threshold, and the ultra-short-term forecast shows that the bearing capacity margin will continue to decline in the next two control cycles.

6. The method for optimizing the operation of distribution network carrying capacity through integrated photovoltaic, energy storage, and charging coordination and control as described in claim 1, characterized in that: The derating rate of the charging pile's active power response in step S4 is determined by the following formula: in: The current charging power of the charging station; The rated power of the charging pile; This is the control coefficient; This represents the thermal load-bearing capacity margin for the current period.

7. The method for optimizing the operation of distribution network carrying capacity through integrated photovoltaic, energy storage, and charging control as described in claim 1, characterized in that: The power regulation law for the dynamic virtual capacity expansion of energy storage in step S4 is as follows: ;in, This is the active power command value for the energy storage converter; This is the proportional control coefficient; For target bearing capacity margin; This represents the current real-time bearing capacity margin; when < When, the energy storage absorbs power during charging; when > Furthermore, when there is a risk of voltage exceeding the upper limit, energy storage discharge provides support.

8. The method for optimizing the operation of distribution network carrying capacity through integrated photovoltaic, energy storage, and charging control as described in claim 1, characterized in that: The specific execution sequence of reactive and active power reduction of the photovoltaic inverter in step S4 is as follows: First, the power factor of the photovoltaic inverter is adjusted from 1 to 0.95 within the capacitive range. If the load-bearing capacity margin has not been restored, the power factor is further adjusted to 0.

9. Finally, automatic power generation control is put into operation to reduce active power.

9. The method for optimizing the operation of distribution network carrying capacity through integrated photovoltaic, energy storage, and charging control as described in claim 1, characterized in that: The actual power distribution network status quantities fed back to step S1 in step S5 include at least: the real-time voltage amplitude of each node, the real-time current amplitude of each branch, the actual charging and discharging power of the energy storage converter, the actual charging power of the charging pile group, and the actual active and reactive power output of the photovoltaic inverter.

10. The method for optimizing the operation of distribution network carrying capacity through integrated photovoltaic, energy storage, and charging control as described in claim 1, characterized in that: The closed-loop self-optimization operation cycle is 5 to 15 minutes. In each cycle, steps S1 to S5 are executed sequentially once. The intraday rolling correction in step S3 and the real-time emergency collaborative control in step S4 are both calculated in parallel based on the latest feedback state variables. The time scale of the intraday rolling correction is 15 minutes, and the time scale of the real-time emergency collaborative control is from the second to the minute level.

Images