A photovoltaic prefabricated cabin type substation intelligent control system

Abstract

The application discloses a kind of photovoltaic prefabricated cabin type substation intelligent control systems, it is related to photovoltaic control technical field, including, the multiple-source operation data of photovoltaic prefabricated cabin type substation is collected, and is gathered according to same control period, generates cabin station multiple-source data set, electrical topology mapping and cabin space mapping are carried out to cabin station multiple-source data set, generates cabin station topology correlation data set;Based on cabin station topology correlation data set constructs multidimensional operation boundary, based on multidimensional operation boundary, the operation state in cabin station topology correlation data set is carried out boundary margin analysis, generates boundary approximation result, according to boundary approximation result, multidimensional operation boundary is dynamically contracted, and generates contraction boundary set;According to execution feedback, boundary recovery evaluation and update are carried out to contraction boundary set, and output cabin station control result.The application is updated to contraction boundary set, and outputs cabin station control result, improves the reliability and stability of photovoltaic prefabricated cabin type substation operation.

Description

Technical Field

[0001] This invention relates to the field of photovoltaic control technology, and in particular to an intelligent control system for a photovoltaic prefabricated substation. Background Technology

[0002] As an important component of renewable energy, photovoltaic power generation is expanding its grid connection scale with the transformation of the energy structure. It is showing a development trend from decentralized grid connection to centralized voltage boosting, local consumption, energy storage collaboration, and remote operation and maintenance. Against this backdrop, prefabricated modular substations have become key supporting facilities for photovoltaic power plants and park-level new energy power supply and distribution due to their advantages such as modular integration, short on-site construction cycle, small footprint, and strong adaptability to complex outdoor environments. As grid connection standards continue to improve the requirements for voltage fluctuation, reactive power support, power quality, energy storage response, and fault isolation capabilities, the control objectives of photovoltaic prefabricated modular substations are gradually developing from single equipment start-up and shutdown or simple alarm to cross-equipment collaborative control.

[0003] However, existing technologies still have the following shortcomings. The control strategies of photovoltaic prefabricated substations mainly rely on fixed threshold judgment, single device linkage, or independent control of subsystems. Although they can meet the basic monitoring and local control needs, they have problems such as insufficient multi-source state correlation, lack of dynamic adjustment of operating boundaries, lack of unified constraints on control actions, and limited closed-loop feedback update capabilities. There is a lack of unified correlation between different data sources based on electrical circuits, energy storage branches, and the space area inside the substation, which makes it difficult to identify the coupling relationship between photovoltaic power output changes, grid connection quality fluctuations, energy storage regulation capabilities, electrical equipment loads, and the state of the environment inside the substation in a timely manner. Summary of the Invention

[0004] In view of the aforementioned existing problems, the present invention is proposed.

[0005] Therefore, this invention provides an intelligent control system for photovoltaic prefabricated substations to solve the problems of insufficient correlation of multi-source states and lack of unified constraints on control actions.

[0006] To solve the above-mentioned technical problems, the present invention provides the following technical solution:

[0007] This invention provides an intelligent control system for a photovoltaic prefabricated substation, comprising:

[0008] The topology mapping module is used to collect multi-source operation data of photovoltaic prefabricated substations and aggregate them according to the same control cycle to generate a multi-source dataset of the substation. The module performs electrical topology mapping and intra-substation space mapping on the multi-source dataset of the substation to generate a topology association dataset of the substation.

[0009] The margin analysis module is used to construct a multidimensional operating boundary based on the cabin topology association dataset, perform boundary margin analysis on the operating status in the cabin topology association dataset based on the multidimensional operating boundary, generate boundary approximation results, and dynamically shrink the multidimensional operating boundary based on the boundary approximation results to generate a shrinking boundary set.

[0010] The control feasibility module is used to perform control action feasibility analysis on the shrinkage boundary set, determine the control actions that are allowed to be executed, the control actions that are prohibited from being executed, and the protection actions that must be executed within the same control cycle, form the control feasibility domain, and perform multi-objective strategy matching and conflict verification to generate a set of collaborative control strategies.

[0011] The recovery evaluation module is used to distribute the collaborative control strategy set to the corresponding control layer of the photovoltaic prefabricated substation, collect execution feedback, perform boundary recovery evaluation and update the shrinkage boundary set based on the execution feedback, and output the substation control results.

[0012] As a preferred embodiment of the intelligent control system for the photovoltaic prefabricated substation described in this invention, the specific steps for generating the multi-source dataset for the substation are as follows:

[0013] The multi-source operational data includes photovoltaic power output data, grid connection quality data, energy storage operation data, electrical equipment operation data, protection action data, and cabin environment data;

[0014] Read the acquisition time identifier and device source identifier of each data item in the multi-source operation data, and merge the data items whose acquisition time identifiers belong to the same control cycle to form a data set of the same cycle;

[0015] The data items in the same period dataset are categorized and organized according to the equipment source identifier to generate a multi-source dataset for the cabin station.

[0016] As a preferred embodiment of the intelligent control system for photovoltaic prefabricated substations according to the present invention, the specific steps for generating the substation topology association dataset are as follows:

[0017] Based on the equipment source identifier, the data items in the multi-source dataset of the cabin station are associated with equipment ownership to generate an equipment association dataset;

[0018] Perform branch attribution parsing on the electrical connection identifiers in the device association dataset to generate a branch attribution dataset;

[0019] The connection order of the branch attribution dataset is parsed to generate loop connection order data, and the power flow direction is parsed to generate loop power flow direction data. The loop connection order data and loop power flow direction data are aggregated to generate branch connection relationships.

[0020] Write the branch affiliation dataset and branch connection relationship into the device association dataset to generate the electrical topology mapping dataset;

[0021] Based on the equipment installation locations in the electrical topology mapping dataset, the electrical topology mapping dataset is spatially associated with the prefabricated cabin area, environmental monitoring points, and control equipment layout locations of the photovoltaic prefabricated cabin substation to generate a cabin-substation topology association dataset.

[0022] As a preferred embodiment of the intelligent control system for photovoltaic prefabricated substations described in this invention, the specific steps for constructing a multi-dimensional operating boundary based on the substation topology association dataset are as follows.

[0023] The photovoltaic output status, grid-connected operation status, energy storage operation status, electrical equipment operation status, and cabin environment status are extracted from the cabin topology association dataset and aggregated to form cabin status aggregation data.

[0024] The operational constraints of the collected data on the status of the cabin are analyzed to determine the power regulation constraints of the photovoltaic output status, the grid connection quality constraints of the grid-connected operation status, the charging and discharging constraints of the energy storage operation status, the equipment load-bearing constraints of the electrical equipment operation status, and the environmental safety constraints of the cabin environment status, thus forming the basis for boundary construction.

[0025] Based on the power regulation constraints, grid connection quality constraints, charge and discharge constraints, equipment load-bearing constraints, and environmental safety constraints in the boundary construction basis, photovoltaic power regulation boundaries, grid connection quality allowable boundaries, energy storage charge and discharge boundaries, electrical equipment load-bearing boundaries, and cabin environmental safety boundaries are constructed respectively, forming multi-dimensional operating boundaries.

[0026] As a preferred embodiment of the intelligent control system for photovoltaic prefabricated substations described in this invention, the specific steps for generating the boundary approximation result are as follows:

[0027] The state of the cabin topology association dataset is classified according to the boundary type in the multidimensional operation boundary to generate a classified operation state set. The classified operation state set is then matched with the multidimensional operation boundary to generate the boundary matching result.

[0028] The boundary matching results are used to calculate the state boundary difference to determine the remaining adjustment margin, remaining load margin, and remaining safety margin of various operating states relative to the corresponding boundaries, thus forming boundary margin data.

[0029] The boundary margin data is divided into normal margin state, contraction margin state and protection margin state. The normal margin state is removed, and the contraction margin state and protection margin state are located according to the electrical topology relationship and the internal space relationship in the compartment topology association dataset to generate the boundary approximation object set.

[0030] The approaching object set is analyzed to determine the operating state category, electrical circuit, and cabin area to which the boundary approach occurs, and the results are aggregated to generate the boundary approach results.

[0031] In a preferred embodiment of the intelligent control system for the photovoltaic prefabricated substation described in this invention, the specific steps for generating the shrinking boundary set are as follows:

[0032] Based on the operational status category in the boundary approximation results, select the boundary type that needs to be shrunk from the multidimensional operational boundaries, and mark the shrunk object by the electrical circuit and the internal area of ​​the compartment to generate a boundary shrinkage object set;

[0033] Boundary location screening is performed on the boundary shrinkage object set to determine the photovoltaic power regulation boundary, grid connection quality allowable boundary, energy storage charging and discharging boundary, electrical equipment bearing boundary or cabin environment safety boundary that need to be shrunk, and a target shrinkage boundary set is generated.

[0034] By applying first-level and second-level shrinkage processing to the target shrinkage boundary set using the shrinkage margin state and the protection margin state, a hierarchical shrinkage boundary set is generated.

[0035] The hierarchical shrinkage boundary set is merged with the non-shrinkage boundaries in the multidimensional running boundary to generate a shrinkage boundary set.

[0036] As a preferred embodiment of the intelligent control system for photovoltaic prefabricated substations described in this invention, the specific steps for forming the control feasible domain are as follows:

[0037] The feasible action set is screened to determine the action constraints of each controlled object within the same control cycle;

[0038] Using the hierarchical shrinkage boundary set as the constraint benchmark, the action restriction conditions are classified and marked. Control actions that satisfy the hierarchical shrinkage boundary set constraints are marked as allowed control actions, control actions that do not satisfy the hierarchical shrinkage boundary set constraints are marked as prohibited control actions, and protection actions used to restore the hierarchical shrinkage boundary set constraints are marked as mandatory protection actions.

[0039] The allowed control actions, prohibited control actions, and mandatory protection actions are grouped together to form the control feasible domain.

[0040] As a preferred embodiment of the intelligent control system for photovoltaic prefabricated substations described in this invention, the specific steps for generating the collaborative control strategy set are as follows:

[0041] Extract the allowed control actions and the necessary protection actions from the control feasible domain to form a set of candidate control strategies;

[0042] Using prohibited control actions in the feasible control domain as constraints, candidate policies containing prohibited actions are removed from the candidate control policy set to generate an executable control policy set.

[0043] The executable control strategy set is arranged according to the controlled object, action type and execution order. Executable control strategies that are repeatedly executed on the same controlled object, simultaneously executed with opposite actions on the same controlled object, or not executed with priority according to the protection actions that must be executed are removed, thus generating a collaborative control strategy set.

[0044] As a preferred embodiment of the intelligent control system for photovoltaic prefabricated substations described in this invention, the acquisition and execution feedback refers to sending the collaborative control strategy set to the corresponding control layer of the photovoltaic prefabricated substation and acquiring the action status, execution completion status, and abnormal feedback status returned by the corresponding control layer.

[0045] As a preferred embodiment of the intelligent control system for the photovoltaic prefabricated substation described in this invention, the specific steps for outputting the substation control results are as follows:

[0046] Based on the hierarchical shrinkage boundaries and boundary margins of the shrinkage boundary set, the allowable state range of each operating state within the current control cycle is set.

[0047] Based on the execution feedback data, determine whether each operating state has recovered to the allowable state range. For boundary states that have not recovered to the allowable state range and are at the first-level contraction boundary, continue to perform first-level contraction processing. For boundary states that have not recovered to the allowable state range and are at the second-level contraction boundary, continue to perform second-level contraction processing, generating conservative update boundary data. For boundary states that have recovered to the allowable state range, relax the boundary level according to the tiered contraction boundary set, generating recovery update boundary data.

[0048] The shrinkage boundary set is updated based on conservative update boundary data and restored update boundary data, and the station control results are output.

[0049] The beneficial effects of this invention are as follows: By performing spatial correlation, a module topology correlation dataset is generated, achieving the effect of accurately locating each operating data source, facilitating refined control and risk identification; by constructing a multi-dimensional operating boundary based on the module topology correlation dataset, the control accuracy of the photovoltaic prefabricated module substation is improved and the operating risk is reduced; by performing approximation object positioning analysis, boundary approximation results are generated, improving the early warning lead time and reducing the risk of exceeding limits; by performing dynamic contraction, a contracted boundary set is generated, improving the system safety margin and reducing potential risks; by combining and coordinating, a collaborative control strategy set is generated, ensuring the coordinated operation of each control objective and improving the effectiveness of control execution; by updating the contracted boundary set, the module control results are output, improving the reliability and stability of the photovoltaic prefabricated module substation operation. Attached Figure Description

[0050] To more clearly illustrate the technical solutions of the embodiments of the present invention, the drawings used in the following description of the embodiments will be briefly introduced. Obviously, the drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0051] Figure 1 This is a schematic diagram of the intelligent control system for a photovoltaic prefabricated substation.

[0052] Figure 2 A flowchart for generating a module topology association dataset.

[0053] Figure 3 A flowchart for constructing multidimensional operational boundaries.

[0054] Figure 4 A flowchart for forming the controllable feasible region. Detailed Implementation

[0055] To make the above-mentioned objects, features and advantages of the present invention more apparent and understandable, the specific embodiments of the present invention will be described in detail below with reference to the accompanying drawings.

[0056] Many specific details are set forth in the following description in order to provide a full understanding of the invention. However, the invention may also be practiced in other ways different from those described herein, and those skilled in the art can make similar extensions without departing from the spirit of the invention. Therefore, the invention is not limited to the specific embodiments disclosed below.

[0057] Secondly, the term "one embodiment" or "embodiment" as used herein refers to a specific feature, structure, or characteristic that may be included in at least one implementation of the present invention. The phrase "in one embodiment" appearing in different places in this specification does not necessarily refer to the same embodiment, nor is it a single or selective embodiment that is mutually exclusive with other embodiments.

[0058] Reference Figures 1-4 As one embodiment of the present invention, this embodiment provides an intelligent control system for a photovoltaic prefabricated substation, comprising the following steps:

[0059] The topology mapping module is used to collect multi-source operation data of photovoltaic prefabricated substations and aggregate them according to the same control cycle to generate a multi-source dataset of the substation. The module performs electrical topology mapping and intra-substation space mapping on the multi-source dataset of the substation to generate a topology association dataset of the substation.

[0060] Read the acquisition time identifier and equipment source identifier of each data item in the multi-source operation data, and merge the data items whose acquisition time identifiers belong to the same control cycle to form a data set of the same cycle; classify and organize the data items in the data set of the same cycle according to the equipment source identifier to generate a multi-source dataset of the cabin station.

[0061] Furthermore, data acquisition terminals are installed on the photovoltaic array side, inverter side, grid connection point side, energy storage interface side, electrical equipment side, protection device side, and cabin environment monitoring side of the photovoltaic prefabricated substation. String acquisition devices on the photovoltaic array side collect string voltage, string current, and string output power; inverter monitoring devices on the inverter side collect inverter input power, inverter output power, and inverter operating status, forming photovoltaic output data; power quality monitoring devices on the grid connection point side collect grid connection voltage, grid connection current, grid connection frequency, active power, reactive power, power factor, and harmonic status, forming grid connection quality data; and energy storage monitoring devices on the energy storage interface side collect energy storage state of charge, storage... Energy storage operation data is generated by collecting charging power, energy storage discharging power, energy storage interface voltage, energy storage interface current, and energy storage interface alarm status. Electrical equipment operation data is generated by collecting data on the load rate of the step-up transformer, the temperature of the step-up transformer, the temperature of the high-voltage switchgear, the opening and closing status of the high-voltage switchgear, and the energized status of the high-voltage switchgear through electrical equipment monitoring devices installed on the side of the step-up transformer and the high-voltage switchgear. Protection operation data is generated by collecting data on protection alarm status, protection lockout status, protection action status, and fault recording status through protection status acquisition devices installed on the side of the protection device. Cabin environment data is generated by collecting data on cabin temperature, cabin humidity, condensation status, water immersion status, smoke status, and ventilation status through cabin environment monitoring devices installed inside the cabin.

[0062] Photovoltaic output data, grid connection quality data, energy storage operation data, electrical equipment operation data, protection action data, and cabin environment data are treated as multi-source operation data. For each data item in the multi-source operation data, an acquisition time identifier and a device source identifier are written. The acquisition time identifier indicates the time when the data item was generated or uploaded, and the device source identifier indicates the identification information of the acquisition object to which the data item belongs. Data items whose acquisition time identifiers belong to the same control cycle are merged into the same data set to form a data set for the same cycle. The photovoltaic output data, grid connection quality data, energy storage operation data, electrical equipment operation data, protection action data, and cabin environment data in the same cycle data set are then categorized and organized according to the device source identifier, retaining the device source identifier and acquisition time identifier for each data item, thus generating a multi-source dataset for the cabin station.

[0063] Based on the equipment source identifier, the data items in the multi-source dataset of the cabin station are associated with equipment ownership to generate an equipment association dataset; the electrical connection identifiers in the equipment association dataset are parsed to generate a branch ownership dataset.

[0064] Furthermore, the device source identifier carried by each data item is read from the multi-source dataset of the prefabricated photovoltaic substation. The device source identifier includes the data source category, device number, and collection point number. According to the device number in the device source identifier, data items with the same device number are grouped into the same device data group. According to the collection point number in the device source identifier, the collection location of each data item in the same device data group is organized so that the data items in the same device data group have clear data source category, device number, and collection point number, generating a device association dataset. The electrical connection identifier bound to each device number is read from the device association dataset. The electrical connection identifier consists of the branch number, circuit number, and connection end identifier in the primary wiring relationship of the photovoltaic prefabricated substation. The branch number is used to distinguish photovoltaic output branches, inverter branches, energy storage branches, step-up circuits, grid connection circuits, and protection circuits. The circuit number is used to distinguish different electrical circuits under the same type of branch. The connection end identifier is used to indicate the access end position of the device in its electrical circuit.

[0065] According to the branch number in the electrical connection identifier, each device data group in the device association dataset is assigned to the corresponding electrical branch. According to the circuit number in the electrical connection identifier, the device data groups in the same electrical branch are assigned to the corresponding electrical circuit. According to the connection end identifier in the electrical connection identifier, the device data groups in the same electrical circuit are organized by access end to generate the branch affiliation dataset.

[0066] The connection order and power flow direction of the branch attribution dataset are parsed to generate branch connection relationships; the branch attribution dataset and branch connection relationships are written into the device association dataset to generate the electrical topology mapping dataset.

[0067] Furthermore, the device data groups, branch numbers, circuit numbers, and connection terminal identifiers within each electrical circuit are read from the branch attribution dataset. The device data groups within the same electrical circuit are arranged according to the adjacent positions of the connection terminal identifiers in the primary wiring relationship, forming circuit connection sequence data. Based on the power transmission direction from the photovoltaic array to the inverter, from the inverter to the step-up transformer, from the step-up transformer to the high-voltage switchgear, and from the high-voltage switchgear to the grid connection point, the device data groups in the circuit connection sequence data are marked with power inflow and power outflow ends, forming circuit power flow direction data. The circuit connection sequence data and circuit power flow direction data are categorized according to the branch number and circuit number to generate branch connection relationships. The branch number, circuit number, and connection terminal identifier from the branch attribution dataset are written into each device data group in the device association dataset. The circuit connection sequence data and circuit power flow direction data from the branch connection relationships are written into the same device data group, generating an electrical topology mapping dataset.

[0068] Based on the equipment installation locations in the electrical topology mapping dataset, the electrical topology mapping dataset is spatially associated with the prefabricated cabin area, environmental monitoring points, and control equipment layout locations of the photovoltaic prefabricated cabin substation to generate a cabin-substation topology association dataset.

[0069] Furthermore, the installation locations of each equipment data group are read from the electrical topology mapping dataset. These locations include the cabin number, cabin area number, and installation coordinates. Each equipment data group is then assigned to a prefabricated cabin area within the photovoltaic prefabricated substation according to its cabin number and cabin area number. The monitoring point number, monitoring range, and installation coordinates of environmental monitoring points within the photovoltaic prefabricated substation are also read. Equipment data groups located within the same prefabricated cabin area and within the monitoring range are linked to environmental monitoring points to form environmental association data. Finally, the layout locations of control equipment within the photovoltaic prefabricated substation are read. The placement locations include the fan placement location, air conditioning placement location, dehumidification equipment placement location, protection interlocking equipment placement location, and switch control equipment placement location. Based on the intra-cabin area relationship between the control equipment placement location and the equipment installation location, the equipment data group is associated with the control equipment that can act on the same prefabricated cabin area to form control association data. The environmental association data and control association data are written into the electrical topology mapping dataset, so that each equipment data group in the electrical topology mapping dataset simultaneously has the electrical circuit location, prefabricated cabin area, environmental monitoring point, and control equipment placement location, generating the cabin topology association dataset.

[0070] It should be noted that the relationship between the areas within the prefabricated module refers to the relationship of the control equipment layout and the equipment installation location within the same prefabricated module, including the area affiliation, spatial distance, and functional coverage. This is obtained by reading the module number, area number, installation coordinates, and the effective range of the control equipment of the photovoltaic prefabricated module substation. The prefabricated module area refers to the internal space unit within the photovoltaic prefabricated module substation, which is divided according to electrical function, equipment layout, and environmental monitoring range. This includes the photovoltaic inverter area, energy storage interface area, step-up transformer area, high-voltage switchgear area, protection and communication area, and internal environmental regulation area.

[0071] The margin analysis module is used to construct a multidimensional operating boundary based on the cabin topology association dataset, perform boundary margin analysis on the operating status in the cabin topology association dataset based on the multidimensional operating boundary, generate boundary approximation results, and dynamically shrink the multidimensional operating boundary according to the boundary approximation results to generate a shrunken boundary set.

[0072] The photovoltaic output status, grid-connected operation status, energy storage operation status, electrical equipment operation status, and cabin environment status are extracted from the cabin topology association dataset to form cabin status aggregation data. The cabin status aggregation data is then subjected to operation constraint analysis to determine the power regulation constraints of the photovoltaic output status, the grid connection quality constraints of the grid-connected operation status, the charging and discharging constraints of the energy storage operation status, the equipment load-bearing constraints of the electrical equipment operation status, and the environmental safety constraints of the cabin environment status, thus forming the basis for boundary construction.

[0073] Furthermore, the data on the photovoltaic array string voltage, string current, and inverter output power are read from the cabin topology association dataset. These data are then organized according to the branch number corresponding to each string. The average voltage of each string under the same branch is calculated and used as the photovoltaic branch voltage state. The current of each string under the same branch is summed to obtain the photovoltaic branch current state. The product of each string voltage and current is obtained to form the output power of each string. The output power of each string is summed to obtain the total output power of the photovoltaic array. The inverter output power is read as the AC output power of the inverter. The difference between the total output power of the photovoltaic array and the AC output power of the inverter is obtained to form the photovoltaic output conversion state. The output power of each string and the photovoltaic array are then compared. The total output power of the photovoltaic array, the AC output power of the inverter, and the photovoltaic output conversion status are collected according to the photovoltaic output branch number to form a photovoltaic output status set. The grid-connected point voltage, current, frequency, active power, reactive power, power factor, and harmonic index are read from the cabin topology association dataset and organized according to the grid-connected circuit number. The average voltage of each sampling point under the same circuit is taken as the grid-connected voltage status, the summation of the current is taken as the grid-connected current status, the average of the frequencies at each point is taken as the grid-connected frequency status, the summation of the active power and reactive power is taken as the circuit power status, the average of the power factor is taken as the power factor status, and the summation of the harmonic index by the square of the amplitude is taken as the total harmonic status, thus forming a grid-connected operation status set.

[0074] The energy storage charge status, charging / discharging power, voltage, current, and alarm status are read from the station topology association dataset. Organized according to the energy storage branch number, the net power is obtained by subtracting the charging / discharging power of each interface, the interface voltage status is obtained by averaging the voltage, and the interface current status is obtained by averaging the current. Alarm status is also retained, forming a set of energy storage operating statuses. The step-up transformer load rate, step-up transformer temperature, high-voltage switchgear temperature, high-voltage switchgear opening / closing status, and energized status are read from the station topology association dataset. Organized according to circuit number, the average load rate of step-up transformers under the same circuit number is taken as the transformer load status, the maximum temperature of step-up transformers under the same circuit number is taken as the transformer temperature status, and the maximum temperature of high-voltage switchgear under the same circuit number is taken as the switchgear temperature status. The switchgear operation status is formed based on the high-voltage switchgear opening / closing status and the high-voltage switchgear energized status. The transformer load status, transformer temperature status, switchgear temperature status, and switchgear operation status are aggregated to form a set of electrical equipment operating statuses. The internal temperature and humidity, condensation status, water immersion status, and smoke status of the cabin are read from the station topology association dataset. Status and ventilation status are organized according to the cabin area number. Temperature and humidity are averaged. The condensation, water immersion, and smoke statuses within the cabin are converted to Boolean values ​​and then merged. When condensation is present, water immersion indicates water accumulation or a water immersion alarm, and smoke indicates a smoke alarm or smoke concentration exceeding safety limits, the corresponding status is recorded as abnormal. When condensation is absent, water immersion indicates no water accumulation and no water immersion alarm, and smoke indicates no smoke alarm and smoke concentration not exceeding safety limits, the corresponding status is recorded as normal. If there are missing data collections, data timeouts, or invalid status identifiers in the condensation, water immersion, or smoke states, the corresponding status identifiers will be treated as abnormal. If any status identifier in the condensation, water immersion, or smoke states is abnormal, the cabin environment safety status is determined to be abnormal; otherwise, the cabin environment safety status is determined to be normal, forming a cabin environment status set. The photovoltaic output status set, grid-connected operation status set, energy storage operation status set, electrical equipment operation status set, and cabin environment status set are collected according to branch number, circuit number, and cabin area number to generate cabin station status collection data.

[0075] The system reads the photovoltaic (PV) output status from the station status collection data, calculates the product of PV branch voltage and PV branch current to form the real-time power of the PV branch, calculates the difference between the inverter's rated output power and the inverter's output power to form the potential power margin, and calculates the difference between the inverter's output power and the minimum allowable output power of the PV branch to form the potential power margin. The real-time power of the PV branch, the potential power margin, and the potential power margin are used as power regulation constraints for the PV output status. Finally, the system calculates the grid connection voltage and the upper limit of the grid connection voltage. The voltage margin is formed by calculating the difference between the grid-connected voltage and the lower limit of the grid-connected voltage, and the frequency margin is formed by calculating the difference between the grid-connected frequency and the upper limit of the grid-connected frequency, and the lower limit of the grid-connected frequency. The power factor margin is formed by calculating the difference between the power factor and the lower limit of the power factor. The harmonic state is formed by calculating the ratio between the harmonic state and the harmonic allowable limit. The voltage margin, frequency margin, power factor margin and harmonic occupancy ratio are used as grid-connected quality constraints for grid-connected operation.

[0076] The difference between the energy storage state of charge (SBC) and the upper limit of the SBC is calculated to form the rechargeable margin. The difference between the energy storage SBC and the lower limit of the SBC is calculated to form the dischargeable margin. The difference between the energy storage charging power and the upper limit of the charging power is calculated to form the charging power margin. The difference between the energy storage discharging power and the upper limit of the discharging power is calculated to form the discharging power margin. Based on the energy storage interface voltage, energy storage interface current, and energy storage interface alarm status, it is determined whether the energy storage interface meets the charging and discharging conditions. When the energy storage interface voltage is within the allowable range, the energy storage interface current is within the allowable range, and the energy storage interface alarm status is zero, the energy storage interface is determined to meet the charging and discharging conditions. When at least one of the energy storage interface voltage, energy storage interface current, and energy storage interface alarm status fails to meet the corresponding allowable conditions, the energy storage interface is determined not to meet the charging and discharging conditions, and the charging and discharging power is limited to zero or limited to the safe derating range, thus forming the charging and discharging constraints for the energy storage operation status.

[0077] The ratio between the load factor and the rated load factor of the step-up transformer is calculated to form the load occupancy ratio. The difference between the temperature of the step-up transformer and its upper limit of allowable temperature is calculated to form the transformer temperature margin. The difference between the temperature of the high-voltage switchgear and its upper limit of allowable temperature is obtained to form the switchgear temperature margin. Based on the opening and closing status and the energized status of the high-voltage switchgear, it is determined whether the opening and closing actions meet the execution conditions. When the energized status of the high-voltage switchgear is in a permissible operating state and the opening and closing status has not reached the target state, the opening and closing actions are deemed to meet the execution conditions. When the energized status of the high-voltage switchgear is in a prohibited operating state, the opening and closing actions are deemed not to meet the execution conditions. The load occupancy ratio, transformer temperature margin, switchgear temperature margin, and... The results of the opening and closing actions are organized to form equipment load constraints for the operating status of electrical equipment. The difference between the internal temperature and the upper limit of the internal temperature and the lower limit of the internal temperature are calculated to form a temperature margin. The difference between the internal humidity and the upper limit of the internal humidity and the lower limit of the internal humidity are calculated to form a humidity margin. When any of the states of condensation, water immersion, and smoke is abnormal, it is determined that there is an abnormal state of the internal environment, forming environmental safety constraints for the internal environment. The power regulation constraints, grid connection quality constraints, charging and discharging constraints, equipment load constraints, and environmental safety constraints are organized according to the branch number, circuit number, and internal area number to form the basis for boundary construction.

[0078] Based on the boundary construction criteria, photovoltaic power regulation boundary, grid connection quality allowable boundary, energy storage charging and discharging boundary, electrical equipment load-bearing boundary, and cabin environment safety boundary are constructed to form a multi-dimensional operation boundary.

[0079] Furthermore, based on the boundary construction criteria, the inverter output power, photovoltaic branch voltage, and photovoltaic branch current are compared with the upper and lower limits of the power regulation constraints to determine the adjustable photovoltaic power range. For example, the adjustable range of inverter output power is 0 to rated power, the allowable range of photovoltaic branch voltage is 350 to 450V, and the allowable range of photovoltaic branch current is 0 to 12A, forming the photovoltaic power regulation boundary. The voltage, frequency, active power, reactive power, power factor, and harmonic indicators under grid-connected operation are compared with the allowable range and the difference is calculated. This involves determining voltage margin, frequency margin, power margin, power factor margin, and harmonic margin. For example, the allowable voltage range is 380–420V, the allowable frequency range is 49.5–50.5Hz, and the power factor range is 0.95–1.0, forming the grid connection quality allowable boundaries. Then, the difference between the energy storage's state of charge, charging / discharging power, voltage, and current and the upper and lower limits of the charging / discharging constraints is calculated to determine the energy storage's rechargeable range, discharging range, and interface safety range. For example, the allowable state of charge range is 20–80%, and the charging power... The allowable power range is 0–50kW, the allowable discharge power range is 0–50kW, and the allowable energy storage interface voltage range is 400–450V, forming the energy storage charging and discharging boundaries. The load rate, equipment temperature, and switch status during the operation of electrical equipment are compared and the differences are calculated with the equipment's load-bearing constraints and safe temperature limits to determine the load-bearing range and equipment safety range. For example, the allowable load rate range for a step-up transformer is 0–90%, the allowable temperature range for a step-up transformer is 0–85℃, and the allowable temperature range for a high-voltage switchgear is 0–70℃. The electrical equipment bearing capacity boundary is formed by comprehensively judging the temperature and humidity difference, condensation, water immersion, smoke and ventilation conditions in the cabin environment. The safe range of each condition is summarized, for example, the allowable range of cabin temperature is 15-35℃ and the allowable range of cabin humidity is 30-70%, forming the cabin environment safety boundary. The photovoltaic power regulation boundary, grid connection quality allowable boundary, energy storage charging and discharging boundary, electrical equipment bearing capacity boundary and cabin environment safety boundary are collected according to branch number, circuit number and cabin area number to form a multi-dimensional operation boundary.

[0080] The terminal topology association dataset is classified according to the boundary type in the multidimensional operation boundary to generate a classified operation state set; the classified operation state set is matched with the multidimensional operation boundary to generate boundary matching results; the state boundary difference is calculated on the boundary matching results to determine the remaining adjustment margin, remaining carrying capacity margin and remaining safety margin of each type of operation state relative to the corresponding boundary, forming boundary margin data.

[0081] Furthermore, according to the boundary types in the multidimensional operational boundary, the photovoltaic output status, grid-connected operation status, energy storage operation status, electrical equipment operation status, and cabin environment status in the cabin topology association dataset are respectively classified into the corresponding boundary types to form a categorized operational status set. Each operational status in the categorized operational status set is matched one by one with the corresponding photovoltaic power regulation boundary, grid-connected quality allowable boundary, energy storage charging and discharging boundary, electrical equipment load-bearing boundary, and cabin environment safety boundary in the multidimensional operational boundary. The difference between the actual operating value and the upper and lower limits of the boundary is compared to generate the boundary matching result. For example, if the actual photovoltaic output is 90kW and the photovoltaic power adjustment boundary is 0-100kW, then the matching difference is 10kW. If the actual energy storage state of charge is 45% and the charge / discharge boundary is 20-80%, then the matching difference is 35%. Based on the boundary matching results, the differences between various operating states and their corresponding boundaries are calculated to obtain the remaining adjustable power, remaining load capacity, and remaining safety margin, forming boundary margin data. For example, the remaining adjustment margin of photovoltaic output is 10kW, the remaining load capacity margin of inverter load is 5%, and the remaining safety margin of cabin temperature is 2℃.

[0082] The boundary margin data is divided into normal margin state, contraction margin state, and protection margin state. The contraction margin state and protection margin state are then grouped according to the electrical topology relationship and intra-cabin spatial relationship in the cabin topology association dataset to generate a boundary approximation object set. The boundary approximation object set is then analyzed to determine the operating state category, electrical circuit, and intra-cabin area to which the boundary approximation occurs, generating the boundary approximation result.

[0083] Furthermore, based on the cabin topology association dataset and the boundary margins of various boundaries in the multi-dimensional operational boundary, the allowable state ranges for each operational state are set. The allowable state ranges for residual adjustment margin, residual load margin, and residual safety margin are between 5% and 30%. Choosing 5%-30% ensures that constraints are applied in advance when the operational state approaches the boundary, preventing over-limits while maintaining normal operating space. The residual adjustment margin, residual load margin, and residual safety margin for each operational state in the boundary margin data are compared with the allowable state ranges. When the residual margin of the operational state... When the remaining margin of the operating state exceeds the upper limit of the allowable state range, the operating state is classified as the normal margin state. When the remaining margin of the operating state is lower than the lower limit of the allowable state range, the operating state is classified as the protection margin state. The remaining operating states are classified as the contraction margin state. Based on the electrical circuit number, branch number, and cabin area number of each equipment data group in the cabin topology association dataset, the operating states of the contraction margin state and the protection margin state are classified and organized. The contraction margin state and the protection margin state located in the same electrical circuit and cabin area are summarized to form the boundary approximation object set.

[0084] The contraction margin status and protection margin status are categorized and organized according to the electrical circuit number, branch number, and cabin area number of each equipment data group in the cabin topology association dataset. Each boundary approximation object is traversed, and the electrical circuit number, branch number, and cabin area number of the equipment group to which the object belongs are used to determine the operating status category of the boundary approximation object as either a contraction margin status or a protection margin status based on the boundary margin data. The corresponding electrical circuit number, branch number, and cabin area number of the object are then bound to the operating status category to generate the boundary approximation result.

[0085] Based on the operational status category in the boundary approximation results, select the boundary type that needs to be contracted from the multidimensional operational boundaries, and mark the boundary type that needs to be contracted as a contraction object by the electrical circuit and the area inside the cabin, generating a boundary contraction object set; perform boundary positioning and screening on the boundary contraction object set to determine the photovoltaic power regulation boundary, grid connection quality allowable boundary, energy storage charging and discharging boundary, electrical equipment bearing boundary or cabin environment safety boundary that needs to be contracted, and generate a target contraction boundary set;

[0086] Furthermore, each boundary approximation object in the boundary approximation result is traversed, the operating status category of each boundary approximation object is read, the boundary type corresponding to the operating status category is selected from the multi-dimensional operating boundary, and the boundary type is marked in the electrical circuit number, branch number and cabin area number to which the boundary approximation object belongs, forming a boundary shrinkage object set.

[0087] Iterate through each boundary shrinkage object in the boundary shrinkage object set, read the object's boundary type, electrical circuit number, branch number, and cabin area number. Match the boundary type of the boundary shrinkage object with the photovoltaic power regulation boundary, grid connection quality allowable boundary, energy storage charging and discharging boundary, electrical equipment bearing capacity boundary, and cabin environment safety boundary in the multi-dimensional operation boundary. Then, compare the electrical circuit number, branch number, and cabin area number of the boundary shrinkage object with the electrical circuit, branch, and cabin area in the boundary approximation results. When the boundary type matches and the position matches, record the corresponding photovoltaic power regulation boundary, grid connection quality allowable boundary, energy storage charging and discharging boundary, electrical equipment bearing capacity boundary, or cabin environment safety boundary as the target boundary to be shrunk, forming a target shrinkage boundary set.

[0088] By applying first-level and second-level contraction processing to the target contraction boundary set using the contraction margin state and protection margin state respectively, a hierarchical contraction boundary set is generated. The hierarchical contraction boundary set is then merged with the non-contracted boundaries in the multidimensional running boundary to generate a contraction boundary set.

[0089] Furthermore, each boundary object in the target contraction boundary set is traversed, and the upper and lower limits of the boundary object in the graded contraction boundary set are read. These correspond to the photovoltaic power regulation boundary, grid connection quality allowable boundary, energy storage charging and discharging boundary, electrical equipment bearing capacity boundary, or cabin environment safety boundary, respectively. For boundary objects in the contraction margin state, the boundary distance between the upper and lower limits is reduced by one level, that is, the upper limit is lowered and the lower limit is raised by 5%-10% each, forming a first-level contraction boundary, reducing the allowable operating range of the boundary. For boundary objects in the protection margin state, the boundary distance between the upper and lower limits is reduced by two levels, that is, the upper limit is lowered and the lower limit is raised by 10%-20% each, forming a second-level contraction boundary, further compressing the allowable operating range of the boundary. Based on the boundary margin data, the boundary distance between the upper and lower limits of each target boundary in the target contraction boundary set, and the rated operating safety margin of the corresponding control object, the adjustment range of the first-level and second-level contraction is set, with a value range of 5%-20%. The first-level and second-level contraction are performed on all boundary objects in the target contraction boundary set in sequence to form a graded contraction boundary set.

[0090] The non-contraction boundaries in the graded contraction boundary set are merged with the non-contraction boundaries in the multi-dimensional operation boundary set. The original state and the state after contraction processing of each boundary are preserved. The boundary is then sorted and classified according to the electrical circuit number, branch number and cabin area number to generate a contraction boundary set.

[0091] The control feasibility module is used to perform control action feasibility analysis on the shrinkage boundary set, determine the control actions that are allowed to be executed, the control actions that are prohibited from being executed, and the protection actions that must be executed within the same control cycle, form the control feasibility domain, and perform multi-objective strategy matching and conflict verification to generate a set of cooperative control strategies.

[0092] The feasible actions of the shrinkage boundary set are screened to determine the action restrictions of each controlled object within the same control cycle. Using the hierarchical shrinkage boundary set as the constraint benchmark, the action restrictions are classified and marked. Control actions that meet the constraints of the hierarchical shrinkage boundary set are marked as allowed control actions, control actions that do not meet the constraints of the hierarchical shrinkage boundary set are marked as prohibited control actions, and protection actions used to restore the constraints of the hierarchical shrinkage boundary set are marked as mandatory protection actions. The allowed control actions, prohibited control actions, and mandatory protection actions are aggregated to form the control feasible domain.

[0093] Furthermore, for each boundary object in the shrinking boundary set, the corresponding upper and lower limits of the shrinking boundary are read. Based on the upper and lower limits, the allowable action range of each controlled object in the current control cycle is set to form action restriction conditions. According to the upper and lower limits of the boundary of the hierarchical shrinking boundary set, the action restriction conditions are compared with the executable actions of each controlled object. Control actions that are completely within the allowable range of the hierarchical shrinking boundary set are marked as allowed control actions. Control actions that may cause the operating state to exceed the allowable range of the hierarchical shrinking boundary set are marked as prohibited control actions. Actions that can be used to restore the operating state to the allowable range of the hierarchical shrinking boundary set are marked as mandatory protection actions. The allowed control actions, prohibited control actions, and mandatory protection actions are organized and collected in turn according to the controlled object and execution order to generate the control feasible domain.

[0094] The feasible control domain is used to extract the allowed control actions and the required protection actions to form a set of candidate control strategies. The prohibited control actions in the feasible control domain are used as constraints to remove candidate strategies containing prohibited actions from the set of candidate control strategies, thereby generating a set of executable control strategies. The set of executable control strategies is combined and coordinated to generate a set of collaborative control strategies.

[0095] Furthermore, according to the controlled object, action type, and execution order, the allowed control actions and the mandatory protection actions are combined to form a candidate control strategy set. Based on the control actions marked as prohibited in the control feasible domain, strategies containing prohibited actions or that may trigger prohibited actions are removed from the candidate control strategy set to generate an executable control strategy set. The executable control strategy set is then grouped according to the controlled object. Executable control strategies that are repeatedly executed for the same controlled object within the same control cycle are deleted. Executable control strategies that contain opposite control actions for the same controlled object within the same control cycle are deleted. Executable control strategies that are not arranged according to the priority execution rule of mandatory protection actions are deleted. The remaining executable control strategies are then organized according to the controlled object and control cycle to generate a collaborative control strategy set.

[0096] The recovery evaluation module is used to distribute the collaborative control strategy set to the corresponding control layer of the photovoltaic prefabricated substation, collect execution feedback, perform boundary recovery evaluation and update the shrinkage boundary set based on the execution feedback, and output the substation control results.

[0097] The collaborative control strategy set is distributed to the corresponding control layer of the photovoltaic prefabricated substation, and the action status, execution completion status and abnormal feedback status returned by the corresponding control layer are collected.

[0098] Furthermore, the collaborative control strategy set is distributed to the corresponding control layers of the photovoltaic prefabricated substation, including the photovoltaic inverter control layer, energy storage management layer, electrical equipment control layer, cabin environment control layer, and communication operation and maintenance management layer. After each control layer executes the control strategy, the execution feedback data returned by the control layer is collected, including the action status, execution completion status, and abnormal feedback status of each control action.

[0099] Based on the hierarchical shrinkage boundaries and boundary margins of the shrinkage boundary set, the allowable state range of each operating state within the current control cycle is set; based on the execution feedback data, it is determined whether each operating state has recovered to the allowable state range; the boundary states that have not recovered to the allowable state range are further tightened to generate conservative update boundary data; the boundary states that have recovered to the allowable state range are gradually relaxed to generate recovery update boundary data.

[0100] Furthermore, based on the hierarchical contraction boundaries and boundary margins of the contraction boundary set, allowable state ranges for each operating state within the same control cycle are set. Specifically, the allowable state range for photovoltaic output is 80%-100% of the inverter's rated output power; the allowable state range for grid-connected operation is ±5% of grid voltage fluctuation; the allowable state range for energy storage operation is 20%-80% of the energy storage's state of charge; the allowable state range for electrical equipment operation is 0%-90% of the equipment load rate; and the allowable state range for the cabin environment is cabin temperature 15-35°C and cabin humidity 30-70%. Based on the action status, execution completion status, and abnormal feedback status in the execution feedback data, it is determined whether each running status has recovered to the allowable state range. For boundary objects that have not recovered to the allowable state range, a graded tightening process is performed according to the contraction margin status or protection margin status marked by the boundary object in the boundary approximation result, generating conservative updated boundary data. Among them, the boundary of the contraction margin status object is tightened according to the first-level contraction process, and the boundary of the protection margin status object is further tightened according to the second-level contraction process. For the boundary status that has recovered to the allowable state range, the boundary restrictions are relaxed step by step to generate restored updated boundary data.

[0101] The shrinkage boundary set is updated based on conservative update boundary data and restored update boundary data, and the station control results are output.

[0102] Furthermore, based on the conservative update boundary data and the recovery update boundary data, the original boundary values ​​in the shrinkage boundary set are replaced and adjusted. The boundary values ​​corresponding to the unrecovered state are replaced with the tightened conservative update boundary, and the boundary values ​​corresponding to the recovered state are adjusted with the relaxed recovery update boundary, thus completing the update of the shrinkage boundary set and outputting the cabin station control results.

[0103] In summary, this invention achieves precise location of various operational data sources, facilitates refined control and risk identification, and generates a module topology association dataset through spatial correlation. It improves the control accuracy and reduces operational risks of photovoltaic prefabricated module substations by constructing multi-dimensional operational boundaries based on the module topology association dataset. Furthermore, it increases early warning lead time and reduces the risk of exceeding limits by performing approximation object location analysis to generate boundary approximation results. Dynamic shrinkage generates a shrinking boundary set, improving system safety margin and reducing potential risks. Combining and coordinating these strategies generates a collaborative control strategy set, ensuring coordinated operation of various control objectives and improving the effectiveness of control execution. Finally, updating the shrinking boundary set outputs module control results, enhancing the reliability and stability of photovoltaic prefabricated module substation operation.

[0104] It should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention and are not intended to limit it. Although the present invention has been described in detail with reference to preferred embodiments, those skilled in the art should understand that modifications or equivalent substitutions can be made to the technical solutions of the present invention without departing from the spirit and scope of the technical solutions of the present invention, and all such modifications or substitutions should be covered within the scope of the claims of the present invention.

Claims

1. A smart control system for a photovoltaic prefabricated substation, characterized in that: include, The topology mapping module is used to collect multi-source operation data of photovoltaic prefabricated substations and aggregate them according to the same control cycle to generate a multi-source dataset of the substation. The module performs electrical topology mapping and intra-substation space mapping on the multi-source dataset of the substation to generate a topology association dataset of the substation. The margin analysis module is used to construct a multidimensional operating boundary based on the cabin topology association dataset, perform boundary margin analysis on the operating status in the cabin topology association dataset based on the multidimensional operating boundary, generate boundary approximation results, and dynamically shrink the multidimensional operating boundary based on the boundary approximation results to generate a shrinking boundary set. The control feasibility module is used to perform control action feasibility analysis on the shrinkage boundary set, determine the control actions that are allowed to be executed, the control actions that are prohibited from being executed, and the protection actions that must be executed within the same control cycle, form the control feasibility domain, and perform multi-objective strategy matching and conflict verification to generate a set of collaborative control strategies. The recovery evaluation module is used to distribute the collaborative control strategy set to the corresponding control layer of the photovoltaic prefabricated substation, collect execution feedback, perform boundary recovery evaluation and update the shrinkage boundary set based on the execution feedback, and output the substation control results.

2. The intelligent control system for photovoltaic prefabricated substation as described in claim 1, characterized in that: The specific steps for generating the multi-source dataset for the module station are as follows: The multi-source operational data includes photovoltaic power output data, grid connection quality data, energy storage operation data, electrical equipment operation data, protection action data, and cabin environment data; Read the acquisition time identifier and device source identifier of each data item in the multi-source operation data, and merge the data items whose acquisition time identifiers belong to the same control cycle to form a data set of the same cycle; The data items in the same period dataset are categorized and organized according to the equipment source identifier to generate a multi-source dataset for the cabin station.

3. The intelligent control system for photovoltaic prefabricated substation as described in claim 2, characterized in that: The specific steps for generating the cabin topology association dataset are as follows: Based on the equipment source identifier, the data items in the multi-source dataset of the cabin station are associated with equipment ownership to generate an equipment association dataset; Perform branch attribution parsing on the electrical connection identifiers in the device association dataset to generate a branch attribution dataset; The connection order of the branch attribution dataset is parsed to generate loop connection order data, and the power flow direction is parsed to generate loop power flow direction data. The loop connection order data and loop power flow direction data are aggregated to generate branch connection relationships. Write the branch affiliation dataset and branch connection relationship into the device association dataset to generate the electrical topology mapping dataset; Based on the equipment installation locations in the electrical topology mapping dataset, the electrical topology mapping dataset is spatially associated with the prefabricated cabin area, environmental monitoring points, and control equipment layout locations of the photovoltaic prefabricated cabin substation to generate a cabin-substation topology association dataset.

4. The intelligent control system for photovoltaic prefabricated substation as described in claim 3, characterized in that: The specific steps for constructing the multidimensional operational boundary based on the cabin topology association dataset are as follows. The photovoltaic output status, grid-connected operation status, energy storage operation status, electrical equipment operation status, and cabin environment status are extracted from the cabin topology association dataset and aggregated to form cabin status aggregation data. The operational constraints of the collected data on the status of the cabin are analyzed to determine the power regulation constraints of the photovoltaic output status, the grid connection quality constraints of the grid-connected operation status, the charging and discharging constraints of the energy storage operation status, the equipment load-bearing constraints of the electrical equipment operation status, and the environmental safety constraints of the cabin environment status, thus forming the basis for boundary construction. Based on the power regulation constraints, grid connection quality constraints, charge and discharge constraints, equipment load-bearing constraints, and environmental safety constraints in the boundary construction basis, photovoltaic power regulation boundaries, grid connection quality allowable boundaries, energy storage charge and discharge boundaries, electrical equipment load-bearing boundaries, and cabin environmental safety boundaries are constructed respectively, forming multi-dimensional operating boundaries.

5. The intelligent control system for photovoltaic prefabricated substation as described in claim 4, characterized in that: The specific steps for generating the boundary approximation result are as follows. The state of the cabin topology association dataset is classified according to the boundary type in the multidimensional operation boundary to generate a classified operation state set. The classified operation state set is then matched with the multidimensional operation boundary to generate the boundary matching result. The boundary matching results are used to calculate the state boundary difference to determine the remaining adjustment margin, remaining load margin, and remaining safety margin of various operating states relative to the corresponding boundaries, thus forming boundary margin data. The boundary margin data is divided into normal margin state, contraction margin state and protection margin state. The normal margin state is removed, and the contraction margin state and protection margin state are located according to the electrical topology relationship and the internal space relationship in the compartment topology association dataset to generate the boundary approximation object set. The approaching object set is analyzed to determine the operating state category, electrical circuit, and cabin area to which the boundary approach occurs, and the results are aggregated to generate the boundary approach results.

6. The intelligent control system for photovoltaic prefabricated substation as described in claim 5, characterized in that: The specific steps for generating the shrinking boundary set are as follows: Based on the operational status category in the boundary approximation results, select the boundary type that needs to be shrunk from the multidimensional operational boundaries, and mark the shrunk object by the electrical circuit and the internal area of ​​the compartment to generate a boundary shrinkage object set; Boundary location screening is performed on the boundary shrinkage object set to determine the photovoltaic power regulation boundary, grid connection quality allowable boundary, energy storage charging and discharging boundary, electrical equipment bearing boundary or cabin environment safety boundary that need to be shrunk, and a target shrinkage boundary set is generated. By applying first-level and second-level shrinkage processing to the target shrinkage boundary set using the shrinkage margin state and the protection margin state, a hierarchical shrinkage boundary set is generated. The hierarchical shrinkage boundary set is merged with the non-shrinkage boundaries in the multidimensional running boundary to generate a shrinkage boundary set.

7. The intelligent control system for photovoltaic prefabricated substation as described in claim 6, characterized in that: The specific steps for forming the controllable feasible region are as follows. The feasible action set is screened to determine the action constraints of each controlled object within the same control cycle; Using the hierarchical shrinkage boundary set as the constraint benchmark, the action restriction conditions are classified and marked. Control actions that satisfy the hierarchical shrinkage boundary set constraints are marked as allowed control actions, control actions that do not satisfy the hierarchical shrinkage boundary set constraints are marked as prohibited control actions, and protection actions used to restore the hierarchical shrinkage boundary set constraints are marked as mandatory protection actions. The allowed control actions, prohibited control actions, and mandatory protection actions are grouped together to form the control feasible domain.

8. The intelligent control system for photovoltaic prefabricated substation as described in claim 7, characterized in that: The specific steps for generating the collaborative control strategy set are as follows: Extract the allowed control actions and the necessary protection actions from the control feasible domain to form a set of candidate control strategies; Using prohibited control actions in the feasible control domain as constraints, candidate policies containing prohibited actions are removed from the candidate control policy set to generate an executable control policy set. The executable control strategy set is arranged according to the controlled object, action type and execution order. Executable control strategies that are repeatedly executed on the same controlled object, simultaneously executed with opposite actions on the same controlled object, or not executed with priority according to the protection actions that must be executed are removed, thus generating a collaborative control strategy set.

9. The intelligent control system for photovoltaic prefabricated substation as described in claim 8, characterized in that: The collection and execution feedback refers to the process of sending the collaborative control strategy set to the corresponding control layer of the photovoltaic prefabricated substation and collecting the action status, execution completion status and abnormal feedback status returned by the corresponding control layer.

10. The intelligent control system for photovoltaic prefabricated substation as described in claim 9, characterized in that: The specific steps for outputting the control results of the module station are as follows. Based on the hierarchical shrinkage boundaries and boundary margins of the shrinkage boundary set, the allowable state range of each operating state within the current control cycle is set. Based on the execution feedback data, determine whether each operating state has recovered to the allowable state range. For boundary states that have not recovered to the allowable state range and are at the first-level contraction boundary, continue to perform first-level contraction processing. For boundary states that have not recovered to the allowable state range and are at the second-level contraction boundary, continue to perform second-level contraction processing, generating conservative update boundary data. For boundary states that have recovered to the allowable state range, relax the boundary level according to the tiered contraction boundary set, generating recovery update boundary data. The shrinkage boundary set is updated based on conservative update boundary data and restored update boundary data, and the station control results are output.

Images