Source network load storage integrated collaborative optimization and control system

Abstract

The application relates to the technical field of power distribution network operation control, and provides a source-network-load-storage integrated collaborative optimization and control system. The system collects multi-source data of a target station area, constructs equivalent electrical position sequences representing the relative relationship of the voltage sensitivity of each inverter in different phases, and quantifies the dynamic coupling relationship between the output change and the voltage response among nodes based on the equivalent electrical position sequences; the system combines historical off-grid and de-rating events to calibrate the reference coupling strength, introduces real-time photovoltaic output, load level and node voltage margin factors for dynamic correction, and generates a dynamic coupling coefficient matrix; the system identifies the coupling weak points when the output climbs based on the dynamic coupling coefficient matrix, matches energy storage or adjustable load in the electrical neighborhood of the weak points by comprehensively considering the dynamic coupling coefficient, real-time availability and voltage regulation priority, and presets the collaborative voltage regulation scheme to the local resource controller, so that the local autonomous execution is realized when the trigger condition is met, and the cascading off-grid risk is blocked at the minimum cost.

Description

Technical Field

[0001] This invention relates to the field of power distribution network operation control technology, and in particular to an integrated source-grid-load-storage coordinated optimization and control system. Background Technology

[0002] Existing source-grid-load-storage coordinated control methods assume that upper-level dispatch commands can directly control photovoltaic (PV) output. However, in rural low-voltage distribution areas, the local protection logic (overvoltage derating, protection disconnection) embedded in household PV inverters has higher priority than any external commands and cannot be remotely modified or blocked. Meanwhile, rural low-voltage distribution areas have long power lines with many branches, and household PV systems are dispersed and connected to different locations. The sensitivity of each inverter to voltage changes varies significantly depending on its connection location. During the continuous ramp-up of PV output, the inverters with higher voltage sensitivity are the first to reach the protection threshold and disconnect from the grid, triggering a chain reaction of disconnections and causing a precipitous drop in the grid-connectable PV capacity of the distribution area. This chain reaction of disconnections stems from the combined driving force of the same output ramp-up process, and the dual constraints of uncontrollable local protection and differences in connection location render traditional dispatch methods ineffective.

[0003] Existing technologies primarily focus on passive mitigation after voltage limits are exceeded or global optimization based on precise topology models. They fail to achieve topology-free identification and ranking of inverter voltage sensitivity without relying on precise topology parameters and line models. Furthermore, they lack a systematic solution that integrates voltage sensitivity ranking, output ramp-up risk quantification, and adjustable resource coordination within a unified framework. Consequently, they cannot effectively prevent cascading grid disconnection risks at minimal cost during output ramp-up phases. Therefore, there is an urgent need for an optimization and control system that can autonomously identify inverter voltage sensitivity ranking without requiring precise topology, and achieve precise on-site voltage regulation during output ramp-up scenarios through source-grid-load-storage coordination. This would enhance the photovoltaic absorption capacity and operational resilience of rural low-voltage distribution areas. Summary of the Invention

[0004] In order to overcome the defects and shortcomings of existing technologies, this invention provides an integrated source-grid-load-storage collaborative optimization and control system.

[0005] To achieve the above objectives, the present invention adopts the following technical solution: In a first aspect, the present invention provides an integrated source-grid-load-storage collaborative optimization and control system, comprising the following modules: Multi-source acquisition module: Simultaneously acquires photovoltaic inverter operation data, distribution transformer electrical operation data, and adjustable resource status data for the target area; Location sequencing module: Identifies the relative voltage sensitivity relationship through photovoltaic inverter operating data, and determines the equivalent electrical position sequence of photovoltaic inverters in each phase of the target area; Coupling matrix module: Combining the equivalent electrical position sequence, it quantifies the coupling relationship between the output change and voltage response of the photovoltaic inverter, calls the historical operating data of the photovoltaic inverter to calculate the reference coupling strength, and introduces the real-time photovoltaic output factor, the area load level factor and the node voltage margin factor for dynamic correction to generate a dynamic coupling coefficient matrix. Voltage regulation configuration module: Based on the dynamic coupling coefficient matrix, identify coupling vulnerability points, call adjustable resource status data to evaluate the comprehensive protection efficiency of each adjustable resource for coupling vulnerability points, and generate a source-grid-load-storage coordinated voltage regulation scheme.

[0006] According to the above technical solution, the steps of synchronously collecting photovoltaic inverter operation data, distribution transformer electrical operation data, and adjustable resource status data of the target area include: S110. Collect the operating data and phase affiliation information of the photovoltaic inverters within the target area, specifically including: The phase attribution information refers to the single-phase connection of each photovoltaic inverter within the target distribution area. The photovoltaic inverter operation data includes the terminal voltage sequence and active power output sequence collected by each photovoltaic inverter in a time series.

[0007] S120. Collect historical operating data of photovoltaic inverters that have experienced protection actions within the target area, including the action time and output change magnitude of each derating event and grid disconnection event. Specifically, this includes: Collect historical operating data of photovoltaic inverters in the target area, including the action time and output change magnitude of each photovoltaic inverter derating and grid disconnection event.

[0008] S130. Collect electrical operation data of the target transformer substation by phase, specifically including: Collect electrical operation data of the target transformer area, including the active power sequence of the three phases of the low-voltage main outgoing circuit of the transformer collected in time series.

[0009] S140. Collect adjustable resource status data reflecting adjustable load storage resources within the target area, specifically including: Collect adjustable resource status data reflecting adjustable load storage resources within the target area, including the state of charge, charging and discharging power and maximum charging and discharging capacity of household energy storage and distribution transformer energy storage collected in time series, as well as the start-stop status, current power and maximum adjustable power of adjustable load collected in time series.

[0010] According to the above technical solution, the step of identifying the relative voltage sensitivity relationship through photovoltaic inverter operating data and determining the equivalent electrical position sequence of photovoltaic inverters in each phase of the target area includes: S210. Using phase attribution information as the classification standard, calculate the active power output fluctuation amplitude using photovoltaic inverter operating data, and filter the effective time period based on the active power output fluctuation amplitude, specifically including: Using the phase assignment information of photovoltaic inverters as the classification standard, the active power output sequence and terminal voltage sequence of each photovoltaic inverter in the photovoltaic inverter operation data of the most recent preset rolling time window are extracted; After removing periods with missing data, the system is divided into multiple operating segments according to a fixed time window; the active power output fluctuation of a single inverter within each period is calculated, and periods with fluctuation amplitudes greater than a preset threshold are selected as valid periods.

[0011] S220. For the selected effective time periods, the photovoltaic inverter operating data is fitted to obtain the instantaneous voltage sensitivity, which characterizes the voltage change caused by a unit change in active power. Then, the active power output fluctuation amplitude of each effective time period is used as the weight to synthesize a weighted comprehensive voltage sensitivity, which specifically includes: For each effective time period, the active power output change sequence of a single inverter is linearly fitted with its own terminal voltage change sequence to obtain the terminal voltage change amplitude corresponding to the unit active power output change within the effective time period, i.e., the instantaneous voltage sensitivity of a single inverter. Calculate the total active power fluctuation amplitude of the corresponding phase photovoltaic system for each effective time period. Using the normalized fluctuation amplitude as the weight, perform a weighted average of the instantaneous voltage sensitivity of each inverter for each effective time period to obtain the comprehensive voltage sensitivity.

[0012] S230. Based on the comprehensive voltage sensitivity, the photovoltaic inverters within each phase are sorted to generate an equivalent electrical position sequence characterizing the relative relationship of voltage sensitivity, specifically including: Based on the comprehensive voltage sensitivity, each photovoltaic inverter in the phase is sorted from low to high, generating an equivalent electrical position sequence from the beginning (transformer side) to the end. The higher the comprehensive voltage sensitivity, the closer the equivalent position is to the end, and the earlier the protection threshold is reached under the same net power back-feed conditions.

[0013] According to the above technical solution, the steps of quantifying the coupling relationship between photovoltaic inverter output change and voltage response by combining equivalent electrical position sequence, calculating the reference coupling strength by calling historical operating data of photovoltaic inverter, and dynamically correcting by introducing real-time photovoltaic output factor, substation load level factor and node voltage margin factor to generate dynamic coupling coefficient matrix include: S310. Determine the relative positional relationships between photovoltaic inverter nodes based on the equivalent electrical position sequence, and construct coupling paths characterizing the voltage coupling relationship between photovoltaic inverter nodes in conjunction with the photovoltaic backflow direction, specifically including: Based on the equivalent electrical position sequence within each phase, the photovoltaic inverter at the end of the equivalent electrical position sequence is defined as the line end node, and the photovoltaic inverter at the beginning is defined as the distribution transformer beginning node. According to the physical law that the photovoltaic reverse power flow flows from the end to the distribution transformer, when the active power output of the line end node increases, it will cause the voltage of all distribution transformer beginning nodes on the path from the line end node to the distribution transformer to rise synchronously; when the output decreases, the voltage will drop synchronously.

[0014] Based on this, a coupling path between output changes and voltage responses among photovoltaic inverter nodes is constructed to quantify the correlation strength of voltage changes among photovoltaic inverter nodes under the output ramp-up scenario.

[0015] Meanwhile, the output changes at the first-end node of the distribution transformer will also affect the voltage at the end-end node of the line through the common line impedance. The above coupling relationship covers all coupling paths with common electrical impedance within the target area.

[0016] S320. Based on historical operating data of photovoltaic inverters, using events where only a single photovoltaic inverter in the same phase experiences grid disconnection and events where only a single photovoltaic inverter experiences effective derating as calibration samples, the output change amplitude and the terminal voltage change of the affected photovoltaic inverters corresponding to the calibration samples are obtained. Weights are dynamically assigned according to the event type of the grid disconnection event and the effective derating event, and then a weighted sum is calculated to obtain the baseline coupling strength of the coupling path, specifically including: The system retrieves historical operating data of the photovoltaic inverter to obtain the action time and corresponding output change of the photovoltaic inverter for each grid disconnection event and effective derating event. Among them, the effective derating event is the derating event in which the output change is greater than the preset minimum derating threshold, which is used to supplement calibration samples when there are insufficient grid disconnection event samples. Events in which only a single inverter experiences a power output step while the power output fluctuations of the other in-phase inverters are less than a preset stability threshold are selected as valid calibration sample events. Based on the aforementioned action time, the changes in terminal voltage at each distribution transformer head-side node before and after the grid disconnection event and the effective derating event are extracted from the terminal voltage sequence of the photovoltaic inverter operation data. The cumulative number of internet disconnection events that have occurred in history is counted, and the calculation weight of internet disconnection events and effective credit reduction events is dynamically adjusted based on the cumulative number of internet disconnection events. The more cumulative the number of internet disconnection events, the higher the weight of internet disconnection events and the lower the weight of effective credit reduction events. For each coupling path, the output change amplitude and corresponding terminal voltage change of each grid disconnection event and effective derating event are weighted and summed separately using the weighted sum of the output change amplitudes as the denominator and the weighted sum of the terminal voltage changes as the numerator. The change amplitude of the inverter terminal voltage at the first end of the distribution transformer caused by the change of active power output per unit is calculated and used as the reference coupling strength of the coupling path.

[0017] S330, the real-time photovoltaic output factor is obtained by summing and normalizing the current active power output of each photovoltaic inverter in the photovoltaic inverter operation data; the distribution area load level factor is obtained by calculating the ratio of the current target distribution area total load to the historical target distribution area maximum load value after deducting the real-time photovoltaic total output and energy storage charging and discharging power from the distribution transformer outlet active power in the distribution transformer electrical operation data; the node voltage margin factor is calculated by the difference between the current terminal voltage of each photovoltaic inverter and the protection action threshold, specifically including: From the active power output sequence of photovoltaic inverter operation data, extract the active power output of each photovoltaic inverter at the current moment and sum them to obtain the real-time total photovoltaic power output, and generate the real-time photovoltaic power output factor; From the active power sequence of the electrical operation data of the distribution transformer, the active power value corresponding to the current moment is extracted as the total active power output of the distribution area. After deducting the total output of real-time photovoltaic power, the total load of the distribution area is obtained, and the load level factor of the distribution area is generated. From the terminal voltage sequence of photovoltaic inverter operation data, extract the terminal voltage of each photovoltaic inverter at the current moment, calculate the difference between the terminal voltage and the protection action threshold as the real-time voltage margin, and generate the node voltage margin factor.

[0018] S340. For each coupling path, the inherent attenuation coefficient is determined based on the comprehensive voltage sensitivity. Then, according to the equivalent position difference of the photovoltaic inverter nodes in the equivalent electrical position sequence, the weights of the real-time photovoltaic output factor, the substation load level factor, and the node voltage margin factor are dynamically allocated, and a weighted sum is obtained to obtain the comprehensive correction coefficient. The reference coupling strength is then corrected by the inherent attenuation coefficient and the comprehensive correction coefficient to generate the dynamic coupling coefficient, forming a dynamic coupling coefficient matrix, which specifically includes: For each coupling path, the inherent attenuation coefficient of the coupling path is calculated based on the comprehensive voltage sensitivity difference between the corresponding nodes. Based on the equivalent position difference between the corresponding upstream and downstream nodes, determine the dynamic weights of the real-time photovoltaic power output factor, the transformer area load level factor, and the node voltage margin factor. The three factors are summed by dynamic weights to obtain the comprehensive correction coefficient of the coupling path; the baseline coupling strength obtained in S320 is multiplied by the inherent attenuation coefficient and the comprehensive correction coefficient to generate the dynamic coupling coefficient of the corresponding coupling path. The dynamic coupling coefficients of all coupling paths constitute a dynamic coupling coefficient matrix, which is dynamically refreshed as the operating data is updated. It reflects the coupling strength of the influence of the output changes of each photovoltaic inverter on the terminal voltage of other photovoltaic inverters under real-time operating conditions.

[0019] According to the above technical solution, the steps of identifying coupling vulnerability points based on the dynamic coupling coefficient matrix, evaluating the comprehensive protection efficiency of each adjustable resource for coupling vulnerability points by calling adjustable resource status data, and generating a source-grid-load-storage coordinated voltage regulation scheme include: S410. Combining the dynamic coupling coefficient matrix with photovoltaic inverter operating data, predict the voltage rise of each photovoltaic inverter node within the same phase during power output ramp-up. Photovoltaic inverter nodes whose voltage rise exceeds their own voltage margin are marked as coupling vulnerability points, specifically including: Based on the dynamic coupling coefficient matrix, the dynamic coupling coefficient of each coupling path is extracted; Extract the terminal voltage of each photovoltaic inverter at the current moment from the terminal voltage sequence of photovoltaic inverter operation data, and use the difference between the protection action threshold and the current terminal voltage as the current voltage margin; For each photovoltaic inverter, the other photovoltaic inverters in the same phase of the target area are taken as associated nodes. The dynamic coupling coefficient of each associated node to the photovoltaic inverter is extracted, multiplied by the current active power output ratio of each associated node, and then the weighted sum is multiplied by the preset unit ramp rate to obtain the expected increase in terminal voltage of the photovoltaic inverter when the output continues to ramp.

[0020] Photovoltaic inverters whose expected voltage rise exceeds their current voltage margin are marked as coupling vulnerabilities.

[0021] S420. For coupled vulnerable points, based on the dynamic coupling coefficient, a set of photovoltaic inverter nodes constituting the electrical neighborhood is selected in the equivalent electrical location sequence. Within the electrical neighborhood, the comprehensive protection efficiency is obtained by multiplying the dynamic coupling coefficient of adjustable resources to the coupled vulnerable points, the real-time availability determined by the adjustable resource status data, and the voltage regulation priority coefficient determined by the equivalent location difference. Adjustable resources are matched according to the comprehensive protection efficiency to generate a collaborative protection scheme, specifically including: For each coupling vulnerability point, a predetermined number of nodes adjacent to the vulnerability point are selected as candidate nodes in the equivalent electrical location sequence; In the dynamic coupling coefficient matrix, the dynamic coupling coefficient between the coupling vulnerability point and each candidate node is extracted, and candidate nodes with dynamic coupling coefficients greater than the preset coupling threshold are selected to form the electrical neighborhood of the coupling vulnerability point. Call the adjustable resource status data, calculate the energy storage availability based on the current state of charge, charging and discharging power and maximum charging and discharging capacity of the energy storage, and calculate the load availability based on the current start and stop status, current power and maximum adjustable power of the adjustable load; For each adjustable resource in the electrical neighborhood, the voltage regulation priority coefficient is determined based on the equivalent location difference between the adjustable resource and the coupling vulnerability point. Multiply the dynamic coupling coefficient, availability and voltage regulation priority coefficient of the adjustable resource access point to the coupled vulnerable point to obtain the comprehensive protection efficiency of the adjustable resource to the coupled vulnerable point. All coupled vulnerable points are sorted in ascending order of voltage margin. When a coupled vulnerable point is located at the intersection of multiple coupling paths, the corresponding coupled vulnerable point is prioritized. According to the sorting results, the adjustable resources with the highest comprehensive protection efficiency are matched to each coupled vulnerable point in turn, and the matched adjustable resources are removed from the subsequent matching.

[0022] S430. The collaborative protection scheme is pre-configured into the local controller of the matched adjustable resource. When the voltage at the coupled vulnerable point reaches the trigger threshold, the adjustable resource autonomously performs voltage regulation protection locally, specifically including: The matching result between the coupled vulnerability and the adjustable resource is transformed into response logic and preset in the local controller of the matched adjustable resource. When the voltage at the vulnerability rises to the preset trigger threshold, the matched adjustable resource starts adjustment, which can be executed locally autonomously without the need for central scheduling instructions.

[0023] Secondly, this application provides an electronic device, including a processor and a memory, wherein the memory stores a computer program that can be called by the processor, and the processor executes a source-grid-load-storage integrated collaborative optimization and control system by calling the computer program stored in the memory.

[0024] Thirdly, this application provides a computer-readable storage medium storing instructions that, when executed on a computer, cause the computer to execute a source-grid-load-storage integrated collaborative optimization and control system.

[0025] Compared with the prior art, this application has the following advantages and beneficial effects: By constructing an equivalent electrical position sequence based on the sensitivity fitting of the active power output and terminal voltage of the photovoltaic inverter itself, the relative ranking of the voltage sensitivity of each photovoltaic inverter in the distribution area and the identification of overvoltage risk nodes in the power output ramping scenario can be realized without relying on precise topology parameters and line models. By constructing a dynamic coupling coefficient matrix that integrates real-time photovoltaic power output, substation load level, and node voltage margin, the coupling impact of power output changes on the voltage of each node is quantified into a dynamic sensitivity index that can be calculated in real time, thereby enabling the identification of coupling vulnerability points and the ranking of ramp margins. By placing the voltage safety margin of coupled vulnerable points with the dynamic coupling coefficient, real-time availability and voltage regulation priority of adjustable resources under a unified framework for coordinated matching and response logic pre-setting, local voltage regulation protection and integrated source-grid-load-storage coordinated control of the risk of cascading grid disconnection caused by power output ramp-up can be achieved. Attached Figure Description

[0026] Figure 1 This is a schematic diagram of the integrated source-grid-load-storage collaborative optimization and control system structure provided in the embodiments of this application; Figure 2This is a schematic diagram of the process for constructing an equivalent electrical position sequence provided in an embodiment of this application; Figure 3 This is a schematic diagram of the process for generating a dynamic coupling coefficient matrix provided in an embodiment of this application. Detailed Implementation

[0027] The technical solution of this application will be described in detail below with reference to the accompanying drawings and specific embodiments. It should be understood that the embodiments and specific features in the embodiments are detailed descriptions of the technical solution of this application, rather than limitations thereof. In the absence of conflict, the embodiments and technical features in the embodiments can be combined with each other.

[0028] Please see Figure 1 , Figure 1 This is a schematic diagram of the integrated source-grid-load-storage collaborative optimization and control system structure provided in this application embodiment, which specifically includes the following modules: Multi-source acquisition module: Simultaneously acquires photovoltaic inverter operation data, distribution transformer electrical operation data, and adjustable resource status data of the target area.

[0029] In this embodiment, the synchronous collection of photovoltaic inverter operation data, distribution transformer electrical operation data, and adjustable resource status data of the target area includes the following specific contents: S110. Collect the operating data and phase affiliation information of the photovoltaic inverters in the target area.

[0030] Using a preset sampling frequency, the system obtains the operating data of each photovoltaic inverter from the local communication module of each photovoltaic inverter in the target area, collects the effective voltage value at the AC side grid connection point of each photovoltaic inverter, and forms a terminal voltage sequence containing the sampling time and the effective voltage value (in volts). The terminal voltage directly reflects the actual voltage level at the grid connection point of the photovoltaic inverter. The active power value (in kilowatts) injected by each photovoltaic inverter into the target area power grid is collected synchronously, forming an active power output sequence that includes the collection time and active power value. The active power output sequence directly reflects the contribution of each photovoltaic inverter to the power injection of the target area. The terminal voltage and active power output are directly measured and obtained by the metering unit built into the photovoltaic inverter, which can truly reflect the actual electrical status of each photovoltaic inverter at the grid connection point. In this embodiment, the preset acquisition frequency is set to 1 time / second. This setting is based on the fact that the photovoltaic output of rural low-voltage distribution areas can significantly increase within seconds to tens of seconds when clouds move or sunlight intensifies, causing the inverter terminal voltage to rise rapidly. Using a second-level acquisition frequency can fully capture the dynamic changes in the voltage and active power output of each inverter during the output increase process, providing sufficiently fine time resolution for subsequent voltage sensitivity fitting and calibration of the coupling matrix's reference strength. Furthermore, current mainstream residential photovoltaic inverters and smart distribution area terminals generally support second-level data reporting, eliminating the need for additional hardware upgrades.

[0031] Obtain the phase assignment information of each photovoltaic inverter from the power grid marketing system. The phase assignment information is the single phase that each photovoltaic inverter is connected to in the low-voltage distribution network of the target area, specifically phase A, phase B, or phase C. Photovoltaic inverter operation data and phase attribution information are associated and stored according to the device identifier, forming a structured dataset with the device identifier as the primary key and containing phase attribution information and acquisition time, including terminal voltage sequence and active power output sequence.

[0032] S120. Collect historical operating data of photovoltaic inverters that have taken protective actions within the target area, including the action time and output change range of each derating event and grid disconnection event.

[0033] Retrieve historical protection action records from the local event logs of each photovoltaic inverter or the target area data concentrator, and extract relevant data for each derating event and grid disconnection event for each photovoltaic inverter.

[0034] For each derating event, record the moment when the photovoltaic inverter derating action occurs as the action time; record the active power output value at the most recent data acquisition moment before the derating action as the active power output before the derating action; Based on the action time, the active power output value corresponding to the first acquisition moment after the action time is extracted from the active power output sequence of the corresponding derating photovoltaic inverter, and is taken as the active power output after derating; the difference between the active power output before derating and the active power output after derating is taken as the output change amplitude.

[0035] For each grid disconnection event, the moment when the grid disconnection action occurs is recorded as the action time; the active power output value of the corresponding photovoltaic inverter at the most recent acquisition moment before the grid disconnection action is recorded as the active power output before grid disconnection, and the active power output before grid disconnection is used as the output change amplitude.

[0036] The historical derating and grid disconnection event records of each photovoltaic inverter are associated and stored according to the device identifier and the action time, forming a historical operation dataset with the device identifier as the primary key and the action time as the index.

[0037] S130. Collect electrical operation data of the target transformer substation by phase.

[0038] The active power values ​​of the low-voltage main outgoing circuit of the distribution transformer are collected from the terminal unit of the target distribution transformer at a preset sampling frequency, and the active power sequence containing the sampling time and the active power value of each phase is formed. The unit is kilowatt. The active power sequence directly reflects the active power transmission level of the main outgoing circuit of the low-voltage side of the distribution transformer. The power direction is positive when the distribution transformer feeds into the distribution area and negative when the distribution area feeds back into the distribution transformer.

[0039] The preset acquisition frequency is consistent with S110, set to 1 time / second. This setting is based on the principle that second-level acquisition can fully capture the dynamic changes in power before and after a grid disconnection event. The active power sequence is stored in association by phase and acquisition time to form a structured distribution transformer electrical operation dataset.

[0040] S140. Collect adjustable resource status data reflecting adjustable load storage resources within the target area.

[0041] Collect status data of each adjustable resource within the distribution area. Adjustable resources are devices within the target distribution area that are connected to the control system, have remote power adjustment capabilities, and whose users have authorized to participate in the control, including residential energy storage, distribution transformer energy storage, and adjustable loads that have been equipped with intelligent control modules.

[0042] At a preset sampling frequency, the state of charge (SOC), current charging / discharging power, and maximum charging / discharging capacity of each energy storage unit are collected from the management systems of residential and distribution transformer energy storage systems. SOC represents the percentage of the energy storage unit's current remaining capacity relative to its rated capacity; current charging / discharging power is the actual charging / discharging power value of the energy storage unit, with positive values ​​indicating charging and negative values ​​indicating discharging, measured in kilowatts (kW); maximum charging / discharging capacity is the maximum charging and discharging power that the residential energy storage unit can achieve under its current healthy state, also measured in kilowatts. This data forms an energy storage status data sequence that includes the sampling time, SOC, current charging / discharging power, and maximum charging / discharging capacity.

[0043] At a preset sampling frequency, the start / stop status, current power, and maximum adjustable power of each adjustable load are collected from the intelligent control module of the adjustable load. Start / stop status is indicated by a switch input indicating whether the load is currently running or switched off; current power is the actual power consumption of the load, in kilowatts; maximum adjustable power is the maximum operating power the load can achieve under rated conditions, in kilowatts. This data forms a load status data sequence that includes the sampling time, start / stop status, current power, and maximum adjustable power.

[0044] The preset acquisition frequency is consistent with S110, set to 1 time / second. If there are no adjustable resources in the target area or the communication link of the adjustable resources is interrupted, the load status data sequence will be empty, and the subsequent voltage regulation configuration module will only rely on the reactive power regulation capability of the photovoltaic inverter itself or the active power limiting method to perform protection.

[0045] Location sequencing module: Identifies the relative voltage sensitivity relationship through photovoltaic inverter operating data, and determines the equivalent electrical position sequence of photovoltaic inverters in each phase of the target area.

[0046] In this embodiment, as Figure 2 As shown, the equivalent electrical position sequence of photovoltaic inverters in each phase of the target area is determined by identifying the relative voltage sensitivity relationship through photovoltaic inverter operating data, including the following specific details: S210. Using phase attribution information as the classification standard, calculate the active power output fluctuation range using photovoltaic inverter operating data, and select the effective time period based on the active power output fluctuation range.

[0047] Using the phase affiliation information of photovoltaic inverters as the classification standard, for each phase, the active power output sequence and terminal voltage sequence of each photovoltaic inverter within the most recent preset rolling time window are extracted.

[0048] The preset rolling time window is a sliding time range that extends backward by a preset number of days from the current time, used to acquire recent operating data covering different photovoltaic power output and load levels. In this embodiment, the preset number of days is set to 7 days by default. The reason for this setting is that a 7-day window can fully cover different weather types (sunny, cloudy, rainy) and load change patterns (weekdays and weekends) within a natural week, providing sufficient diversity of operating conditions for the subsequent selection of effective time periods; at the same time, the rolling time window length is moderate, which can avoid the decrease in the representativeness of historical data due to seasonal power output characteristics drift or changes in transformer topology.

[0049] In practice, the preset number of days can be determined to other values ​​according to actual needs. For example, in areas where there are significant seasonal differences in photovoltaic output, it can be extended to 30 days to include a wider range of operating modes. When there are recent changes in the topology of the distribution area or the connection of the inverter, the window can be shortened or the data can be re-accumulated from the time of the change to ensure that the data within the window reflects the current system structure.

[0050] To eliminate data gaps, the active power output sequence and terminal voltage sequence are traversed to detect discontinuous data intervals caused by communication interruptions, equipment outages, or abnormal acquisition. Continuous periods with missing data at any acquisition time in the sequence are removed from the analysis scope, retaining only continuous running periods with complete and uninterrupted data.

[0051] For individual isolated data collection points that are missing during a continuous running session (in this embodiment, the number of consecutive missing points is assumed to be no more than 3), linear interpolation is used to fill in the missing points, which does not affect the effectiveness of the continuous running session.

[0052] After removing periods with missing data, the continuous running data is divided into multiple running segments with a fixed duration. The fixed duration is the preset time span of each analysis window, which is set to 15 minutes by default in this embodiment. The basis for this setting is that a 15-minute window can cover the typical fluctuation process of photovoltaic output caused by cloud movement or changes in sunshine, and the number of data points within the window (approximately 900 sampling points at a collection frequency of 1 time / second) is sufficient to support the statistical sample size required for subsequent correlation calculations. In specific implementations, the fixed duration can also be adjusted to other values ​​according to the photovoltaic installed capacity and output fluctuation characteristics of the distribution area.

[0053] Calculate the active power output fluctuation of each photovoltaic inverter during each operating period, that is, the difference between the maximum and minimum active power output during the operating period.

[0054] The effective time periods are selected based on the fluctuation range of active power output exceeding a preset threshold. The preset threshold is based on two conditions: the fluctuation range of active power output must be greater than both the relative fluctuation threshold and the absolute fluctuation threshold.

[0055] The relative threshold for fluctuation is based on the rated power of the corresponding inverter itself. In this embodiment, it is set to 20% by default. The reason for this setting is that when the active power output fluctuation of a single inverter is less than 20% of its rated power, the voltage change caused by the output change is small and is on the same order of magnitude as the voltage measurement noise collected at the second level. The instantaneous voltage sensitivity obtained by linear fitting is easily affected by noise and has poor repeatability. Setting it to 20% can achieve a balance between excluding low signal-to-noise ratio periods and retaining enough effective samples.

[0056] The absolute value threshold for fluctuation is set to 1kW by default in this embodiment. The reason for this setting is that there are large differences in the specifications of rural household photovoltaic inverters. For low-power inverters, relying solely on the percentage threshold may result in a fluctuation threshold that is too low and difficult to distinguish from measurement noise. Increasing the lower limit of the absolute value can ensure that all inverters have a sufficient signal-to-noise ratio during their effective time periods.

[0057] In practice, both the relative fluctuation threshold and the absolute fluctuation threshold can be adjusted to other values ​​based on the actual photovoltaic output fluctuation characteristics and data quality of the distribution area.

[0058] S220. For the selected effective time periods, the photovoltaic inverter operation data is fitted to obtain the instantaneous voltage sensitivity, which characterizes the voltage change caused by a unit change in active power. Then, the active power output fluctuation amplitude of each effective time period is used as the weight to synthesize the comprehensive voltage sensitivity.

[0059] For each effective time period, the active power output change sequence of each photovoltaic inverter is linearly fitted with its own terminal voltage change sequence to obtain the instantaneous voltage sensitivity of the inverter within that effective time period.

[0060] The active power output change sequence is constructed as follows: The active power output sequence collected by S110 is subjected to moving average filtering. The filter window width is set to 5 sampling points by default in this embodiment to reduce noise and smooth the signal. On the filtered active power output sequence, sampling points are taken at equal intervals. The interval duration is set to 1 minute by default in this embodiment. The active power output difference between adjacent sampling points is calculated to form the active power output change sequence, with the unit being kilowatts.

[0061] The terminal voltage change sequence is constructed as follows: the same moving average filtering and equal-interval interpolation processing are applied to the terminal voltage sequence acquired by S110 to obtain the terminal voltage change sequence, in volts. Using a 1-minute interval achieves a balance between capturing photovoltaic power output fluctuation characteristics and suppressing high-frequency measurement noise.

[0062] Linear fitting employs the least squares method, with the change in active power output as the independent variable and the change in terminal voltage as the dependent variable. The slope of the fitted straight line represents the instantaneous voltage sensitivity of the photovoltaic inverter within the effective time period, calculated using the following formula: ; in, For the first Within the first effective time period The instantaneous voltage sensitivity of the photovoltaic inverter, in volts per kilowatt; For the first The sequence of changes in active power output within the effective time period. Each sample value, For the first The mean of all sampled values ​​of the active power output change sequence within a valid time period; For the first The sequence of terminal voltage changes of the i-th photovoltaic inverter within a valid time period. Each sample value, For the first The mean of all sampled values ​​of the terminal voltage change sequence within a valid time period; For the first The total number of sampling points for changes within a valid time period.

[0063] Simultaneous calculation of the fitted Pearson correlation coefficient ,like Below the preset relevant threshold, or If the value is negative, the fitting result for that effective time period corresponding to the inverter is deemed invalid and will not be included in subsequent calculations. When the value is negative, it indicates that the voltage drops as the output increases, which does not conform to the physical law that the voltage rises due to the increase of photovoltaic reverse power. This anomaly is usually caused by load change interference or data anomaly and should be eliminated.

[0064] The preset correlation threshold is the minimum standard for judging whether the linear fitting result has statistical significance; in this embodiment, it is set to 0.5 by default. The rationale for this setting is that when the Pearson correlation coefficient is below 0.5, the linear relationship between the change in active power output and the change in terminal voltage is weak, indicating that the change in terminal voltage within this effective time period may be significantly interfered with by other factors such as load fluctuations and measurement noise. The confidence level of the instantaneous voltage sensitivity obtained from the fitting is insufficient, and it is not suitable for inclusion in the comprehensive voltage sensitivity calculation. In specific implementations, the preset correlation threshold can also be adjusted to other values ​​based on the actual data quality and fitting effect of the transformer area.

[0065] Instantaneous voltage sensitivity The physical meaning of instantaneous voltage sensitivity is: the magnitude of the change in the terminal voltage of a photovoltaic inverter caused by a change in unit active power output within the effective time period. The higher the instantaneous voltage sensitivity, the greater the equivalent electrical impedance between the grid connection point of the photovoltaic inverter and the low-voltage bus of the distribution transformer, and the more significant the voltage rise at the photovoltaic inverter node caused by the same increase in active power output.

[0066] Calculate the total active power fluctuation amplitude and total load fluctuation amplitude of photovoltaic power generation in each effective time period within the target transformer area corresponding to the phase.

[0067] The total active power fluctuation range of photovoltaics is the difference between the maximum and minimum values ​​after summing the active power output of each photovoltaic inverter in the corresponding phase within the effective time period at the same time.

[0068] The total load fluctuation range is the difference between the maximum and minimum values ​​of the total load of the target distribution area within the same time period. The total load of the distribution area is obtained by subtracting the energy storage charging and discharging power collected by S140 from the active power of the distribution transformer collected by S130 and summing the active power output of each photovoltaic inverter.

[0069] The ratio of the total active power fluctuation of photovoltaic power to the total load fluctuation is used as the basic weight of the corresponding effective time period. The larger the value, the stronger the dominance of photovoltaic power output change within the corresponding effective time period, the smaller the interference of load fluctuation on sensitivity fitting, and the higher the data quality.

[0070] The basic weights for each effective time period are normalized so that the sum of all weights is 1, thus obtaining the first... Normalized weights for each time period The instantaneous voltage sensitivity of each photovoltaic inverter during each effective time period is weighted and averaged using normalized weights to obtain the comprehensive voltage sensitivity of the photovoltaic inverter: ; in, For the first The comprehensive voltage sensitivity of a photovoltaic inverter is expressed in volts per kilowatt. It comprehensively reflects the equivalent electrical impedance level between the grid connection point of the photovoltaic inverter and the low-voltage busbar of the distribution transformer. The higher the value, the closer the equivalent electrical position is to the end of the line. This represents the total number of effective time periods for the photovoltaic inverter, i.e., the number of time periods retained after fluctuation threshold screening, fitting correlation coefficient verification, and negative value removal. For the first Normalized weights for each effective time period; For the first The instantaneous voltage sensitivity of the inverter within a valid time period is expressed in volts per kilowatt.

[0071] S230. Based on the comprehensive voltage sensitivity, sort the photovoltaic inverters in each phase to generate an equivalent electrical position sequence that characterizes the relative relationship of voltage sensitivity.

[0072] Based on comprehensive voltage sensitivity, the photovoltaic inverters within each phase of the target distribution area are sorted to generate an equivalent electrical position sequence for each phase from the beginning (low-voltage busbar side of the distribution transformer) to the end (far end of the line). The sorting rules are as follows: The lower the overall voltage sensitivity, the closer the equivalent electrical position is to the beginning; the higher the overall voltage sensitivity, the closer the equivalent electrical position is to the end. When the difference in overall voltage sensitivity of multiple photovoltaic inverters is less than the preset differentiation threshold, they are considered to have parallel electrical positions and belong to the same position level in the equivalent electrical position sequence.

[0073] The preset distinction threshold is uniformly set at 5% of the average comprehensive voltage sensitivity of all photovoltaic inverters in the corresponding phase. The basis for this setting is as follows: considering the inherent error sources of instantaneous sensitivity fitting in S220, including terminal voltage measurement noise, small load fluctuation interference, and sliding filter residuals, the random fluctuation amplitude of comprehensive voltage sensitivity under normal operating conditions is about 3% to 5% of the average value; taking 5% as the preset distinction threshold can cover most random estimation deviations, avoid misjudgment of sorting caused by measurement noise and data fluctuations, avoid frequent jumps in the equivalent position sequence with the rolling window update, and ensure the stability of the voltage regulation strategy operation.

[0074] Coupling matrix module: Combining the equivalent electrical position sequence, it quantifies the coupling relationship between the output change and voltage response of the photovoltaic inverter, calls the historical operating data of the photovoltaic inverter to calculate the reference coupling strength, and introduces the real-time photovoltaic output factor, the area load level factor and the node voltage margin factor for dynamic correction, generating a dynamic coupling coefficient matrix.

[0075] In this embodiment, as Figure 3 As shown, by combining the equivalent electrical position sequence, the coupling relationship between the output change and voltage response of the photovoltaic inverter is quantified. The baseline coupling strength is calculated by calling the historical operating data of the photovoltaic inverter, and the real-time photovoltaic output factor, the area load level factor, and the node voltage margin factor are introduced for dynamic correction, generating a dynamic coupling coefficient matrix including the following specific contents: S310. Determine the relative positional relationship between photovoltaic inverter nodes based on the equivalent electrical position sequence, and construct a coupling path characterizing the voltage coupling relationship between photovoltaic inverter nodes in conjunction with the photovoltaic backflow direction.

[0076] Obtain the equivalent electrical position sequence generated by S230. In the equivalent electrical position sequence, the photovoltaic inverters at the end are collectively referred to as line end-side nodes, and the photovoltaic inverters at the beginning are collectively referred to as distribution transformer beginning-side nodes. The photovoltaic inverters are arranged in order from the beginning to the end.

[0077] Based on the physical law of photovoltaic reverse power flow from the end of the line to the distribution transformer, for any end node of the line, when the active power output of the end node of the line increases, the reverse power flows along the line towards the distribution transformer, the current flowing through the line increases, the line voltage drop increases accordingly, and the voltage of all distribution transformer head nodes on the path from the end node of the line to the distribution transformer rises synchronously; when the output decreases, the line current decreases, the line voltage drop decreases, and the voltage of each distribution transformer head node decreases synchronously.

[0078] Similarly, changes in the output of the transformer's head-end nodes will also affect the voltage of the line's end-end nodes through the common line impedance: when the output of the transformer's head-end nodes increases, the injected current also flows through the common line impedance, causing the voltage of all line's end-end nodes that pass through the common line impedance to rise synchronously.

[0079] Based on this, a coupling path between output changes and voltage responses of photovoltaic inverter nodes is constructed. The specific construction rule is as follows: traverse any two photovoltaic inverter nodes within the corresponding phase. If one photovoltaic inverter node is the transformer head-end node of the other photovoltaic inverter node, that is, it is closer to the head end in the equivalent electrical position sequence, then there is a common electrical impedance between the two photovoltaic inverter nodes. The output changes of the two will affect each other's terminal voltage through the common impedance.

[0080] Photovoltaic inverter node pairs that meet the above conditions are recorded as a coupling path. The coupling path is a bidirectional path, where a change in the output of any photovoltaic inverter node will cause a voltage response at the other photovoltaic inverter node. The set of all coupling paths constitutes the coupling path reference framework for the corresponding phase, covering all node pairs with a common electrical impedance within the target area, and is used to quantify the correlation strength of voltage changes of each photovoltaic inverter node under the output ramp-up scenario.

[0081] S320. Based on the historical operating data of photovoltaic inverters, the events of only one photovoltaic inverter disconnecting from the grid and only one photovoltaic inverter effectively derating within the same phase are used as calibration samples. The output change amplitude and the terminal voltage change of the affected photovoltaic inverters are obtained according to the calibration samples. The weights are dynamically allocated according to the event types of disconnection events and effective derating events, and then the weighted sum is obtained to obtain the reference coupling strength of the coupling path.

[0082] Based on historical operating data of photovoltaic inverters, the action time of each grid disconnection event and effective derating event and the corresponding output change of the photovoltaic inverter are obtained.

[0083] Valid derating events are those whose output change is greater than the preset minimum derating threshold. They are used to supplement calibration samples when there are insufficient off-grid event samples. The preset minimum derating threshold is set according to a preset ratio of the inverter's rated power. In this embodiment, it is 10% of the rated power to eliminate derating events with too small an amplitude and insufficient signal-to-noise ratio.

[0084] Based on the action times of each grid disconnection event and effective derating event, the corresponding action time is located in the terminal voltage sequence collected by S110. For each event, an independence check is first performed: taking the action time as the center, the active power output sequence of all other photovoltaic inverters in the same phase within a 30-second time window before and after it is extracted, and the total active power fluctuation amplitude of the other photovoltaic inverters is calculated. If the total active power fluctuation amplitude is greater than the preset stability threshold (in this embodiment, the preset stability threshold is 5% of the total rated capacity of the corresponding phase), it is determined that the grid disconnection event and effective derating event have multiple machines causing common disturbances, and they are removed and not included in the subsequent calibration calculation. Events that pass the independence check are valid calibration samples.

[0085] Among them, the total rated capacity of each phase refers to the sum of the rated power of all connected photovoltaic inverters in the corresponding phase, in kilowatts. The corresponding data is obtained from the grid connection file of the target area or the sum of the rated power of each photovoltaic inverter. The 30-second window for independence verification is set based on the following principles: it can cover the voltage transient transition process after the inverter protection action, ensure the acquisition of steady-state voltage values, and avoid the introduction of slow load fluctuation interference due to an excessively long window.

[0086] Using the photovoltaic inverter that has experienced a grid disconnection event or an effective derating event as a reference, all distribution transformer head-end nodes whose sorting positions are closer to the head end than the reference inverter are determined according to the equivalent electrical position sequence. Extract the voltage change of each distribution transformer's head-end node during this grid disconnection event or effective derating event: take the average terminal voltage within 30 seconds before the action time as the steady-state voltage before the action, and take the average terminal voltage within 30 seconds after the action time as the steady-state voltage after the action. The absolute value of the difference between the two is the voltage change of the distribution transformer's head-end node during this grid disconnection event or effective derating event.

[0087] The cumulative number of valid network disconnection events that have passed independence verification in the target area throughout history. ,according to The calculation weights for offline events and effective rate reduction events are dynamically allocated. In this embodiment, the weights for offline events are... The effective reduction event weight is determined by the following formula: 1 minus the offline event weight. : ; in, The preset reference sample size is used in this embodiment. This means that when the cumulative number of offline events reaches 5, the weight of the offline event reaches 0.5; when the cumulative number of offline events is small, the effective de-rated events bear the main calculation weight to ensure that the benchmark coupling strength can still be calculated in the stage of scarce samples; as offline events gradually accumulate, the weight of offline events gradually increases, and the weight of effective de-rated events decreases accordingly.

[0088] For each coupling path, all disconnection events and effective derating events on the coupling path are assigned corresponding weights according to their respective types. The output change amplitude and terminal voltage change of each disconnection event and effective derating event are weighted and summed separately. Using the total output change amplitude after weighted summing as the denominator and the total terminal voltage change after weighted summing as the numerator, the average change amplitude of the terminal voltage at the distribution transformer head-end node caused by a unit change in active power output is calculated as the reference coupling strength of the corresponding coupling path.

[0089] The reference coupling strength reflects the degree of influence of the change in unit active power output of one node in the coupling path on the terminal voltage of the other node under average operating conditions. It is the reference value for S340 to construct the dynamic coupling coefficient matrix.

[0090] S330, the real-time photovoltaic output factor is obtained by summing and normalizing the active power output of each photovoltaic inverter at the current moment in the photovoltaic inverter operation data; the distribution area load level factor is obtained by subtracting the real-time total photovoltaic output and energy storage charging and discharging power from the active power output of the distribution transformer outlet in the distribution transformer electrical operation data, and then calculating the ratio of the current target distribution area total load to the historical target distribution area maximum load value; the node voltage margin factor is calculated by the difference between the current terminal voltage of each photovoltaic inverter and the protection action threshold.

[0091] From the active power output sequence of the photovoltaic inverter operating data collected by S110, the active power output value of each photovoltaic inverter at the current moment is extracted and summed to obtain the real-time total photovoltaic output. The unit is kilowatts. Real-time photovoltaic output factor is generated based on the real-time total photovoltaic output. The calculation function is: ; in, The total rated capacity of all photovoltaic inverters connected to the corresponding phase is the sum of the rated power of all connected photovoltaic inverters within the corresponding phase, expressed in kilowatts.

[0092] Real-time photovoltaic power output factor The value ranges from [0,1]. The larger the value, the higher the current photovoltaic output level, and the more significant the voltage response effect caused by the output change.

[0093] From the active power sequence of the electrical operation data of the distribution transformer collected by S130, the active power value corresponding to the current moment is extracted. This value is the total active power sent out by the target transformer area, in kilowatts. The power direction is positive when the distribution transformer feeds into the target transformer area and negative when the target transformer area feeds back into the distribution transformer area.

[0094] The current charging and discharging power of the energy storage is collected by S140, and the unit is kilowatts. Positive values ​​indicate energy storage charging, and negative values ​​indicate energy storage discharging. The total load of the target area is obtained by subtracting the real-time total photovoltaic output from the total active power transmitted out of the target area, and then subtracting the current charging and discharging power of the energy storage. The unit is kilowatts. The load level factor for the target distribution area is generated based on the total load of the target distribution area. The calculation function is: ; in The target transformer area's historical maximum load value is measured in kilowatts and is taken from the transformer area's distribution transformer archives or the power grid marketing system, which record the target transformer area's historical annual maximum load data. Transformer Area Load Level Factor The value ranges from [0,1]. The larger the value, the lower the current load level, the weaker the absorption capacity of the transformer area, and the more significant the voltage rise effect caused by changes in photovoltaic output.

[0095] Extract the terminal voltage value of each photovoltaic inverter at the current moment from the terminal voltage sequence of the photovoltaic inverter operation data collected by S110. The unit is volts. The node voltage margin factor is generated based on the difference between the voltage at each inverter terminal and the protection action threshold. The calculation function is: ; in, For the first The current voltage margin of the photovoltaic inverter. The larger the value, the more voltage headroom the corresponding photovoltaic inverter has before protection trips.

[0096] The protection action threshold, in volts, is the voltage setting value for the inverter's local overvoltage protection action. It is written into the firmware by the inverter manufacturer when the equipment leaves the factory, or set by the operation and maintenance personnel according to the grid connection requirements. It is an inherent parameter of the inverter's local protection logic and is obtained from the inverter equipment parameter ledger or grid connection acceptance data.

[0097] As a safety voltage margin threshold, this embodiment defaults to 5V. When the voltage margin is greater than or equal to 5V, the node voltage margin factor is 0, indicating that the photovoltaic inverter node has sufficient tolerance to voltage rise under the current operating conditions and does not require special attention. When the voltage margin drops below 5V, the node voltage margin factor increases linearly. When the voltage margin drops to 0 or below, the node voltage margin factor takes the maximum value of 1, indicating that the photovoltaic inverter node has reached or exceeded the protection threshold and is extremely sensitive to any voltage rise.

[0098] The node voltage margin factor ranges from [0,1]. The larger the value, the closer the current voltage of the photovoltaic inverter node is to the protection threshold, and the weaker its ability to withstand voltage rise.

[0099] Real-time photovoltaic power output factor District load level factor and node voltage margin factor The three real-time factors in S340 are used to dynamically correct the baseline coupling strength of each coupling path. They are all dynamically refreshed at a frequency of seconds as the running data is updated.

[0100] S340. For each coupling path, the inherent attenuation coefficient is determined based on the comprehensive voltage sensitivity. According to the equivalent position difference of the photovoltaic inverter nodes in the equivalent electrical position sequence, the weights of the real-time photovoltaic output factor, the substation load level factor, and the node voltage margin factor are dynamically allocated and weighted to obtain the comprehensive correction coefficient. The reference coupling strength is corrected by the inherent attenuation coefficient and the comprehensive correction coefficient to generate the dynamic coupling coefficient, forming a dynamic coupling coefficient matrix.

[0101] For each coupling path, the corresponding inherent attenuation coefficient is first calculated. The inherent attenuation coefficient reflects the natural attenuation of the voltage coupling effect between photovoltaic inverter nodes with electrical distance, and its value is directly determined by the comprehensive voltage sensitivity of the upstream and downstream nodes of the coupling path: take the comprehensive voltage sensitivity of the distribution transformer head-end node and divide it by the comprehensive voltage sensitivity of the line end-end node, and the result is the inherent attenuation coefficient of the coupling path.

[0102] Since the overall voltage sensitivity of nodes closer to the head is lower, the calculation results are in the range of 0 to 1: the smaller the difference in overall voltage sensitivity between two nodes, the closer the calculation result is to 1, indicating that the equivalent electrical distance between nodes is closer and the attenuation of voltage coupling effect is weaker; the larger the difference, the smaller the calculation result, indicating that the equivalent electrical distance between nodes is farther and the attenuation of voltage coupling effect is more significant.

[0103] Secondly, based on the equivalent position difference between upstream and downstream nodes in the coupling path, the dynamic weights of the three real-time factors generated by S330 are determined.

[0104] In this embodiment, the default dynamic weight allocation rule is: based on the equivalent position difference of the coupling path. The maximum equivalent position difference between the phases The ratio is the adjustment benchmark, where the equivalent position difference The difference between the position sequence numbers of the end-of-line node and the beginning-of-line node in the equivalent electrical position sequence, representing the maximum equivalent position difference. This represents the positional difference between the last photovoltaic inverter node and the first photovoltaic inverter node of the corresponding phase.

[0105] In this embodiment, the weight of the node voltage margin factor is set to be... The equivalent location difference decreases linearly within the range of 0.7 to 0.2; the remaining weight is evenly distributed by the real-time photovoltaic output factor and the area load level factor, i.e.: The sum of the weights of the three is 1.

[0106] The rationale for this setting is as follows: When the equivalent position difference is small, the electrical distance between the two photovoltaic inverter nodes is close, and the voltage coupling effect is direct and strong. The voltage tolerance margin of the photovoltaic inverter node itself is the core factor determining the degree of coupling influence. Therefore, the voltage margin factor is given a higher initial weight. When the equivalent position difference is large, the electrical range traversed by the path is wide, and the voltage coupling effect is more significantly affected by the overall photovoltaic back-feeding level and load absorption capacity of the distribution area. Therefore, the weights of the two types of macroscopic operating condition factors are gradually increased to match the dominant influence law under different electrical distances.

[0107] In practice, the upper and lower limits of the weights of 0.7 and 0.2 can be adjusted according to the actual topology characteristics of the transformer area and the voltage regulation control requirements. The allocation ratio of the two types of macro factors can also be modified as needed to adapt to the operating characteristics of transformer areas with different installed capacity and line structures. Next, the real-time photovoltaic output factor, the substation load level factor, and the node voltage margin factor are weighted and summed according to the above dynamic weights to obtain the comprehensive correction coefficient of the coupling path. : ; in, The real-time photovoltaic output factor, transformer area load level factor, and the first generation generated by S330 are respectively... i The voltage margin factor at each node has a value range of [0,1]. These are the dynamic weights corresponding to the real-time photovoltaic output factor, the transformer area load level factor, and the node voltage margin factor, respectively, and they all satisfy the condition that the sum of the three is one.

[0108] Comprehensive correction coefficient Similarly, the value is between 0 and 1. The physical meaning of the comprehensive correction coefficient is: under the combined effect of the current photovoltaic power output, the absorption capacity of the distribution area and the voltage withstand capability of the node, the correction coefficient of the voltage coupling strength of the coupling path relative to the reference operating condition. The closer it is to 1, the more significant the voltage rise effect of the output change under the current operating condition; The closer it is to 0, the stronger the current operating condition's suppression of voltage rise.

[0109] Finally, the reference coupling strength of the coupling path obtained from S320 is... Multiplying by the inherent attenuation coefficient and the comprehensive correction coefficient, we obtain the dynamic coupling coefficient of the coupling path at the current moment. : ; in, The reference coupling strength is expressed in volts per kilowatt. The inherent attenuation coefficient; This is a comprehensive correction factor. Dynamic coupling factor. This indicates the magnitude of the impact of each unit change in active power output at the head-end node of the distribution transformer on the voltage at the end-end node of the line under the current actual operating conditions.

[0110] The dynamic coupling coefficients of all coupling paths constitute a dynamic coupling coefficient matrix. As the operating data is updated, the three factors of S330 are refreshed in real time, the dynamic weights and comprehensive correction coefficients of each path are updated synchronously, and the dynamic coupling coefficient matrix is ​​also refreshed dynamically at a frequency of seconds, dynamically reflecting the degree of coupling influence of the output changes of each photovoltaic inverter on the voltage of other nodes under the current operating conditions.

[0111] Voltage regulation configuration module: Based on the dynamic coupling coefficient matrix, identify coupling vulnerability points, call adjustable resource status data to evaluate the comprehensive protection efficiency of each adjustable resource for coupling vulnerability points, and generate a source-grid-load-storage coordinated voltage regulation scheme.

[0112] In this embodiment, based on the dynamic coupling coefficient matrix, vulnerable points are identified. Adjustable resource status data is used to evaluate the overall protection efficiency of each adjustable resource for these vulnerable points, generating a source-grid-load-storage coordinated voltage regulation scheme, including the following specific contents: S410. Combining the dynamic coupling coefficient matrix with the photovoltaic inverter operating data, predict the voltage rise of each photovoltaic inverter node in the same phase when the output is ramped up, and mark the photovoltaic inverter node whose voltage rise exceeds its own voltage margin as a coupling vulnerability point.

[0113] From the dynamic coupling coefficient matrix generated by S340, the dynamic coupling coefficients between photovoltaic inverters in the same phase are extracted; from the terminal voltage sequence collected by S110, the terminal voltage of each photovoltaic inverter at the current moment is extracted, and the difference between the protection action threshold defined by S330 and the current terminal voltage is used as the current voltage margin of each photovoltaic inverter node.

[0114] For each photovoltaic inverter, all other photovoltaic inverters in the same phase are treated as associated nodes. The current active power output ratio of each associated node is weighted and summed with the corresponding dynamic coupling coefficient, and then multiplied by the preset unit ramp-up rate to obtain the expected increase in terminal voltage of the photovoltaic inverter under the overall photovoltaic output ramp-up scenario in the target area.

[0115] Among them, the current active power output ratio is the proportion of the current active power output of the corresponding associated node to the total active power output of all associated nodes, reflecting the contribution weight of each photovoltaic inverter node to the overall coupling boosting effect. The preset unit ramp-up margin is the typical ramp-up increment of photovoltaic output per unit time in a preset area. In this embodiment, it is taken as 10% of the rated installed capacity of the same phase, which is used to quantify the uniform output change range under the output ramp-up scenario. In specific implementation, the preset unit ramp-up margin can be adjusted according to the statistical data of the actual photovoltaic output ramp-up rate of the area: for example, in cloudy areas with drastic fluctuations in photovoltaic output, it can be taken as 15% of the total rated capacity of the same phase to cover the rapid ramp-up scenario; in sunny areas with gentle changes in photovoltaic output, it can be taken as 5% of the total rated capacity of the same phase to reduce conservatism.

[0116] Photovoltaic inverters whose expected voltage rise exceeds their current voltage margin are marked as coupling vulnerable points. Coupling vulnerable points are those nodes that are significantly boosted by the output of other photovoltaic inverter nodes, making them prone to triggering overvoltage protection first and causing a chain reaction of grid disconnection.

[0117] S420. For coupled vulnerable points, on the equivalent electrical position sequence, the set of photovoltaic inverter nodes constituting the electrical neighborhood is selected based on the dynamic coupling coefficient. Within the electrical neighborhood, the comprehensive protection efficiency is obtained by multiplying the dynamic coupling coefficient of the adjustable resource to the coupled vulnerable point, the real-time availability determined by the adjustable resource status data, and the voltage regulation priority coefficient determined by the equivalent position difference. Adjustable resources are matched according to the comprehensive protection efficiency to generate a collaborative protection scheme.

[0118] On the equivalent electrical location sequence, the candidate range is defined with the location of the coupling vulnerability point as the center. Photovoltaic inverter nodes with a dynamic coupling coefficient greater than the preset coupling threshold are selected to form the electrical neighborhood of the coupling vulnerability point. Only adjustable resources within the electrical neighborhood are included in the candidates, and distant resources with negligible coupling effects are excluded.

[0119] The preset coupling threshold is the critical value for determining whether the electrical coupling between photovoltaic inverter nodes has voltage regulation value. In this embodiment, it is taken as 15% of the maximum dynamic coupling coefficient of the corresponding phase by default. When it is lower than the coupling threshold, the coupling effect is weak and there is no need for regulation.

[0120] Use the adjustable resource status data collected by S140 to calculate the availability of each type of resource: Energy storage availability: Energy storage is used to absorb excess active power from photovoltaics, and only charging regulation is effective. Availability equals the remaining rechargeable capacity of the energy storage divided by the rated charging capacity of the energy storage; the remaining rechargeable capacity is calculated by subtracting the current state of charge from the upper limit of the energy storage's charging state of charge, combined with the rated capacity, while deducting the currently occupied charging power. The value ranges from 0 to 1, and the larger the value, the more sufficient the regulation margin for absorbing excess photovoltaic power.

[0121] Adjustable load availability: Adjustable loads rely on increasing the power absorption capacity to suppress overvoltage, and only have adjustment capabilities during operation. When the load is shut down, the availability is directly set to 0; when the load is running, the availability is equal to the current adjustable load margin divided by the maximum adjustable load power. The adjustable margin is the upper limit adjustable power of the load minus the current actual power consumption, with a value range of 0 to 1. The larger the value, the more power can be increased.

[0122] For each adjustable resource in the electrical neighborhood, the voltage regulation priority coefficient is determined based on the equivalent position difference between the adjustable resource and the coupled vulnerable point; the dynamic coupling coefficient, availability and voltage regulation priority coefficient of the adjustable resource access point to the coupled vulnerable point are multiplied together to obtain the comprehensive protection efficiency of the adjustable resource to the coupled vulnerable point.

[0123] All coupled vulnerable points are sorted in ascending order of voltage margin. When a coupled vulnerable point is located at the intersection of multiple coupling paths, the overall ranking of the coupled vulnerable point is moved up one position. According to the ranking results, the adjustable resources with the highest comprehensive protection efficiency are matched to each coupled vulnerable point in turn. The matched resources will not participate in the subsequent allocation to avoid duplicate scheduling.

[0124] S430: The collaborative protection scheme is preset into the local controller of the matched adjustable resource. When the voltage at the coupled vulnerable point reaches the trigger threshold, the adjustable resource will automatically perform voltage regulation protection locally.

[0125] The matching results between coupling vulnerabilities and adjustable resources are transformed into response logic and pre-set in the local controller of the matched adjustable resources.

[0126] For each matching relationship, the protection action threshold voltage of the coupled vulnerable point is used as a reference, and a preset proportion of the protection threshold voltage is set as the trigger threshold. In this embodiment, 95% of the protection threshold voltage is used by default. When the voltage at the vulnerable point rises and exceeds the trigger threshold, the matched adjustable resource immediately performs an adjustment action according to its maximum adjustable capability.

[0127] The aforementioned response logic is pre-configured into the local controller of the matched adjustable resource. The local controller continuously monitors the terminal voltage data of the corresponding vulnerable point. When the terminal voltage reaches the trigger threshold, it immediately executes the adjustment action without waiting for central dispatch instructions, thus realizing local autonomous voltage regulation protection for coupled vulnerable points in power ramping scenarios.

[0128] Embodiments of the present invention also provide an electronic device, including a memory, a processor, and a communication bus; the memory and the processor are connected via the communication bus. The memory stores a source-grid-load-storage integrated collaborative optimization and control system, which can be loaded by the processor and executed as provided in the above embodiments.

[0129] The memory can be used to store instructions, programs, code, code sets, or instruction sets. The memory may include a program storage area and a data storage area. The program storage area may store instructions for implementing an operating system, instructions for at least one function, and instructions for implementing the integrated source-grid-load-storage collaborative optimization and control system provided in the above embodiments. The data storage area may store data involved in the integrated source-grid-load-storage collaborative optimization and control system provided in the above embodiments.

[0130] A processor may include one or more processing cores. The processor executes instructions, programs, code sets, or instruction sets stored in memory, and calls data stored in memory to perform various functions and process data as described in this application. The processor may be at least one of a specific application-specific integrated circuit, a digital signal processor, a digital signal processing device, a programmable logic device, a field-programmable gate array, a central processing unit, a controller, a microcontroller, and a microprocessor. It is understood that, for different devices, the electronic devices used to implement the above-described processor functions may also be other types, and the embodiments of this application do not specifically limit the specific implementation.

[0131] A communication bus may include a pathway for transmitting information between the aforementioned components. The communication bus can be a PCI bus or an EISA bus, etc. Communication buses can be categorized into address buses, data buses, control buses, etc.

[0132] This application provides a computer-readable storage medium storing a computer program that can be loaded by a processor and executed as described in the above embodiments, based on the integrated source-grid-load-storage collaborative optimization and control system.

[0133] In this embodiment, a computer-readable storage medium can be a tangible device that holds and stores instructions used by an instruction execution device. A computer-readable storage medium can be, but is not limited to, an electrical storage device, a magnetic storage device, an optical storage device, an electromagnetic storage device, a semiconductor storage device, or any combination thereof. Specifically, a computer-readable storage medium can be a portable computer disk, a hard disk, a USB flash drive, a random access memory, a read-only memory, an erasable programmable read-only memory, a podium random access memory, a portable compressed disk read-only memory, a digital multifunction disk, a memory stick, a floppy disk, an optical disk, a magnetic disk, a mechanical encoding device, or any combination thereof.

[0134] The terms “comprising,” “including,” or any other variations thereof are intended to cover non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements includes not only those elements but also other elements not expressly listed, or elements inherent to such process, method, article, or apparatus.

[0135] The above description is merely a preferred embodiment of this application and an explanation of the technical principles employed. Those skilled in the art should understand that the scope of this application is not limited to technical solutions formed by specific combinations of the above-described technical features, but should also cover other technical solutions formed by arbitrary combinations of the above-described technical features or their equivalents without departing from the foregoing application concept. For example, technical solutions formed by substituting the above-described features with (but not limited to) technical features with similar functions claimed in this application.

Claims

1. An integrated source-grid-load-storage collaborative optimization and control system, characterized in that, Includes the following modules: Multi-source acquisition module: Simultaneously acquires photovoltaic inverter operation data, distribution transformer electrical operation data, and adjustable resource status data for the target area; Location sequencing module: Identifies the relative voltage sensitivity relationship through photovoltaic inverter operating data, and determines the equivalent electrical position sequence of photovoltaic inverters in each phase of the target area; Coupling matrix module: Combining the equivalent electrical position sequence, it quantifies the coupling relationship between the output change and voltage response of the photovoltaic inverter, calls the historical operating data of the photovoltaic inverter to calculate the reference coupling strength, and introduces the real-time photovoltaic output factor, the area load level factor and the node voltage margin factor for dynamic correction to generate a dynamic coupling coefficient matrix. Voltage regulation configuration module: Based on the dynamic coupling coefficient matrix, it identifies coupling vulnerability points, calls adjustable resource status data to evaluate the comprehensive protection efficiency of each adjustable resource for coupling vulnerability points, and generates a source-grid-load-storage coordinated voltage regulation scheme.

2. The integrated source-grid-load-storage collaborative optimization and control system according to claim 1, characterized in that, Synchronously collect photovoltaic inverter operation data, distribution transformer electrical operation data, and adjustable resource status data for the target area, including the following steps: Collect operating data and phase affiliation information of photovoltaic inverters within the target area; Collect historical operating data of photovoltaic inverters that have triggered protection actions within the target area, including the action time and output change magnitude of each derating event and grid disconnection event; Collect electrical operation data of the target transformer substation area by phase; Collect adjustable resource status data reflecting adjustable load storage resources within the target area.

3. The integrated source-grid-load-storage collaborative optimization and control system according to claim 2, characterized in that, By identifying the relative voltage sensitivity relationships through photovoltaic inverter operating data, the equivalent electrical position sequence of photovoltaic inverters in each phase of the target distribution area is determined, including the following steps: Using phase attribution information as the classification standard, the active power output fluctuation amplitude is calculated using photovoltaic inverter operating data, and the effective time period is selected based on the active power output fluctuation amplitude. For the selected effective time periods, the photovoltaic inverter operation data is fitted to obtain the instantaneous voltage sensitivity, which characterizes the voltage change caused by a unit change in active power. Then, the active power output fluctuation amplitude of each effective time period is used as the weight to synthesize the comprehensive voltage sensitivity. Based on the comprehensive voltage sensitivity, the photovoltaic inverters within each phase are sorted to generate an equivalent electrical position sequence that characterizes the relative relationship of voltage sensitivity.

4. The integrated source-grid-load-storage collaborative optimization and control system according to claim 3, characterized in that, By combining the equivalent electrical position sequence, the coupling relationship between the output change and voltage response of the photovoltaic inverter is quantified, and the baseline coupling strength is calculated by calling the historical operating data of the photovoltaic inverter, including the following steps: The relative positional relationship between photovoltaic inverter nodes is determined based on the equivalent electrical position sequence, and a coupling path characterizing the voltage coupling relationship between photovoltaic inverter nodes is constructed in combination with the photovoltaic backflow direction. Based on historical operating data of photovoltaic inverters, the calibration samples are the events in which only a single photovoltaic inverter in the same phase is disconnected from the grid and the events in which only a single photovoltaic inverter is effectively dated. The output change amplitude and the terminal voltage change of the affected photovoltaic inverter are obtained according to the calibration sample. The weights are dynamically allocated according to the event type of the disconnection event and the effective derating event, and then the weighted sum is obtained to obtain the reference coupling strength of the coupling path.

5. The integrated source-grid-load-storage collaborative optimization and control system according to claim 4, characterized in that, The system incorporates real-time photovoltaic output factor, transformer area load level factor, and node voltage margin factor for dynamic correction, generating a dynamic coupling coefficient matrix. This process includes the following steps: The real-time photovoltaic output factor is obtained by summing and normalizing the active power output of each photovoltaic inverter in the photovoltaic inverter operation data at the current moment; the distribution area load level factor is obtained by subtracting the real-time total photovoltaic output and energy storage charging and discharging power from the active power output of the distribution transformer outlet in the distribution transformer electrical operation data, and then calculating the ratio of the current target distribution area total load to the historical target distribution area maximum load value; the node voltage margin factor is calculated by the difference between the current terminal voltage of each photovoltaic inverter and the protection action threshold. For each coupling path, the inherent attenuation coefficient is determined based on the comprehensive voltage sensitivity. According to the equivalent position difference of the photovoltaic inverter nodes in the equivalent electrical position sequence, the weights of the real-time photovoltaic output factor, the area load level factor, and the node voltage margin factor are dynamically allocated and weighted to obtain the comprehensive correction coefficient. The reference coupling strength is then corrected by the inherent attenuation coefficient and the comprehensive correction coefficient to generate the dynamic coupling coefficient, forming a dynamic coupling coefficient matrix.

6. The integrated source-grid-load-storage collaborative optimization and control system according to claim 5, characterized in that, Identifying coupling vulnerabilities based on the dynamic coupling coefficient matrix includes the following steps: By combining the dynamic coupling coefficient matrix with photovoltaic inverter operating data, the voltage rise of each photovoltaic inverter node in the same phase is predicted during the power output ramp-up. Photovoltaic inverter nodes whose voltage rise exceeds their own voltage margin are marked as coupling weak points.

7. The integrated source-grid-load-storage collaborative optimization and control system according to claim 6, characterized in that, The process involves evaluating the overall protection efficiency of each adjustable resource for coupled vulnerable points by calling adjustable resource status data, and generating a source-grid-load-storage coordinated voltage regulation scheme, including the following steps: For coupled vulnerable points, the set of photovoltaic inverter nodes constituting the electrical neighborhood is selected based on the dynamic coupling coefficient on the equivalent electrical location sequence. Within the electrical neighborhood, the comprehensive protection efficiency is obtained by multiplying the dynamic coupling coefficient of the adjustable resources to the coupled vulnerable points, the real-time availability determined by the adjustable resource status data, and the voltage regulation priority coefficient determined by the equivalent location difference. Adjustable resources are matched according to the comprehensive protection efficiency to generate a collaborative protection scheme. The collaborative protection scheme is pre-configured into the local controller of the matched adjustable resource. When the voltage at the coupled vulnerable point reaches the trigger threshold, the adjustable resource will automatically perform voltage regulation protection locally.

8. An electronic device, comprising: A processor and a memory, wherein the memory stores a computer program that can be called by the processor; characterized in that the processor implements the integrated source-grid-load-storage collaborative optimization and control system as described in any one of claims 1-7 by calling the computer program stored in the memory.

Images