A power voltage collaborative support control system for a light storage cluster

Abstract

This invention relates to the field of photovoltaic and energy storage equipment control technology, and particularly to a power and voltage coordinated support control system for photovoltaic and energy storage clusters. The system includes: collecting multi-source heterogeneous data from distribution network branches to obtain a hierarchical standardized electrical dataset; constructing a branch-coupled electrical characteristic model to obtain disturbance location results; quantifying the multi-dimensional aging characteristics of energy storage equipment, and combining medium- and long-term output predictions to divide the dual-layer regulation capacity constraint range for each distribution network branch, outputting the branch regulation resource boundary; using the branch regulation resource boundary as a constraint to solve for the initial branch control commands, eliminating commands with the risk of exceeding limits, and outputting compliant coordinated control commands to simultaneously execute DC damping regulation and AC power regulation, obtaining a control operation dataset; obtaining the control deviation, and using the control deviation to simultaneously correct the parameters of the branch-coupled electrical characteristic model and the dual-layer regulation capacity constraint range. This scheme improves the voltage regulation accuracy of multi-branch photovoltaic and energy storage distribution networks through an adaptive iterative mechanism.

Description

Technical Field

[0001] This invention relates to the field of photovoltaic and energy storage equipment control technology, and in particular to a power and voltage coordinated support control system for photovoltaic and energy storage clusters. Background Technology

[0002] With the large-scale integration of distributed photovoltaic (PV) and energy storage devices into branch-type distribution networks, the penetration rate of PV-storage clusters in distribution networks continues to increase. This leads to significant differences in branch operating conditions, frequent power fluctuations, and prominent AC / DC coupling disturbances in grid operation. Existing distribution network voltage regulation methods often employ unified modeling and fixed capacity constraints across the entire network, failing to fully consider the impedance characteristics, load fluctuations, and individual differences in PV and energy storage output among different branches. This easily results in inaccurate voltage regulation in local branches and poor overall network compatibility.

[0003] Meanwhile, existing control strategies generally neglect the constraint relationship between the aging and degradation of energy storage cells and the voltage regulation capability. They fail to dynamically adjust the voltage regulation boundary based on the number of energy storage cycles, internal resistance increments, and capacity degradation status. Under long-term operation, this can easily lead to problems such as overload control of aging energy storage, increased losses, and accelerated lifespan degradation. In addition, traditional control focuses on single voltage correction on the AC side, lacking the coordinated cooperation between DC damping suppression and AC power regulation. Under extreme conditions such as sudden cloud cover changes and load fluctuations, this can easily trigger secondary risks such as AC / DC coupling oscillations, short-term voltage over-limits, and power backfeed.

[0004] Furthermore, the control mode with fixed model parameters and fixed voltage regulation capacity range cannot adapt to the dynamic operating conditions of the distribution network and the aging changes of energy storage throughout its entire life cycle. Its adaptive capability is weak, making it difficult to meet the safe, economical, and stable coordinated control operation requirements of high-penetration photovoltaic-storage distribution networks. Summary of the Invention

[0005] This invention constructs a closed-loop adaptive iterative mechanism for branch-level AC / DC coupling coordinated regulation and energy storage aging constraints, which solves the technical problem of poor adaptability of traditional fixed modeling and regulation of the entire distribution network, and significantly improves the voltage regulation accuracy and operating condition adaptive regulation capability of multi-branch photovoltaic-storage distribution networks.

[0006] The technical solution proposed in this invention is: a power-voltage coordinated support control system for photovoltaic-storage clusters, comprising: The data acquisition module is used to collect multi-source heterogeneous data from distribution network branches, unify the time series reference, match the time-aligned multi-source heterogeneous data to the corresponding distribution network branches, and obtain a hierarchical standardized electrical dataset. The disturbance tracing module constructs a branch coupled electrical feature model based on a hierarchical standardized electrical dataset, extracts the branch time-series residuals, and reverse-matches the distribution network branch to which the disturbance belongs to obtain the disturbance location result. The capacity constraint planning module quantifies the multi-dimensional aging characteristics of energy storage equipment based on the disturbance location results, and divides the dual-layer regulation capacity constraint interval of each distribution network branch in combination with the medium and long-term output forecast, and outputs the branch regulation resource boundary. The simulation pre-verification module uses the branch adjustment resource boundary as a constraint to solve the initial control instructions of the branch. It combines the branch simulation model to perform working condition simulation and verification, eliminates instructions with the risk of exceeding the limit, and outputs compliant collaborative control instructions. The AC / DC coordinated control module synchronously executes DC damping adjustment and AC power adjustment based on compliant coordinated control commands, and obtains a control operation dataset by combining the actual operating parameters of the equipment. The adaptive iterative module calculates the difference between the control operation dataset and the control target value to obtain the control deviation. It then uses the control deviation to synchronously correct the parameters of the branch coupled electrical characteristic model and the two-layer control capacity constraint range.

[0007] Preferably, the process for obtaining the hierarchical standardized electrical dataset is as follows: Collect AC side voltage, branch power, and load electrical data of each branch of the distribution network, and simultaneously collect multi-source heterogeneous data of DC side dynamic impedance, ripple current, and cell status of photovoltaic and energy storage PCS. By unifying the global timing reference based on the acquisition frequency of multi-source heterogeneous data, the sampling timing deviation of different devices is calibrated to achieve synchronization and alignment of the timing of data across all branches. Based on the physical topology of the distribution network, a unique index is established for each branch. The time-aligned multi-source heterogeneous data is mapped to the corresponding topology branch. The data is collected and organized according to the branch index to form a hierarchical standardized electrical dataset at the branch level.

[0008] Preferably, the process for obtaining the disturbance localization result is as follows: Read the hierarchical standardized electrical dataset branch by branch and extract the AC / DC electrical characteristics, power fluctuation characteristics, and voltage offset characteristics of each branch. Based on the independent electrical parameters of each branch, a branch coupled electrical characteristic model is constructed, and the real-time electrical operation sequence residual of each branch is calculated. By comparing the timing residuals of each branch with the preset disturbance threshold, the target distribution network branch corresponding to the voltage disturbance and AC / DC coupled oscillation is matched in reverse to locate the disturbance and obtain the branch disturbance location result.

[0009] Preferably, the process for obtaining the branch adjustment resource boundary is as follows: Based on the branch disturbance location results, the incremental internal resistance of the energy storage cell, the cumulative number of cycles and the remaining capacity decay coefficient are collected for each branch to obtain the energy storage aging characteristic parameters of each branch. Historical photovoltaic power output time series data and historical load fluctuation time series data of each branch are collected, and medium- and long-term prediction data of branch photovoltaic power output and load fluctuation are generated through time series extrapolation. Establish a closed mapping relationship between the aging characteristic parameters of energy storage and the upper limit of the voltage regulation output of the branch, limit the regulation output threshold of the high-aging energy storage branch, and combine the medium and long-term forecast data of the branch photovoltaic output and load fluctuation to define the steady-state voltage regulation capacity range and the dynamic emergency voltage regulation capacity range of the branch, forming a nested double-layer capacity constraint range, and output the branch regulation resource boundary.

[0010] Preferably, the solution process for the preliminary control command is as follows: Based on the branch regulation resource boundary, the double-layer capacity constraint interval is used as a hard constraint condition, and the regulation objective function is constructed by combining the branch real-time voltage deviation and power fluctuation. Solve the control objective function to obtain the initial power and voltage regulation commands that are adapted to the single-branch disturbance condition and conform to the upper limit of the branch capacity, which serve as the initial control commands for the branch. The preliminary control commands from each branch are aggregated to form a set of preliminary control commands for the entire cluster.

[0011] Preferably, the process of obtaining the compliance-coordinated control instruction is as follows: The preset branch simulation model is invoked to synchronously replicate the real-time topology, optical storage parameters, and line impedance operation status of each branch. The initial control commands for each branch are input into the corresponding branch simulation model to perform branch operating condition simulation and complete multi-dimensional risk verification of power backfeed, voltage reverse over-limit, and energy storage overload. The initial control instructions that pose secondary risks to the power grid are removed, and the remaining instructions are adjusted and adapted across the entire domain to output compliant and coordinated control instructions without operational risks.

[0012] Preferably, the process of obtaining the control operation dataset is as follows: The compliant and coordinated control command is transmitted to the corresponding branch photovoltaic inverter and energy storage PCS equipment, and the branch DC damping regulation and AC power voltage regulation coordinated action are triggered simultaneously. The DC side dynamically adjusts the PCS virtual damping parameters to suppress AC / DC coupling oscillations in the branch, while the AC side matches the reactive and active power outputs to complete the bus voltage correction. After real-time collection and control, the AC / DC electrical parameters, energy storage operation parameters, and bus voltage parameters of each branch are collected and organized to form a control operation dataset.

[0013] Preferably, the process for obtaining the control deviation is as follows: The preset voltage stability target value, power smoothing target value, and energy storage loss control target value of each branch are retrieved to form a set of control benchmark targets; By performing a difference calculation on the actual operating parameters of the control operation dataset and the corresponding control benchmark target set in each dimension, the voltage deviation, power deviation and energy storage operation deviation are obtained respectively. By integrating voltage deviation, power deviation, and energy storage operation deviation, and normalizing them, the real-time control deviation of each branch is obtained.

[0014] Preferably, the correction process for the branch coupling electrical characteristic model parameters and the two-layer regulating capacity constraint range is as follows: The control deviation is used as a correction factor to iteratively update the coefficient parameters of the branch coupled electrical characteristic model parameter by parameter; Based on the energy storage operation loss and operating condition changes corresponding to the real-time deviation, the upper and lower limit thresholds of the dual-layer regulation capacity constraint range of each branch are dynamically fine-tuned. After correcting the parameters of the branch coupled electrical characteristic model and the dual-layer regulating capacity constraint range, the updated data will be sent back to the corresponding functional module.

[0015] The present invention also provides a computer-readable storage medium storing a computer program that is executed by a processor to implement the aforementioned power voltage coordinated support control system for photovoltaic energy storage clusters.

[0016] The beneficial effects of this invention are: 1. The zoned control mode, which uses independent modeling and management of branch lines, can accurately preserve the impedance characteristics, load fluctuations, and differentiated power output of each distribution network branch. This effectively solves the problems of loss of branch features, fuzzy disturbance location, strong generality of control commands, and insufficient specificity in traditional full-domain modeling. It enables precise source tracing of faults and disturbances, and refined identification of branch operating conditions, significantly improving the accuracy of voltage regulation and power optimization in multi-branch distribution networks, and adapting to the differentiated operating conditions of multiple branches.

[0017] 2. By using a reverse constraint voltage regulation capacity mechanism based on the aging state of energy storage, the degree of energy storage aging is quantified from multiple dimensions, including cell internal resistance, cycle count, and remaining capacity. The upper limit of branch voltage regulation is dynamically locked based on the aging coefficient. Simultaneously, a double-layer nested voltage regulation capacity range of steady state and dynamic is constructed to adapt to different operating conditions. This effectively solves the problems of traditional fixed voltage regulation capacity thresholds not adapting to the aging and degradation characteristics of energy storage throughout its entire life cycle, insufficient voltage regulation margin under extreme operating conditions, and wasted steady-state control redundancy. It avoids overload operation, increased losses, and accelerated lifespan degradation of aging energy storage, while maximizing the exploitation of adjustable photovoltaic and energy storage resources in the branch, balancing the safety of distribution network regulation and the economic efficiency of equipment operation.

[0018] 3. A closed-loop mechanism for AC / DC coordinated damping control and parameter adaptive iteration is established. This mechanism suppresses AC / DC coupling oscillations through DC virtual damping, corrects bus voltage on the AC side, and continuously iterates and optimizes model parameters and capacity constraint boundaries based on control deviations. This effectively addresses the shortcomings of traditional single AC voltage regulation, such as weak anti-disturbance capability, difficulty in suppressing AC / DC coupling oscillations, and the inability of fixed models to adapt to dynamic grid conditions. It can quickly smooth power and voltage fluctuations caused by sudden load changes and sudden changes in sunlight, avoid secondary risks such as power backfeeding and voltage exceeding limits, and continuously improve the adaptive and coordinated control capability of distribution network photovoltaic-storage clusters. Attached Figure Description

[0019] Figure 1 A flowchart of a power-voltage coordinated support control system for photovoltaic-storage clusters; Figure 2 A flowchart illustrating branch-independent modeling of a power-voltage coordinated support control system for photovoltaic-storage clusters; Figure 3 A flowchart of a two-layer capacity configuration for energy storage aging constraints in a power-voltage coordinated support control system for photovoltaic-storage clusters. Figure 4 This is a flowchart of AC / DC coordinated closed-loop iterative control of a power-voltage coordinated support control system for photovoltaic-storage clusters. Detailed Implementation

[0020] The following description is intended to disclose the present invention and enable those skilled in the art to implement it. The preferred embodiments described below are merely examples, and other obvious variations will occur to those skilled in the art. The basic principles of the invention defined in the following description can be applied to other embodiments, modifications, improvements, equivalents, and other technical solutions that do not depart from the spirit and scope of the invention.

[0021] It is understood that the term "a" should be understood as "at least one" or "one or more," that is, in one embodiment, the number of an element can be one, while in another embodiment, the number of the element can be multiple, and the term "a" should not be understood as a limitation on the number.

[0022] like Figure 1 As shown, a power-voltage coordinated support control system for photovoltaic-storage clusters includes: The data acquisition module collects multi-source heterogeneous data from distribution network branches, unifies the time-series benchmark, and matches the time-aligned multi-source heterogeneous data to the corresponding distribution network branches to obtain a hierarchical standardized electrical dataset. The disturbance tracing module constructs a branch-coupled electrical feature model based on the hierarchical standardized electrical dataset, extracts the branch time-series residuals, and reverse-matches the distribution network branch to which the disturbance belongs to obtain the disturbance location result. The capacity constraint planning module quantifies the multi-dimensional aging characteristics of energy storage equipment based on the disturbance location result, combines medium- and long-term output forecasts to divide the dual-layer regulation capacity constraint interval for each distribution network branch, and outputs the branch regulation resource boundary. Simulation and pre-processing modules are also included. The verification module uses the branch regulation resource boundary as a constraint to solve the initial branch regulation command, and performs operating condition simulation verification in conjunction with the branch simulation model. It eliminates commands with the risk of exceeding limits and outputs compliant coordinated regulation commands. The AC / DC coordinated regulation module executes DC damping regulation and AC power regulation synchronously based on the compliant coordinated regulation commands, and obtains the regulation operation dataset by combining the actual operating parameters of the equipment. The adaptive iteration module calculates the regulation deviation by performing a difference calculation between the regulation operation dataset and the regulation target value, and uses the regulation deviation to synchronously correct the parameters of the branch coupled electrical characteristic model and the double-layer regulation capacity constraint range.

[0023] This embodiment is applicable to distributed photovoltaic-storage clusters in multi-branch tree-structured distribution networks. A complete set of multi-source data acquisition hardware modules is deployed at each branch node, photovoltaic grid connection point, and energy storage compartment throughout the entire region. The complete set of multi-source data acquisition hardware modules includes a branch electrical acquisition terminal, a photovoltaic inverter status monitoring module, an energy storage PCS monitoring unit, an energy storage BMS cell status acquisition terminal, a branch load monitoring terminal, and an ambient light acquisition module. Each hardware module independently acquires data, summarizes data at different levels, and communicates with each other.

[0024] The system comprises several components: a branch electrical data acquisition terminal and a branch load monitoring terminal collaboratively collect AC side electrical data from the distribution network branches, including PCC bus three-phase voltage, branch active power, branch reactive power, and branch load current, to reflect the overall voltage and power operating status of the distribution network branches; a photovoltaic inverter status monitoring module collects photovoltaic DC side operating data, including photovoltaic DC bus voltage, DC output current, DC dynamic impedance, and output power time-series data, to reflect the DC side operating condition fluctuations and impedance characteristic changes of the photovoltaic unit; an energy storage PCS monitoring unit and an energy storage BMS cell status acquisition terminal collect energy storage AC and DC operating data in layers, with the PCS monitoring unit collecting PCS AC and DC power and PCS DC side ripple current, and the BMS cell status acquisition terminal specifically collecting cell internal resistance, cell cumulative cycle count, and cell remaining capacity, covering the operating status and aging characteristics of energy storage equipment; and an ambient light acquisition module collects branch operating environment data, including real-time light intensity, ambient temperature, and load time-series fluctuation data, to match the external environmental impact conditions of photovoltaic and energy storage output.

[0025] To balance data acquisition accuracy with the computing load on edge devices, this embodiment configures differentiated fixed sampling periods for various acquisition terminals based on operating parameters with different response speeds and variation characteristics. Specifically, distribution network voltage, branch power, and PCS operating status are instantaneously fluctuating parameters, and are sampled continuously at a high frequency of 10ms / sample to capture details of instantaneous voltage fluctuations and power disturbances in the branches. Photovoltaic DC impedance and energy storage cell status are slowly varying steady-state parameters, and are sampled at a medium frequency of 1s / sample to effectively reduce the computing power overhead of edge terminals while ensuring the accuracy of device status updates. Irradiance and ambient temperature are externally slowly varying environmental parameters, and are sampled at a low frequency of 1min / sample to stably accumulate long-term steady-state operating condition time-series samples. All raw heterogeneous data collected by the terminals are uniformly uploaded to the edge controller in real time via a dedicated communication protocol for the distribution IoT.

[0026] Based on the completion of comprehensive multi-source heterogeneous data collection, this embodiment accurately divides branch units according to the actual physical topology of the distribution network. This scheme uses distribution network branch switches, line branch joints, and transformer nodes as physical dividing nodes to split the entire tree-structured distribution network into several independent topology branch units. Simultaneously, each distribution network branch is uniquely coded and marked, establishing a one-to-one correspondence between the topology branch index and equipment, data, and operating conditions.

[0027] Furthermore, the process of obtaining the hierarchical standardized electrical dataset is as follows: The system collects AC side voltage, branch power, and load condition electrical data for each branch of the distribution network, and simultaneously collects multi-source heterogeneous data on DC side dynamic impedance, ripple current, and cell status of photovoltaic and energy storage PCS. A global timing benchmark is unified based on the collection frequency of the multi-source heterogeneous data, and the sampling timing deviation of different devices is calibrated to achieve time synchronization alignment of data across all branches. A unique branch index is established based on the physical topology of the distribution network, and the time-aligned multi-source heterogeneous data is mapped one by one to the corresponding topology branch. All-dimensional data is collected and organized according to the branch index to form a hierarchical standardized electrical dataset at the branch level.

[0028] like Figure 2As shown, this embodiment uses the local industrial clock of the edge controller as the sole timing reference for the entire system, unifying the overall minimum timing primitive as 10ms. Layered timing calibration and synchronization alignment are performed on the raw data with different sampling periods and timing accuracies. For high-frequency electrical data such as branch voltage, power, and PCS operating status with a 10ms sampling period, the terminal's local clock offset error is directly corrected based on the edge controller's standard clock, and the original data timestamp is calibrated without smoothing or noise reduction. For medium-speed steady-state data such as photovoltaic DC impedance and energy storage cell status with a 1s sampling period, a piecewise linear interpolation completion method is used. Based on the numerical fitting of the effective steady-state sampling points before and after, missing data in the 10ms timing nodes is completed, densifying the medium-speed steady-state data into a timing sequence with the same granularity as the electrical data, achieving precise matching of steady-state parameters and transient electrical parameters. For low-frequency environmental data such as light intensity and ambient temperature with a 1min sampling period, a timing window interval assignment mechanism is used to bind a single environmental sampling data point to all corresponding 10ms timing nodes within one minute.

[0029] The system collects AC electrical data, photovoltaic DC operation data, energy storage AC / DC status data, and branch environmental condition data for each branch. Simultaneously, multi-dimensional data cleaning and standardization are performed. Data is screened batch by batch, removing duplicate time-series data, out-of-limit abnormal data, and invalid null data caused by communication interruptions, momentary equipment anomalies, or transmission distortions. Valid operating condition data that conforms to the physical constraints of equipment operation is retained. After data cleaning, the storage field format, physical parameter units, and numerical calculation precision of all branch data are standardized to eliminate format differences and accuracy deviations in the output data from different acquisition terminals. After a complete process of time-series alignment, branch aggregation, data cleaning, and format standardization, a hierarchical standardized electrical dataset is finally generated, with each distribution network branch as an independent analysis and storage unit. The dataset covers four core dimensions: branch AC operation characteristics, photovoltaic DC impedance output characteristics, energy storage cell aging and operating status characteristics, and branch time-series environmental conditions.

[0030] Furthermore, the process of obtaining the perturbation localization results is as follows: The hierarchical standardized electrical dataset is read branch by branch, and the AC / DC electrical characteristics, power fluctuation characteristics, and voltage offset characteristics of each branch are extracted. Based on the independent electrical parameters of each branch, a branch coupled electrical characteristic model is constructed, and the real-time electrical operation sequence residual of each branch is calculated. The time sequence residual of each branch is compared with the preset disturbance threshold, and the target distribution network branch corresponding to the voltage disturbance and AC / DC coupled oscillation is matched in reverse to locate the disturbance in the branch and obtain the branch disturbance location result.

[0031] This embodiment extracts three types of core branch operation characteristics: first, branch AC / DC electrical characteristics, covering the branch AC bus voltage amplitude, AC voltage phase angle, photovoltaic DC-side dynamic impedance, and real-time steady-state value and dynamic change of energy storage PCS DC ripple current; second, branch power fluctuation characteristics, including branch active power fluctuation rate, reactive power offset, and real-time output fluctuation amplitude of photovoltaic and energy storage; and third, branch voltage offset characteristics, including the deviation amplitude of bus voltage relative to rated voltage, voltage duration of voltage offset, and instantaneous voltage jump characteristics. Based on the above multi-dimensional independent branch characteristics, a branch coupled electrical characteristic model is built to reflect the strong coupling correlation mechanism among the dynamic change of AC voltage, DC impedance drift, and DC ripple disturbance of a single branch, and to characterize the AC / DC linkage anomaly characteristics caused by local disturbances in a single branch.

[0032] Based on the branch-coupled electrical characteristic model, the real-time electrical timing residuals of each branch are calculated window by window. The formula for calculating the real-time electrical timing residuals of the branches is as follows: ; In the formula: This represents the electrical timing residual of the current branch at time t. This represents the actual measured voltage value of the PCC bus at time t for the current branch. The theoretical steady-state voltage values ​​for the branch circuit were obtained by fitting the historical steady-state operating conditions over the past 72 hours. These theoretical steady-state voltage values ​​exclude historical instantaneous disturbances and abnormal fluctuations.

[0033] This embodiment, based on the actual operating conditions of multiple branches in the distribution network, sets different steady-state disturbance judgment thresholds for each branch. The specific process for setting the preset disturbance threshold is as follows: using the theoretical steady-state voltage value of the branch as a benchmark, the allowable steady-state voltage residual threshold is set to ±2%. , The rated voltage of the distribution network branch is used; simultaneously, auxiliary thresholds for DC-side disturbances are set, with a DC ripple current distortion rate threshold of 5% and a DC dynamic impedance offset threshold of 8%. Based on these, branch disturbance judgment and source tracing are performed. The specific process is as follows: the real-time electrical timing residual of the branch is compared with the preset steady-state threshold window by window. If the branch timing residual exceeds ±2% for three or more consecutive 10ms timing windows... If the steady-state threshold is met, and the auxiliary judgment conditions of DC ripple current distortion rate exceeding 5% or DC dynamic impedance offset exceeding 8% are also met, it can be determined that there is an abnormal AC / DC coupling disturbance in the current branch. The timing residuals and coupling characteristic parameters of all distribution network branches in the entire area are traversed sequentially, and the deviation of each branch's operating condition is compared one by one. The target branches that meet the disturbance triggering conditions are screened out, and the interference of normal branches without abnormal residuals and DC coupling distortion is excluded. The branch where the disturbance occurs is locked, and the branch-level reverse source tracing of the disturbance source is completed, and the branch disturbance location result is finally determined.

[0034] Furthermore, the process of obtaining the resource boundary for branch adjustment is as follows: Based on the branch disturbance location results, the incremental internal resistance of the energy storage cells, the cumulative number of cycles, and the remaining capacity decay coefficient are collected for each branch to obtain the energy storage aging characteristic parameters of each branch. Historical photovoltaic power output time-series data and historical load fluctuation time-series data of each branch are collected, and medium- and long-term prediction data of branch photovoltaic power output and load fluctuation are generated through time-series extrapolation. A closed mapping relationship between the energy storage aging characteristic parameters and the upper limit of branch voltage regulation output is established to limit the regulation output threshold of highly aging energy storage branches. Combined with the medium- and long-term prediction data of branch photovoltaic power output and load fluctuation, the steady-state voltage regulation capacity range and the dynamic emergency voltage regulation capacity range of the branch are delineated respectively to form a nested double-layer capacity constraint range and output the branch regulation resource boundary.

[0035] like Figure 3 As shown, this embodiment constructs a comprehensive aging quantification index for energy storage by using a reverse constraint voltage regulation capacity mechanism based on the aging state of branch energy storage, thereby mapping the degree of aging to the upper limit of voltage regulation. The comprehensive aging coefficient calculation formula is as follows: ; In the formula: Let be the comprehensive aging coefficient of the energy storage of the i-th branch, with a value of [0,1]. The current internal resistance of the branch cell, , These represent the maximum and minimum internal resistance values ​​of the entire energy storage system. The cumulative number of cycles for the battery cell. To design the maximum number of loops; This represents the current remaining capacity. Rated capacity; , , These are the weighting coefficients, and their sum is 1.

[0036] A voltage regulation output locking mapping formula is constructed based on the aging coefficient: ; In the formula: This represents the maximum allowable voltage regulation output of the i-th branch; The rated adjustable capacity for branch energy storage. The higher the aging coefficient, the lower the upper limit of branch voltage regulation, thus achieving active load limiting protection for aging branches.

[0037] Based on the solution for the voltage regulation upper limit of energy storage aging interlock, this embodiment, for each independent distribution network branch, uses the time-series data of the branch's historical photovoltaic output and load fluctuation over the past 72 hours to perform medium- and long-term operating condition extrapolation of the branch's photovoltaic output and load fluctuation, obtaining stable medium- and long-term prediction data that closely matches the branch's operating patterns. The time-series moving average prediction algorithm in this embodiment is as follows: ; ; In the formula: The long-term forecast output of the branch photovoltaic system at time t; Let t be the medium-to-long-term predicted power of the branch load at time t; , The actual photovoltaic output and branch load power at historical time points are represented by M, which is the total number of sampling points in the 72-hour time series sliding window. By smoothing the time series through a large window, short-term sudden noise is filtered out, and the steady-state change trend of photovoltaic and energy storage output and load is preserved, resulting in a stable and reliable medium- and long-term prediction sequence.

[0038] Based on the aforementioned medium- and long-term steady-state prediction data and branch short-term real-time fluctuation data, this embodiment delineates nested, double-layered voltage regulation capacity constraint intervals to adapt to different operating conditions. These intervals are the steady-state voltage regulation capacity interval and the dynamic emergency voltage regulation capacity interval. The two types of intervals have different constraint dimensions, are nested and complementary, and together constitute the complete adjustable resource constraint range for the branch. Specifically, the steady-state voltage regulation capacity interval adapts to the steady-state operating conditions of small daily voltage deviations and normalized photovoltaic and energy storage output fluctuations in the branch. Using the medium- and long-term photovoltaic and load prediction steady-state values ​​as a benchmark, combined with the allowable range of conventional voltage deviations in the branch, the upper and lower limits of the steady-state adjustable capacity are defined. The calculation formula for the steady-state voltage regulation capacity interval is as follows: ; In the formula: , These are the lower and upper limits of the steady-state voltage regulation capacity range, respectively. , The steady-state voltage regulation margin coefficient is a fixed value determined based on the normal fluctuation characteristics of the distribution network branches. It is used to define the safe and continuous steady-state voltage regulation range of energy storage under daily operating conditions.

[0039] The dynamic emergency voltage regulation capacity range is adapted to extreme dynamic operating conditions such as sudden changes in cloud cover, sudden increases or decreases in load, and short-term disturbances. Based on the extreme values ​​of real-time output fluctuations of branches and the amplitude of short-term load fluctuations, it extends the dynamic emergency adjustable margin outward from the steady-state capacity range to offset voltage over-limits and power oscillations caused by short-term severe operating condition fluctuations. The calculation formula for the dynamic emergency voltage regulation capacity range is as follows: ; In the formula: , These are the lower and upper limits of the dynamic emergency pressure regulation capacity range, respectively. This refers to the peak deviation of short-term load fluctuations in branch circuits. This refers to the peak deviation of short-term photovoltaic power output. This is a dynamic margin adjustment coefficient used to adapt to emergency pressure regulation needs under extreme operating conditions.

[0040] In this embodiment, routine steady-state control employs an inner-layer steady-state range constraint to ensure both control efficiency and equipment stability. Under extreme disturbance conditions, the outer-layer dynamic range margin is activated to enhance the branch's disturbance immunity. Simultaneously, the overall upper and lower limits of the dual-layer capacity range are combined with the maximum voltage regulation output locked by energy storage aging. A secondary threshold clamping correction is performed to deeply couple the hardware constraints of energy storage aging and degradation with the software constraints of operating condition fluctuations, eliminating the theoretically adjustable capacity that exceeds the aging capacity of the equipment.

[0041] Furthermore, the solution process for the preliminary control command is as follows: Based on the branch regulation resource boundary, and with the double-layer capacity constraint interval as a hard constraint condition, a regulation objective function is constructed by combining the branch real-time voltage offset and power fluctuation. The regulation objective function is solved to obtain the initial power and voltage regulation commands that are adapted to the single branch disturbance condition and conform to the branch capacity limit, which serve as the branch preliminary regulation commands. The preliminary regulation commands of each branch are summarized to form a set of cluster-wide preliminary regulation commands.

[0042] This embodiment constructs a multi-objective collaborative regulation and optimization model based on the dual-layer regulation capacity constraint range and the energy storage aging lockout output boundary, combined with the real-time operating deviation conditions of the branch, thereby minimizing the steady-state voltage deviation of the distribution network branch, optimizing the suppression of power fluctuations in the photovoltaic-storage cluster, and controlling the operating losses of energy storage voltage regulation. The multi-objective collaborative regulation and optimization model uses the real-time voltage offset and real-time power fluctuation of the branch as optimization inputs, and the dual-layer nested capacity range, the upper limit of equipment aging output, and the distribution network voltage operating threshold as global hard constraints. Through step-by-step optimization, the optimal active and reactive power regulation components adapted to the single-branch disturbance conditions are obtained, and finally, the initial power regulation command and voltage regulation command of the branch are obtained.

[0043] First, calculate the real-time operating deviation parameter of the branch. The specific calculation formula is as follows: ; ; In the formula: Let be the real-time voltage offset of the i-th branch; This is the measured voltage of the current PCC bus in the branch. This is the steady-state setpoint for the branch voltage. This refers to the real-time power fluctuation of the branch circuit. This represents the current actual output power of the branch optical storage; The steady-state reference power of the branch is obtained based on medium- and long-term forecasts.

[0044] A multi-objective optimization objective function is constructed by combining real-time deviation parameters. Voltage deviation and power fluctuations are suppressed through a squared penalty, while energy storage voltage regulation losses are constrained to ensure both precise regulation and economical equipment operation. The objective function expression is as follows: ; In the formula: The target value for comprehensive regulation of branch roads; Normalized data for branch voltage offset; Normalized data for branch power fluctuations; Normalized data for the operating losses during this voltage regulation process of energy storage; , , The weights are voltage regulation weight, power smoothing weight, and loss constraint adaptive weight, respectively. These three types of weights are normalized in real time and can be dynamically and adaptively adjusted according to the branch disturbance intensity. Under disturbance conditions, the voltage and power regulation weights are increased, while under steady-state conditions, the loss constraint weight is increased. In this embodiment... , , .

[0045] The model solution relies on setting multi-level hard constraints based on the boundary of branch regulation resources, while also superimposing distribution network voltage safety operation constraints and energy storage aging lockout constraints. The complete constraint conditions are as follows: ; In the formula: , The upper and lower limits for safe operation of distribution network branch voltage; , The upper and lower limits of the steady-state voltage regulation capacity range of the branch circuit are used to constrain the daily steady-state control output. , The upper and lower limits of the dynamic emergency voltage regulation capacity range of the branch line are used to adapt to the output adjustment under extreme disturbance conditions. The maximum voltage regulation output for energy storage aging lockout serves as the ultimate output limit of the equipment and must not be exceeded under any operating conditions.

[0046] Based on the established multi-objective collaborative control optimization model, a sequential quadratic programming algorithm is used for iterative solution to obtain the initial control commands for each branch in stages. The first step solves for the optimal active and reactive power compensation control components of the branch to match power fluctuation suppression requirements. The second step calculates the branch voltage correction control amount based on the coupling relationship between the active and reactive power control components. The third step performs amplitude limiting verification on the solution results in conjunction with the dual-layer capacity constraint interval, forcibly constraining the optimized output within the adjustable resource boundary range of the branch. Finally, the voltage correction control command and power smoothing control command are integrated to form a complete preliminary control command for each branch that adapts to the current branch disturbance conditions, conforms to the aging state of energy storage, and meets the distribution network safety constraints. After completing the optimization solution for each branch, the preliminary control commands of all branches across the entire region are summarized to form a set of preliminary control commands for the entire photovoltaic-storage cluster.

[0047] Furthermore, the process for obtaining compliance and coordination control instructions is as follows: The system calls upon a pre-defined branch simulation model to synchronously replicate the real-time topology, photovoltaic and energy storage parameters, and line impedance operating status of each branch. It inputs the preliminary control commands of each branch into the corresponding branch simulation model to perform branch operating condition simulation and complete multi-dimensional risk verification of power backfeed, voltage reverse limit exceedance, and energy storage overload. It eliminates preliminary control commands that pose secondary risks to the power grid, performs full-domain collaborative adaptation and correction on the remaining commands, and outputs compliant collaborative control commands without operational risks.

[0048] like Figure 4As shown, this embodiment builds a separate branch simulation model for each independent topology distribution network branch. This branch simulation model adopts a layered modular architecture, consisting of a real-time parameter synchronization module, four core simulation calculation units, a dynamic operating condition deduction module, and a risk assessment output module. The four core simulation units are the core operating carriers of the model, and their specific components and functions are as follows: The topology simulation unit collects real-time topology information such as the switching status of distribution network branches, branch topology connection nodes, cable connections, and photovoltaic and energy storage device mounting points, dynamically updating the branch physical topology architecture and restoring the actual grid connection form of each branch; the line impedance simulation unit stores inherent parameters such as branch line length, rated resistance, and rated reactance, and dynamically corrects line impedance micro-offsets by combining real-time ambient temperature and branch load current fluctuation data, replicating the actual impedance operating characteristics of the line and avoiding the influence of fixed impedance parameters. The simulation unit addresses simulation deviations; it inputs the rated capacity, power conversion efficiency, and maximum output limit of the branch photovoltaic inverter, as well as the rated capacity, PCS charge and discharge response rate, short-term overload threshold, and continuous operation tolerance parameters of the energy storage system, and synchronizes the current operating status of the equipment in real time to simulate the power output, response lag, and overload constraint characteristics of the photovoltaic and energy storage equipment; the load condition simulation unit relies on historical time-series data to solidify the typical load characteristic curves of the branch, and updates the branch baseline load, load fluctuation range, and peak-valley load characteristics in real time to dynamically replicate the real-time power consumption conditions of the branch, ensuring that the simulated load completely matches the actual operating conditions on site.

[0049] This embodiment imports the preliminary control commands obtained from solving all branches across the entire domain into the branch simulation model one by one according to the branch correspondence and time sequence. It fully simulates the dynamic response process of branch AC voltage, AC / DC power, energy storage PCS operating status, and line power flow after the control commands are issued. It specifically performs multi-dimensional verification of three typical secondary grid risks: power backfeed, voltage reverse limit exceedance, and instantaneous energy storage overload. The specific verification process and judgment criteria for each type of risk are as follows: First, power backfeed risk verification: During the simulation, the power flow direction and power value at the branch grid connection point are monitored in real time, and the line power flow changes after the control action are continuously tracked. The power flow direction of the branch's photovoltaic-energy storage output power is compared with the input power of the upstream feeder. The allowed reverse transmission threshold is set to 5% of the branch's rated capacity. If, after control, the branch's photovoltaic-energy storage output continues to be transmitted in reverse to the upstream main grid, and the backfeed power amplitude exceeds the preset reverse transmission threshold for two or more consecutive 10ms time windows, it is determined that the preliminary control command for that branch has a secondary risk of power backfeed. Secondly, voltage reverse limit risk verification: A multi-point synchronous monitoring mechanism is adopted to conduct full-area voltage timing monitoring at the branch grid connection point, the middle section of the line, and the load concentration node at the end. The ±7% rated voltage specified in the distribution network regulations is used as the safe operating range to verify the dynamic voltage change trend after the execution of the control command. If the control action causes abnormal reverse voltage deviation in the branch, resulting in excessive voltage rise or fall, or continuously exceeding the safe operating range, the command is deemed to have a secondary risk of voltage reverse limit exceedance. Thirdly, energy storage instantaneous overload risk verification: Real-time simulation tracks the instantaneous charging and discharging power and AC / DC output current parameters of the energy storage PCS. Combined with the short-term overload constraint standard of the energy storage equipment, the instantaneous overload judgment threshold is set to 1.2 times the rated output power, with a continuous withstand time not exceeding 50ms. If the initial control command requires the instantaneous output of the energy storage to exceed this threshold, or the short-term impact load exceeds the equipment's withstand time, it will cause overload impact on the energy storage devices and aggravate device losses. The command is deemed to have an energy storage instantaneous overload operation risk.

[0050] Based on the above multi-dimensional simulation and quantitative risk verification results, compliant collaborative control commands without operational risks are obtained. The first step involves risk command elimination and blocking. All preliminary control commands are screened branch by branch and time sequence by time sequence. If a single command triggers any secondary risk judgment condition, it is directly judged as an invalid risk command. Severely excessive limits are directly eliminated, and amplitude reduction blocking is implemented for critically excessive limits. The second step involves global collaborative adaptation and correction. For quasi-compliant commands without independent risks in a single branch but with superimposed disturbances in multi-branch synchronous control, global collaborative optimization is carried out based on the voltage coupling correlation characteristics between branches and the complementary characteristics of photovoltaic and energy storage power. By fine-tuning the output amplitude of each branch's control command and adjusting the start-up sequence, the voltage and power distribution of each branch in the entire domain are balanced, resolving the superimposed problems such as implicit voltage deviation, global power imbalance, and local operating condition resonance caused by multi-branch coordinated control, ensuring the stability of multi-branch collaborative control. The third step is to standardize and output compliant instructions, summarize all effective control instructions that have eliminated risks and completed global collaborative correction, unify the instruction timing format, output amplitude accuracy and control priority, and standardize them to form a set of compliant collaborative control instructions.

[0051] Furthermore, the process of obtaining the dataset for the control operation is as follows: The compliant coordinated control command is transmitted to the corresponding branch photovoltaic inverter and energy storage PCS equipment, and the branch DC damping adjustment and AC power voltage regulation are triggered simultaneously. The DC side dynamically adjusts the PCS virtual damping parameters to suppress the AC-DC coupling oscillation of the branch, and the AC side matches the reactive and active power output to complete the bus voltage correction. After the control is completed, the AC-DC electrical parameters, energy storage operation parameters and bus voltage parameters of each branch are collected in real time and compiled into a control operation dataset.

[0052] This embodiment employs an AC / DC coordinated control mechanism that links DC damping oscillation suppression with AC voltage and power correction. Relying on compliant coordinated control commands that have undergone verification, it achieves precise coordinated control of multi-branch photovoltaic-storage clusters. After the compliant coordinated control commands for the entire domain are properly configured and output, each branch's dedicated control command is precisely sent point-to-point to the corresponding photovoltaic grid-connected inverter and energy storage PCS equipment according to the unique mapping relationship of the branch equipment. Upon receiving the command, the equipment immediately responds synchronously, simultaneously initiating branch-side DC damping adjustment and AC-side power voltage regulation, achieving AC / DC coordinated control and solving the problem that a single control method cannot simultaneously address oscillation suppression and precise voltage stabilization.

[0053] During DC-side regulation, to address the AC / DC coupled oscillation problem caused by the DC impedance deviation of photovoltaic power distribution branches and the DC ripple disturbance of energy storage, adaptive oscillation suppression is achieved by dynamically adjusting the virtual damping parameters of the energy storage PCS in real time. This scheme constructs a calculation formula for the virtual damping adjustment of the PCS DC side, relying on the real-time DC ripple current to dynamically correct the damping impedance. The specific formula is as follows: ; In the formula: The virtual damping impedance is updated in real time for PCS; The inherent basic damping impedance of the PCS is a fixed parameter of the equipment. It is a damping adaptive adjustment coefficient that can be dynamically adapted according to the branch oscillation intensity; This represents the current real-time DC ripple current of the energy storage PCS. During the control process, the equipment collects the DC side ripple current fluctuation value in real time. When there is AC / DC coupling oscillation in the branch and the ripple current distortion increases, the virtual damping impedance is automatically increased to suppress the oscillation divergence caused by DC side power disturbance and impedance drift. When the branch operation tends to be steady-state and the ripple current is stable, the damping parameter falls back to the base value to avoid increased energy storage operation losses caused by excessive damping.

[0054] During AC-side regulation, bus voltage correction and power smoothing are carried out simultaneously in conjunction with DC damping vibration suppression. Based on the active and reactive power output ratios of compliant regulation commands, the AC output characteristics of the photovoltaic inverter and the energy storage PCS are matched in a layered manner. Specifically, by adjusting the reactive power generation and absorption capacity of the photovoltaic inverter, the reactive power deficit in the branch is quickly compensated, resolving minor voltage deviations. By adjusting the active power charging and discharging output of the energy storage PCS, the active power imbalance caused by fluctuations in branch photovoltaic output and sudden load changes is smoothed, correcting significant over-limit issues of the PCC bus voltage. The timing of the AC and DC-side regulation actions is synchronized and their functions are complementary. The DC side is responsible for suppressing high-frequency coupled oscillations and solidifying the foundation for steady-state operation of the branch, while the AC side is responsible for accurately correcting voltage deviations and smoothing power fluctuations.

[0055] After the entire AC / DC coordinated control operation is completed and the branch operating conditions stabilize, the full-dimensional operation parameter collection and aggregation process is initiated to complete the standardization and construction of the control operation dataset. Through the full-domain acquisition terminals of each branch, electrical parameters such as AC bus voltage, three-phase current, and active and reactive power of the controlled branch are collected uniformly. Simultaneously, photovoltaic operating parameters such as actual output and operating efficiency of the photovoltaic inverter are collected, as well as energy storage control parameters such as real-time damping parameters of the energy storage PCS, charging and discharging power, DC ripple current, and cell operating status are collected, while matching the current branch environmental operating conditions. All collected real-time parameters after control are filtered, deduplicated, and standardized. Invalid data with transient changes during the control transition phase are removed, and steady-state valid operating data is retained. The data is then categorized and aggregated according to branch dimensions, with unified field formats and precision, ultimately forming a branch control operation dataset that is comprehensive in dimensions, continuous in time sequence, and accurately matches the control effect.

[0056] Furthermore, the process of obtaining the control deviation is as follows: The preset voltage stability target value, power smoothing target value, and energy storage loss control target value of each branch are retrieved to form a control benchmark target set. The actual operating parameters of the control operation dataset are compared with the corresponding control benchmark target set dimension by dimension to obtain the voltage deviation, power deviation, and energy storage operation deviation. The voltage deviation, power deviation, and energy storage operation deviation are integrated, normalized, and then the real-time control deviation of each branch is obtained.

[0057] First, a set of benchmark targets for regulation is constructed. These benchmark targets are all generated based on the branch's rated operating parameters, historical best steady-state operating conditions, and the standard setpoints for distribution network operation. Specifically, the voltage stability target is the standard steady-state voltage of the PCC bus under the branch's rated operating conditions; the power smoothing target is the steady-state benchmark output obtained from the branch's medium- and long-term forecasts, used to characterize the branch's optimal power operation state without disturbance; and the energy storage loss control target is the optimal steady-state operating loss of energy storage under the current aging state, obtained by fitting the energy storage cell aging coefficient and conventional charge-discharge efficiency. These three types of benchmark parameters together constitute the set of benchmark targets for evaluating the regulation effect of this branch.

[0058] Based on the control operation dataset and benchmark target parameters, the original deviations after branch control are calculated dimension by dimension, yielding the original voltage deviation, original power deviation, and original energy storage operation deviation. The specific calculation formulas are as follows: ; ; ; In the formula: This represents the original deviation of the branch voltage. To regulate the actual operating voltage of the downstream branch PCC bus; This is the steady-state reference target value for the branch voltage; This represents the original deviation of the branch power. To regulate the actual output power of the downstream branch optical storage; The target value for smoothing branch power; This represents the initial deviation in energy storage operation. This represents the actual operating losses of energy storage during this regulation process; This represents the optimal baseline loss for energy storage under current operating and aging conditions.

[0059] This embodiment uses extreme value normalization to perform dimensionless compression on various original deviations, mapping all deviations to the 0-1 interval to obtain normalized deviation parameters. The normalization calculation formula is as follows: ; ; ; In the formula: , , These are the normalized deviations for voltage, power, and energy storage loss, respectively. This represents the maximum allowable controllable loss of the branch energy storage under its current aging state.

[0060] Finally, based on the normalized three types of deviation parameters, multi-dimensional fusion calculation is performed using adaptive weights to obtain a real-time comprehensive control deviation that can comprehensively characterize the branch control effect. The multi-dimensional control deviation normalization fusion formula in this embodiment is as follows: ; In the formula: This represents the deviation of the branch line's real-time integrated control. The larger the value, the more significant the deviation between the actual operating conditions and the ideal reference operating conditions.

[0061] Furthermore, the correction process for the branch-coupled electrical characteristic model parameters and the two-layer regulating capacity constraint range is as follows: The control deviation is used as a correction factor to iteratively update the coefficient parameters of the branch coupled electrical characteristic model. Based on the energy storage operation loss and operating condition changes corresponding to the real-time deviation, the upper and lower limit thresholds of the dual-layer regulation capacity constraint range of each branch are dynamically fine-tuned. After the correction of the branch coupled electrical characteristic model parameters and the dual-layer regulation capacity constraint range is completed, the updated data is sent back to the corresponding functional module.

[0062] Based on the control deviation, this embodiment constructs an adaptive closed-loop iterative correction mechanism for operating conditions, and simultaneously realizes dynamic iterative updates of branch coupled electrical characteristic model parameters and adaptive fine-tuning of the dual-layer regulation capacity constraint range.

[0063] The first step is the iterative correction of the branch coupling electrical characteristic model parameters. In this embodiment, the gradient descent algorithm is used to iteratively update the model parameters, with the optimization objective of minimizing the overall control deviation. This involves reversely correcting the core parameters of the branch AC / DC coupling model, including the branch equivalent impedance characteristic coefficient, voltage coupling correlation coefficient, DC ripple disturbance weight, power fluctuation response coefficient, and other key modeling parameters, to adapt to the electrical characteristic deviations during long-term operation of the branch. The iterative update formula for the coupling model parameters is as follows: ; In the formula: These are the original core parameters of the branch coupling electrical characteristic model for the current iteration cycle; Adjust parameters for the iteratively updated model; The learning rate is used to control the magnitude of parameter correction in a single iteration, avoiding excessive correction that could lead to model oscillation and instability. In this embodiment, a fixed learning rate is set. The model parameter gradient represents the sensitivity of each core parameter to the overall control deviation. During the iteration process, the system modifies various model parameters differently according to the parameter gradient weights, increasing the correction intensity for electrical coupling parameters with high deviation sensitivity and making small adjustments to parameters with low steady-state sensitivity, ensuring the targeted and accurate nature of the model correction. After the iteration is completed, the branch coupling electrical characteristic model is updated.

[0064] The second stage involves dynamic correction of the branch's dual-layer regulation capacity constraint range. Based on the actual operating losses of the energy storage after regulation, voltage and power deviation trends, and changes in energy storage aging status, the upper and lower limits of the steady-state voltage regulation capacity range and the dynamic emergency voltage regulation capacity range are fine-tuned in conjunction. First, considering the actual energy storage loss deviation during this regulation, the maximum voltage regulation output threshold for energy storage aging lockout is corrected to adapt to the real-time aging and degradation characteristics of the battery cells. Second, based on the long-term voltage deviation trend of the branch and the fluctuation amplitude of photovoltaic-energy storage output, the upper and lower limits of the steady-state capacity range are fine-tuned to adapt to changes in the branch's normal steady-state operating conditions. Finally, based on the intensity of short-term extreme operating condition disturbances, the margin of the dynamic emergency capacity range is dynamically corrected to improve the regulation adaptability under extreme operating conditions.

[0065] After completing the model parameter iteration and dual-layer capacity range correction, the updated branch coupled electrical characteristic model and the new dual-layer regulation resource boundary parameters are uniformly stored back to the corresponding functional modules of each branch, replacing the original old model parameters and old capacity constraint thresholds, thus completing the adaptive update and iteration of the entire branch regulation system. At the same time, based on the real-time regulation loss and deviation trend, the upper and lower limits of the branch steady-state and emergency dual-layer capacity range are dynamically fine-tuned, realizing the dynamic self-updating of model parameters and capacity boundaries according to the distribution network operating conditions and energy storage aging status.

[0066] The processes described above with reference to the flowcharts in the embodiments disclosed in this invention can be implemented as computer software programs. The embodiments disclosed in this invention include a computer program product comprising a computer program carried on a computer-readable medium, the computer program containing program code for performing the methods shown in the flowcharts. In such embodiments, the computer program can be downloaded and installed from a network via a communication component, and / or installed from a removable medium. When the computer program is executed by a central processing unit (CPU), it performs the functions defined in the methods of this application. It should be noted that the computer-readable medium described above in this application can be a computer-readable signal medium or a computer-readable storage medium, or any combination of the two. A computer-readable storage medium can be, for example, but not limited to, an electrical, magnetic, optical, electromagnetic, infrared, or semiconductor system, apparatus, or device, or any combination thereof. More specific examples of computer-readable storage media may include, but are not limited to: electrical connections having one or more wire segments, portable computer disks, hard disks, random access memory (RAM), read-only memory (ROM), erasable programmable read-only memory (EPROM or flash memory), optical fiber, portable compact disk read-only memory (CD-ROM), optical storage devices, magnetic storage devices, or any suitable combination thereof. In this application, a computer-readable storage medium can be any tangible medium containing or storing a program that can be used by or in connection with an instruction execution system, apparatus, or device. In this application, a computer-readable signal medium may include a data signal propagated in baseband or as part of a carrier wave, carrying computer-readable program code. Such propagated data signals can take various forms, including but not limited to electromagnetic signals, optical signals, or any suitable combination thereof. A computer-readable signal medium can also be any computer-readable medium other than a computer-readable storage medium, which can send, propagate, or transmit a program for use by or in connection with an instruction execution system, apparatus, or device. The program code contained on a computer-readable medium may be transmitted using any suitable medium, including but not limited to: wireless segments, wire segments, optical fibers, RF, etc., or any suitable combination thereof.

[0067] The flowcharts and block diagrams in the accompanying drawings illustrate the architecture, functionality, and operation of possible implementations of systems, methods, and computer program products according to various embodiments of the present invention. In this regard, each block in a flowchart or block diagram may represent a module, segment, or portion of code containing one or more executable instructions for implementing a specified logical function. It should also be noted that in some alternative implementations, the functions indicated in the blocks may occur in a different order than those indicated in the drawings. For example, two consecutively indicated blocks may actually be executed substantially in parallel, and they may sometimes be executed in reverse order, depending on the functions involved. It should also be noted that each block in the block diagrams and / or flowcharts, and combinations of blocks in the block diagrams and / or flowcharts, can be implemented using a dedicated hardware-based system that performs the specified function or operation, or using a combination of dedicated hardware and computer instructions.

[0068] Those skilled in the art should understand that the embodiments of the present invention described above and shown in the accompanying drawings are merely examples and do not limit the present invention. The purpose of the present invention has been fully and effectively achieved. The functions and structural principles of the present invention have been shown and explained in the embodiments. Without departing from the stated principles, the implementation of the present invention may have any variations or modifications.

Claims

1. A power-voltage co-support control system for an optical storage cluster, characterized by, include: The data acquisition module is used to collect multi-source heterogeneous data from distribution network branches, unify the time series reference, match the time-aligned multi-source heterogeneous data to the corresponding distribution network branches, and obtain a hierarchical standardized electrical dataset. The disturbance tracing module constructs a branch coupled electrical feature model based on a hierarchical standardized electrical dataset, extracts the branch time-series residuals, and reverse-matches the distribution network branch to which the disturbance belongs to obtain the disturbance location result. The capacity constraint planning module quantifies the multi-dimensional aging characteristics of energy storage equipment based on the disturbance location results, and divides the dual-layer regulation capacity constraint interval of each distribution network branch in combination with the medium and long-term output forecast, and outputs the branch regulation resource boundary. The simulation pre-verification module uses the branch adjustment resource boundary as a constraint to solve the initial control instructions of the branch. It combines the branch simulation model to perform working condition simulation and verification, eliminates instructions with the risk of exceeding the limit, and outputs compliant collaborative control instructions. The AC / DC coordinated control module synchronously executes DC damping adjustment and AC power adjustment based on compliant coordinated control commands, and obtains a control operation dataset by combining the actual operating parameters of the equipment. The adaptive iterative module calculates the difference between the control operation dataset and the control target value to obtain the control deviation. It then uses the control deviation to synchronously correct the parameters of the branch coupled electrical characteristic model and the two-layer control capacity constraint range.

2. The power-voltage co-support control system for optical cluster storage according to claim 1, wherein The process of obtaining the hierarchical standardized electrical dataset is as follows: Collect AC side voltage, branch power, and load electrical data of each branch of the distribution network, and simultaneously collect multi-source heterogeneous data of DC side dynamic impedance, ripple current, and cell status of photovoltaic and energy storage PCS. By unifying the global timing reference based on the acquisition frequency of multi-source heterogeneous data, the sampling timing deviation of different devices is calibrated to achieve time synchronization and alignment of data across all branches. Based on the physical topology of the distribution network, a unique index is established for each branch. The time-aligned multi-source heterogeneous data is mapped to the corresponding topology branch. The data is collected and organized according to the branch index to form a hierarchical standardized electrical dataset at the branch level.

3. The power-voltage co-support control system for optical cluster storage according to claim 2, wherein The process of obtaining the disturbance localization result is as follows: Read the hierarchical standardized electrical dataset branch by branch and extract the AC / DC electrical characteristics, power fluctuation characteristics, and voltage offset characteristics of each branch. Based on the independent electrical parameters of each branch, a branch coupled electrical characteristic model is constructed, and the real-time electrical operation sequence residual of each branch is calculated. By comparing the timing residuals of each branch with the preset disturbance threshold, the target distribution network branch corresponding to the voltage disturbance and AC / DC coupled oscillation is matched in reverse to locate the disturbance and obtain the branch disturbance location result.

4. The power-voltage co-support control system for optical cluster storage according to claim 3, wherein The process of obtaining the boundary of the branch adjustment resource is as follows: Based on the branch disturbance location results, the incremental internal resistance of the energy storage cell, the cumulative number of cycles and the remaining capacity decay coefficient are collected for each branch to obtain the energy storage aging characteristic parameters of each branch. Historical photovoltaic power output time series data and historical load fluctuation time series data of each branch are collected, and medium- and long-term prediction data of branch photovoltaic power output and load fluctuation are generated through time series extrapolation. Establish a closed mapping relationship between the aging characteristic parameters of energy storage and the upper limit of the voltage regulation output of the branch, limit the regulation output threshold of the high-aging energy storage branch, and combine the medium and long-term forecast data of the branch photovoltaic output and load fluctuation to define the steady-state voltage regulation capacity range and the dynamic emergency voltage regulation capacity range of the branch, forming a nested double-layer capacity constraint range, and output the branch regulation resource boundary.

5. The power-voltage co-support control system for optical cluster storage according to claim 4, wherein The solution process for the preliminary control command is as follows: Based on the branch regulation resource boundary, the double-layer capacity constraint interval is used as a hard constraint condition, and the regulation objective function is constructed by combining the branch real-time voltage deviation and power fluctuation. Solve the control objective function to obtain the initial power and voltage regulation commands that are adapted to the single-branch disturbance condition and conform to the upper limit of the branch capacity, which serve as the initial control commands for the branch. The preliminary control commands from each branch are aggregated to form a set of preliminary control commands for the entire cluster.

6. The power-voltage co-support control system for optical cluster storage according to claim 5, wherein The process of obtaining the compliance-coordinated control instructions is as follows: The preset branch simulation model is invoked to synchronously replicate the real-time topology, optical storage parameters, and line impedance operation status of each branch. The initial control commands for each branch are input into the corresponding branch simulation model to perform branch operating condition simulation and complete multi-dimensional risk verification of power backfeed, voltage reverse over-limit, and energy storage overload. The initial control instructions that pose secondary risks to the power grid are removed, and the remaining instructions are adjusted and adapted across the entire domain to output compliant and coordinated control instructions without operational risks.

7. The power-voltage co-support control system for optical cluster storage according to claim 6, wherein The process of obtaining the control operation dataset is as follows: The compliant and coordinated control command is transmitted to the corresponding branch photovoltaic inverter and energy storage PCS equipment, and the branch DC damping regulation and AC power voltage regulation coordinated action are triggered simultaneously. The DC side dynamically adjusts the PCS virtual damping parameters to suppress AC / DC coupling oscillations in the branch, while the AC side matches the reactive and active power outputs to complete the bus voltage correction. After real-time collection and control, the AC / DC electrical parameters, energy storage operation parameters, and bus voltage parameters of each branch are collected and organized to form a control operation dataset.

8. The power-voltage co-support control system for optical cluster storage according to claim 7, wherein The process for obtaining the control deviation is as follows: The preset voltage stability target value, power smoothing target value, and energy storage loss control target value of each branch are retrieved to form a set of control benchmark targets; By performing a difference calculation on the actual operating parameters of the control operation dataset and the corresponding control benchmark target set in each dimension, the voltage deviation, power deviation and energy storage operation deviation are obtained respectively. By integrating voltage deviation, power deviation, and energy storage operation deviation, and normalizing them, the real-time control deviation of each branch is obtained.

9. The power-voltage co-support control system for optical cluster storage according to claim 8, wherein, The correction process for the branch-coupled electrical characteristic model parameters and the two-layer regulating capacity constraint range is as follows: The control deviation is used as a correction factor to iteratively update the coefficient parameters of the branch coupled electrical characteristic model parameter by parameter; Based on the energy storage operation loss and operating condition changes corresponding to the real-time deviation, the upper and lower limit thresholds of the dual-layer regulation capacity constraint range of each branch are dynamically fine-tuned. After correcting the parameters of the branch coupled electrical characteristic model and the dual-layer regulating capacity constraint range, the updated data will be sent back to the corresponding functional module.

10. A computer-readable storage medium, characterized in that, The computer-readable storage medium stores a computer program that is executed by a processor to implement a power voltage coordinated support control system for photovoltaic-storage clusters as described in any one of claims 1-9.

Images