Power equipment selection optimization method and system based on load demand matching

Abstract

The application discloses a power utilization equipment selection optimization method and system based on load demand matching, and relates to the technical field of data processing.The method comprises the following steps: acquiring a load change trend sequence, a starting impact feature sequence and a load coordination relationship sequence; performing aging dynamic prediction to establish a power supply aging dynamic atlas; performing multi-level dynamic adjustment according to the load change trend sequence and the power supply aging dynamic atlas to obtain a power supply selection optimization strategy; performing starting impact dynamic compensation on the power supply selection optimization strategy according to the starting impact feature sequence to obtain a power supply selection intermediate strategy; performing mutual exclusion identification and three-phase balance optimization on the power supply selection intermediate strategy according to the load coordination relationship sequence to obtain a power supply selection target strategy; and performing power supply configuration optimization deployment of a target power utilization area.The application solves the technical problem of low matching precision of power utilization equipment selection and actual load demand in the prior art, and achieves the technical effect of improving the accuracy of power utilization equipment selection and the adaptability of power supply configuration.

Description

Technical Field

[0001] This invention relates to the field of data processing technology, and more specifically to a method and system for optimizing the selection of electrical equipment based on load demand matching. Background Technology

[0002] In the process of configuring electrical equipment, equipment selection is usually based on empirical parameters or static load data, making it difficult to fully consider the dynamic changes in load operation, startup impact, and the collaborative relationships of multiple loads for comprehensive analysis. When the load structure of the target power consumption area changes, the original selection scheme is prone to problems such as unreasonable equipment capacity configuration, insufficient power supply adaptability, and decreased operational stability, thereby affecting the overall operation of the power supply system and the efficiency of equipment utilization. Summary of the Invention

[0003] This application provides a method and system for optimizing the selection of electrical equipment based on load demand matching, which is used to address the technical problem of low accuracy in matching electrical equipment selection with actual load demand in the prior art.

[0004] In view of the above problems, this application provides a method and system for optimizing the selection of electrical equipment based on load demand matching.

[0005] The first aspect of this application provides a method for optimizing the selection of electrical equipment based on load demand matching, the method comprising: Based on the load operation data of the target power consumption area, time-series evolution analysis is performed to obtain the load change trend sequence, startup impact characteristic sequence, and load coordination relationship sequence. Aging dynamic prediction is performed based on the power supply configuration strategy of the target power consumption area to establish a power supply aging dynamic map. The power supply configuration strategy is dynamically adjusted at multiple levels according to the load change trend sequence and the power supply aging dynamic map to obtain a power supply selection optimization strategy, which includes a distribution selection optimization strategy, an electrical selection optimization strategy, and a protection selection optimization strategy. Startup impact dynamic compensation is performed on the power supply selection optimization strategy based on the startup impact characteristic sequence to obtain an intermediate power supply selection strategy. Startup impact dynamic compensation includes hierarchical correlation compensation of the power supply selection optimization strategy based on the startup impact characteristic sequence. Mutual exclusion identification and three-phase balance optimization are performed on the intermediate power supply selection strategy based on the load coordination relationship sequence to obtain a target power supply selection strategy. The power supply configuration optimization deployment of the target power consumption area is then executed according to the target power supply selection strategy.

[0006] A second aspect of this application provides a power equipment selection optimization system based on load demand matching, the system comprising: The system includes the following modules: an evolution analysis module for performing time-series evolution analysis based on load operation data of the target power consumption area to obtain load change trend sequences, startup impact characteristic sequences, and load coordination relationship sequences; a dynamic prediction module for performing dynamic aging prediction based on the power supply configuration strategy of the target power consumption area to establish a power supply aging dynamic map; a dynamic adjustment module for performing multi-level dynamic adjustment of the power supply configuration strategy based on the load change trend sequence and the power supply aging dynamic map to obtain a power supply selection optimization strategy, which includes a distribution selection optimization strategy, an electrical selection optimization strategy, and a protection selection optimization strategy; a dynamic compensation module for performing startup impact dynamic compensation on the power supply selection optimization strategy based on the startup impact characteristic sequence to obtain an intermediate power supply selection strategy, which includes hierarchical correlation compensation based on the startup impact characteristic sequence; and a strategy execution module for performing mutual exclusion identification and three-phase balance optimization on the intermediate power supply selection strategy based on the load coordination relationship sequence to obtain a target power supply selection strategy and executing the power supply configuration optimization deployment of the target power consumption area according to the target power supply selection strategy.

[0007] One or more technical solutions provided in this application have at least the following technical effects or advantages: This application performs time-series evolution analysis on load operation data of the target power consumption area to obtain load change trend sequences, startup impact characteristic sequences, and load coordination relationship sequences; performs aging dynamic prediction based on the power supply configuration strategy of the target power consumption area to establish a power supply aging dynamic map; performs multi-level dynamic adjustment of the power supply configuration strategy according to the load change trend sequence and the power supply aging dynamic map to obtain a power supply selection optimization strategy, which includes a distribution selection optimization strategy, an electrical selection optimization strategy, and a protection selection optimization strategy; performs startup impact dynamic compensation on the power supply selection optimization strategy according to the startup impact characteristic sequence to obtain an intermediate power supply selection strategy, which includes hierarchical correlation compensation on the power supply selection optimization strategy according to the startup impact characteristic sequence; performs mutual exclusion identification and three-phase balance optimization on the intermediate power supply selection strategy according to the load coordination relationship sequence to obtain a target power supply selection strategy, and executes the power supply configuration optimization deployment of the target power consumption area according to the target power supply selection strategy. Compared to existing technologies that primarily rely on empirical parameters or static load data for equipment selection, this invention performs time-series evolution analysis on load operation data to comprehensively characterize load change trends, startup impact characteristics, and multi-load collaborative relationships. It also incorporates the aging risks and topological correlation status of power supply equipment to implement hierarchical dynamic optimization of distribution equipment, electrical equipment, and protection equipment. Furthermore, it corrects the selection results through startup impact compensation, load mutual exclusion identification, and three-phase balance adjustment. This ensures that equipment capacity, rated parameters, protection configuration, and load connection methods match future load demands and the actual operating status of the power supply system, reducing risks such as unreasonable equipment capacity configuration, localized overload, startup impact, protection malfunctions, and three-phase imbalance. Ultimately, this improves the accuracy of equipment selection and the adaptability of power supply configuration. Attached Figure Description

[0008] To more clearly illustrate the technical solutions in the embodiments of the present invention, the accompanying drawings used in the description of the embodiments will be briefly introduced below. Obviously, the accompanying 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.

[0009] Figure 1 A schematic diagram of the process for optimizing the selection of electrical equipment based on load demand matching provided in this application embodiment; Figure 2 An architecture diagram of the aging risk prediction model in the equipment selection optimization method based on load demand matching provided in the embodiments of this application; Figure 3 This is a schematic diagram of the system structure for optimizing the selection of electrical equipment based on load demand matching, provided in an embodiment of this application.

[0010] Figure labeling: Evolutionary analysis module 11, dynamic prediction module 12, dynamic adjustment module 13, dynamic compensation module 14, strategy execution module 15. Detailed Implementation

[0011] This application provides a method and system for optimizing the selection of electrical equipment based on load demand matching. It addresses the technical problem of low accuracy in matching electrical equipment selection with actual load demand in the prior art by performing dynamic analysis and selection optimization based on load operation data and power supply configuration strategies, thereby improving the accuracy of electrical equipment selection and the adaptability of power supply configuration.

[0012] The technical solutions of the embodiments of this application will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only a part of the embodiments of this application, and not all of them. All other embodiments obtained by those skilled in the art based on the embodiments of this application without creative effort are within the scope of protection of this application.

[0013] It should be noted that any variation of the terms "comprising" and "having" is intended to cover non-exclusive inclusion, for example, a process, method, system, product, or server that includes a series of steps or units is not necessarily limited to those steps or units that are explicitly listed, but may include other steps or modules that are not explicitly listed or that are inherent to such processes, methods, products, or devices.

[0014] Example 1, as Figure 1 As shown, this application provides a method for optimizing the selection of electrical equipment based on load demand matching, the method comprising: Step S100: Perform time-series evolution analysis based on the load operation data of the target power consumption area to obtain the load change trend sequence, the start-up impact characteristic sequence, and the load coordination relationship sequence.

[0015] In this embodiment of the application, when performing time-series evolution analysis based on the load operation data of the target power consumption area, time-series operation curves of each load are constructed, and time-series evolution prediction of each load time-series operation curve is performed in combination with future windows to obtain the load change trend sequence; features of the load change trend sequence are extracted based on the start-up impact characteristic index to obtain the start-up impact characteristic sequence; features of the load change trend sequence are extracted based on the load coordination characteristic index to obtain the load coordination relationship sequence.

[0016] Furthermore, the method provided in the application embodiments, which performs time-series evolution analysis based on the load operation data of the target power consumption area to obtain a load change trend sequence, a startup impact characteristic sequence, and a load coordination relationship sequence, also includes: Based on the load operation data, construct the time-series operation curves for each load; predict the time-series evolution of each load operation curve based on the future window to obtain the load change trend sequence; extract features from the load change trend sequence based on the startup impact characteristic index to obtain the startup impact characteristic sequence; extract features from the load change trend sequence based on the load coordination characteristic index to obtain the load coordination relationship sequence.

[0017] Furthermore, the method provided in the application embodiments also includes: The starting impact characteristic indicators include the peak value of the instantaneous starting current, the starting duration, the starting frequency, the continuous starting interval, and the starting impact energy.

[0018] Furthermore, the method provided in the application embodiments also includes: The load coordination characteristic indicators include load running time overlap characteristics, load start-stop correlation characteristics, and load power superposition characteristics.

[0019] In this embodiment of the application, the collected data of each load in the target power consumption area is classified according to the load identification based on the load operation data, and the operating current, operating voltage, operating power and start-stop status of the same load are sorted according to the collection time. The collection time is used as the time axis, and the operating current, operating voltage, operating power and start-stop status are used as operating status parameters. The continuous operating change trajectory of each load is formed in time sequence, and the time sequence operation curve of each load is constructed. For example, the target power consumption area includes motor M1, air conditioning unit AC1, and lighting circuit LT1, with a data acquisition period of 5 minutes. At the three data acquisition time points of 08:00, 08:05, and 08:10, the operating current of motor M1 is 42A, 36A, and 18A, respectively, and the operating power is 15kW, 14kW, and 8kW, respectively, with start / stop states of running, running, and running, respectively. The operating current of air conditioning unit AC1 is 0A, 28A, and 22A, respectively, and the operating power is 0kW, 10kW, and 9kW, respectively, with start / stop states of stopped, running, and running, respectively. The operating current of lighting circuit LT1 is 6A, 6A, and 6A, with an operating power of 3kW for each, and the start / stop state is running for all of them. After sorting the above data according to load identification and acquisition time, time-series operating curves for motor M1, air conditioning unit AC1, and lighting circuit LT1 are generated respectively.

[0020] Next, when predicting the time-series operation curves of each load based on the future window, the historical operation change trajectory before the current moment is used as the prediction basis. A historical time segment with the same length as the future window is extracted, and the average operating power, average operating current, and number of start-stop states at each collection point within the historical time segment are calculated. The average operating power, average operating current, and the most frequently occurring start-stop state at the same time position in the historical time segment are used as the predicted operating power, predicted operating current, and predicted start-stop state at the corresponding prediction time point in the future window, and arranged in the order of prediction time to obtain the load change trend sequence. For example, the future window is set to 08:00 to 09:00, and historical time segments from 08:00 to 09:00 in the previous three working days are selected as the basis for prediction. For motor M1 at the prediction time point of 08:10, the operating power at the same time position in the previous three working days is 8.2kW, 7.8kW, and 8.0kW, respectively, and the operating current is 18.5A, 17.5A, and 18.0A, respectively. The start-stop state is running. Therefore, the predicted operating power at 08:10 is 8.0kW, the predicted operating current is 18.0A, and the predicted start-stop state is running. The predicted operating power, predicted operating current, and predicted start-stop state for each prediction time point in the future window are obtained in the same way, forming a load change trend sequence.

[0021] When extracting features from the load change trend sequence based on the start-up impact characteristic indicators, the predicted operating current and predicted start-up / stop status are read from the load change trend sequence. The time point when the predicted start-up / stop status changes from a stopped state to an operating state is identified as the start-up start time, and the time point when the predicted operating current falls back to the rated operating current range is identified as the start-up end time. Between the start-up start time and the start-up end time, the peak value of the instantaneous start-up current, the start-up duration, the start-up frequency, the consecutive start-up interval, and the start-up impact energy are extracted. Among them, the peak value of the instantaneous start-up current is the maximum value of the predicted operating current within this time period, the start-up duration is the time difference between the start-up end time and the start-up time, the start-up frequency is the number of start-up times within the future window, the consecutive start-up interval is the time difference between adjacent start-up times, and the start-up impact energy is the cumulative value of the portion of the predicted operating current exceeding the rated operating current over time. The extraction results are arranged by load identifier and prediction time to obtain the start-up impact characteristic sequence.

[0022] When extracting features from the load change trend sequence based on load coordination characteristic indicators, the predicted operating power and predicted start / stop status of each load are aligned according to the same predicted time point within the future window. Based on the alignment results, the time overlap interval of different loads simultaneously in the operating state is calculated to obtain the load operating time overlap feature. The start and stop times of different loads are compared to obtain the load start / stop association feature. The predicted operating power of multiple loads in the operating state at the same predicted time point is summed to obtain the load power superposition feature. The load operating time overlap feature, load start / stop association feature, and load power superposition feature are arranged according to the load combination relationship and the predicted time to obtain the load coordination relationship sequence. For example, within the future window from 08:00 to 09:00, if the predicted start / stop state of motor M1 is from 08:00 to 08:40, the predicted start / stop state of air conditioning unit AC1 is from 08:05 to 08:45, and the predicted start / stop state of lighting circuit LT1 is from 08:00 to 09:00, then the time interval during which motor M1 and air conditioning unit AC1 are simultaneously in operation is from 08:05 to 08:40, corresponding to a load operation time overlap characteristic of 35 minutes; the start time of motor M1 is 08:00, and the start time of air conditioning unit AC1 is 08:05. The two start times differ by 5 minutes, corresponding to the load start-stop correlation characteristic that the air conditioning unit AC1 starts within 5 minutes after motor M1 starts; at the predicted time point of 08:10, the predicted operating power of motor M1 is 8kW, the predicted operating power of air conditioning unit AC1 is 9kW, and the predicted operating power of lighting circuit LT1 is 3kW, so the load power superposition characteristic at this predicted time point is 20kW; by arranging the above 35-minute time overlap interval, 5-minute start-stop time difference, and 20kW power summation result according to the load combination relationship and prediction time, the load coordination relationship sequence is obtained.

[0023] Step S200: Based on the power supply configuration strategy of the target power consumption area, perform dynamic prediction of aging and establish a dynamic power supply aging map.

[0024] In this embodiment of the application, when performing dynamic aging prediction based on the power supply configuration strategy of the target power consumption area, firstly, multi-device dynamic monitoring is performed based on the power supply configuration strategy of the target power consumption area to obtain monitoring data of multiple devices; then, aging-related attention allocation is performed on the monitoring data of multiple devices to establish multiple device monitoring prominence vectors; next, machine learning is performed based on the aging logs of multiple devices to generate an aging risk prediction model; finally, the multiple device monitoring prominence vectors are input into the aging risk prediction model to obtain multiple aging risk prediction results, and topological association binding is performed on the multiple aging risk prediction results to generate a power supply aging dynamic map.

[0025] Furthermore, the method provided in the application embodiments, which performs dynamic aging prediction based on the power supply configuration strategy of the target power consumption area and establishes a dynamic power supply aging map, further includes: According to the power supply configuration strategy, multi-device dynamic monitoring is performed to obtain monitoring data of multiple devices; aging-related attention allocation is performed on the monitoring data of multiple devices to establish multiple device monitoring prominence vectors; machine learning is performed on the aging logs of multiple devices to generate an aging risk prediction model; the multiple device monitoring prominence vectors are input into the aging risk prediction model to obtain multiple aging risk prediction results, and topology association binding is performed on the multiple aging risk prediction results to generate the power supply aging dynamic map.

[0026] In this embodiment, when performing multi-device dynamic monitoring according to the power supply configuration strategy, the equipment selection configuration, equipment rated parameters, equipment connection relationships, power supply paths, and protection coordination relationships in the power supply configuration strategy are first read to determine multiple power distribution devices, multiple electrical devices, and multiple protection devices in the target power consumption area, and a monitoring object list is established according to the device identification. Then, the monitoring location of each device is determined according to the equipment connection relationship and power supply path. With a 5-minute acquisition cycle, the operating current, operating voltage, operating power, load rate, temperature rise, running time, number of disconnections, number of protection actions, and fault records of each device are collected. Among them, the load rate is obtained by dividing the operating power by the rated power of the device, the temperature rise is obtained by subtracting the ambient temperature from the operating temperature of the device, the running time is obtained according to the cumulative time the device is in the operating state, the number of disconnections is obtained according to the cumulative opening and closing records of the switch equipment, the number of protection actions is obtained according to the cumulative action records of the protection equipment, and the fault records are recorded according to the number of fault occurrences and the duration of the fault, thereby obtaining monitoring data for multiple devices.

[0027] Next, when allocating attention based on aging-related data from multiple devices, the following steps are taken: First, the operating current, operating voltage, operating power, load rate, temperature rise, operating duration, number of interruptions, number of protection actions, and fault records for each device are normalized to obtain characteristic values ​​for operating current, operating voltage, operating power, load rate, temperature rise, operating duration, number of interruptions, number of protection actions, and fault records. Then, historical monitoring characteristic values ​​and historical aging risk values ​​for similar devices are extracted from the aging logs of multiple devices. The historical aging risk value is calculated using historical fault counts, historical fault durations, historical maintenance counts, and historical replacement records. Specifically, the historical fault counts are divided by a preset fault count limit to obtain a normalized fault count value; the historical fault durations are divided by a preset fault duration limit to obtain a normalized fault duration value; the historical maintenance counts are divided by a preset maintenance count limit to obtain a normalized maintenance count value; when historical replacement records exist, the replacement record value is recorded as 1, and when no historical replacement records exist... When recording historical replacement records, the replacement record value is recorded as 0. The normalized values ​​of the number of faults, the duration of faults, the number of maintenance, and the replacement record value are multiplied by 0.30, 0.25, 0.20, and 0.25 respectively, and then summed to obtain the historical aging risk value. The correlation coefficient between each historical monitoring feature value and the historical aging risk value is calculated. The calculation first calculates the average value of the corresponding historical monitoring feature value and the average value of the historical aging risk value. Then, the difference between the historical monitoring feature value and its average value and the difference between the historical aging risk value and its average value in each historical sample are multiplied and summed to obtain the co-variance. At the same time, the sum of the squares of the differences between the historical monitoring feature values ​​and the sum of the squares of the differences between the historical aging risk values ​​are multiplied and squared to obtain the normalized variation. The co-variance is divided by the normalized variation to obtain the corresponding correlation coefficient. The absolute value of each correlation coefficient is divided by the sum of the absolute values ​​of all correlation coefficients to obtain the aging association weight corresponding to each monitoring feature value. The monitoring feature value is multiplied by the corresponding aging association weight and combined according to the equipment identifier to establish multiple equipment monitoring prominence vectors.

[0028] Finally, when performing machine learning based on the aging logs of multiple devices, the historical operating current characteristic values, historical operating voltage characteristic values, historical operating power characteristic values, historical load rate characteristic values, historical temperature rise characteristic values, historical running time characteristic values, historical interruption count characteristic values, historical protection action count characteristic values, and historical fault record characteristic values ​​from the aging logs of multiple devices are used as training input data, and the historical aging risk values ​​of the corresponding devices are used as training output data; for example... Figure 2As shown, the architecture of the aging risk prediction model includes an input layer, a hidden layer, and an output layer. The input layer receives nine types of historical monitoring feature values, the hidden layer performs nonlinear mapping on these nine types of historical monitoring feature values, and the output layer outputs the aging risk value of the corresponding device. The aging risk prediction model adopts a three-layer feedforward neural network structure, including an input layer, a hidden layer, and an output layer. The input layer includes 9 input nodes, the hidden layer includes 12 hidden nodes, and the output layer includes 1 output node, which outputs the aging risk value between 0 and 1. The model parameters include a 9×12 input-to-hidden layer connection weight matrix, a 1×12 hidden layer bias vector, a 12×1 hidden-to-output layer connection weight matrix, a 1×1 output layer bias, a learning rate of 0.01, 500 iterations, a batch size of 32 samples, and an error threshold of 0.05. During training, the nine types of historical monitoring feature values ​​for each historical sample are... The eigenvalues ​​are input into the input layer, and the aging risk values ​​of the hidden layer output and the output layer are calculated sequentially. Based on the error between the aging risk value of the output layer and the historical aging risk value in the training output data, the connection weight matrix and bias vector are adjusted in reverse until the number of iterations is reached or the error is less than the error threshold, thus generating an aging risk prediction model. Multiple device monitoring prominence vectors are input into the aging risk prediction model, and multiple aging risk prediction results are output. These multiple aging risk prediction results include the aging risk values ​​corresponding to each power distribution device, each electrical device, and each protection device. Based on the device connection relationship, power supply path, and protection coordination relationship in the power supply configuration strategy, multiple power distribution devices, multiple electrical devices, and multiple protection devices are used as topology nodes, and the electrical connection relationship, upstream and downstream power supply relationship, and protection coordination relationship between devices are used as topology edges. The aging risk value corresponding to each device is bound to the corresponding topology node, generating a dynamic power supply aging graph.

[0029] The power supply aging dynamic map includes equipment nodes, topology connection edges, and node aging attributes. Equipment nodes include distribution equipment nodes, electrical equipment nodes, and protection equipment nodes. Topology connection edges include electrical connection edges, upstream and downstream power supply edges, and protection coordination edges. Node aging attributes include equipment identifier, equipment type, equipment monitoring highlight vector, aging risk value, acquisition time, and associated power supply path. When generating the power supply aging dynamic map, equipment nodes are first established according to the equipment connection relationships in the power supply configuration strategy. Then, upstream and downstream power supply edges are established according to the power supply path, and protection coordination edges are established according to the protection coordination relationships. Subsequently, the aging risk value output by the aging risk prediction model is written into the corresponding equipment node according to the equipment identifier. The equipment node is then topologically associated and bound to its upstream power supply equipment, downstream power receiving equipment, and corresponding protection equipment, thereby forming a power supply aging dynamic map that can simultaneously represent the equipment aging status and power supply correlation relationships.

[0030] The power supply aging dynamic map established by the above aging dynamic prediction can intuitively show the aging status of each device and its correlation in the power supply network. This helps to identify high-risk devices and potential fault paths in advance, optimize equipment selection and power supply layout, improve the reliability and stability of the power supply system, and provide data support for dynamic adjustment and capacity expansion decisions.

[0031] For example, the target power consumption area includes transformer T1, distribution cabinet P1, power distribution line L1, power distribution line L2, motor M1, air conditioning unit AC1, and circuit breakers QF1 and QF2. T1 supplies power to P1, and P1 supplies power to downstream loads via L1 and L2. L1 connects M1 and AC1, and QF1 and QF2 form an upstream and downstream protection coordination relationship. T1, P1, L1, L2, M1, AC1, QF1, and QF2 are respectively designated as equipment nodes. T1 to P1, P1 to L1, P1 to L2, L1 to M1, and L1 to AC1 are designated as upstream and downstream power supply edges, and QF1 to QF2 are designated as protection coordination edges. The aging risk value corresponding to each device is written into the corresponding equipment node. If the aging risk value of L1 is greater than or equal to the preset aging risk threshold, then L1 and its power supply path are recorded, and this power supply path is restricted from being used as an expansion carrying line for new loads during subsequent power supply selection optimization. When M1 and AC1 are in operation at the same time period, and the sum of their predicted operating power during that time period is greater than the allowable load value calculated by L1 according to the load safety factor, L1 is not used as an expansion carrying line for new loads. Instead, the adjustment target is determined from the branches whose aging risk value is less than the preset aging risk threshold and whose remaining allowable load value meets the expansion requirements. If the aging risk value of QF2 is greater than or equal to the preset aging risk threshold, then the predicted operating current of the branch where QF2 is located at each predicted time point within the future window is read, and the predicted operating current with the largest value is taken; this predicted operating current with the largest value is divided by the current setting current of QF2 to obtain the corresponding ratio; the ratio is compared one by one with the current multiple range in the pre-stored action characteristic table of QF2, and when the ratio falls into a certain current multiple range, the action time corresponding to that current multiple range is read; the current action time of QF1 is subtracted from the action time corresponding to that current multiple range to obtain the difference between the two; when the difference is less than QF1... When considering the time difference in the pre-stored coordination requirements between upper and lower level protections and QF2, the operating time of QF1 is adjusted to the sum of the operating time corresponding to the current multiple range and the time difference in the coordination requirements between upper and lower level protections. When the operating time corresponding to the current multiple range is greater than the pre-stored allowable clearing time in the current protection selection scheme, protection devices with rated current greater than or equal to the largest predicted operating current, breaking capacity greater than or equal to the pre-stored short-circuit current verification value of the branch where QF2 is located, and operating time less than or equal to the allowable clearing time are selected from the protection device parameter table, and QF2 is replaced by the selected protection device. Thus, the power supply aging dynamic map can provide aging constraints for the selection of distribution expansion locations, adjustment of electrical equipment access paths, and optimization of protection parameters, enabling the power supply selection optimization strategy to simultaneously meet load demands and equipment aging status requirements, improving the stability and adaptability of the optimized deployment of power supply configuration in the target power consumption area.

[0032] Step S300: Based on the load change trend sequence and the power supply aging dynamic spectrum, the power supply configuration strategy is dynamically adjusted at multiple levels to obtain a power supply selection optimization strategy, which includes a power distribution selection optimization strategy, an electrical selection optimization strategy, and a protection selection optimization strategy.

[0033] In this embodiment, when the power supply configuration strategy is dynamically adjusted in multiple levels based on the load change trend sequence and the power supply aging dynamic graph, the power supply configuration strategy is first decomposed to obtain the current power distribution selection scheme, the current electrical selection scheme, and the current protection selection scheme. Then, the power supply aging dynamic graph is decomposed in multiple levels to obtain the power distribution aging dynamic layer, the electrical aging dynamic layer, and the protection aging dynamic layer. Afterward, based on the load change trend sequence and combined with the power distribution aging dynamic layer, the electrical aging dynamic layer, and the protection aging dynamic layer, the current power distribution selection scheme is reconstructed through capacity expansion analysis, the current electrical selection scheme is reconstructed through overload analysis, and the current protection selection scheme is reconstructed through protection trigger association. The output is a power supply selection optimization strategy that includes the power distribution selection optimization strategy, the electrical selection optimization strategy, and the protection selection optimization strategy.

[0034] Furthermore, in the method provided in the application embodiment, the power supply configuration strategy is dynamically adjusted at multiple levels based on the load change trend sequence and the power supply aging dynamic spectrum to obtain a power supply selection optimization strategy. The power supply selection optimization strategy includes a distribution selection optimization strategy, an electrical selection optimization strategy, and a protection selection optimization strategy, and further includes: The power supply configuration strategy is disassembled to obtain the current power distribution selection scheme, the current electrical selection scheme, and the current protection selection scheme. The power supply aging dynamic graph is disassembled at multiple levels to obtain the power distribution aging dynamic layer, the electrical aging dynamic layer, and the protection aging dynamic layer. Based on the load change trend sequence and the power distribution aging dynamic layer, the current power distribution selection scheme is expanded and reconstructed, and the power distribution selection optimization strategy is output. Based on the load change trend sequence and the electrical aging dynamic layer, the current electrical selection scheme is overload analyzed and reconstructed, and the electrical selection optimization strategy is output. Based on the load change trend sequence and the protection aging dynamic layer, the current protection selection scheme is reconstructed using protection trigger association, and the protection selection optimization strategy is output.

[0035] In this embodiment, when disassembling the power supply configuration strategy, the equipment selection configuration, equipment rated parameters, equipment connection relationships, power supply paths, and protection coordination relationships within the target power consumption area are first read. The power supply configuration strategy is then divided into a distribution configuration section, an electrical configuration section, and a protection configuration section according to the equipment function type. The distribution configuration section includes the capacity level, rated current, rated voltage, laying path, and connection relationships of transformers, distribution cabinets, busbars, and power distribution lines, from which the current distribution selection scheme is obtained. The electrical configuration section includes the equipment model, rated power, operating current, starting method, and connection location of motors, air conditioning units, lighting circuits, and production electrical equipment, from which the current electrical selection scheme is obtained. The protection configuration section includes the rated current, setting value, operating time, breaking capacity, and upper / lower level protection coordination relationships of circuit breakers, fuses, relay protection devices, and leakage current protection devices, from which the current protection selection scheme is obtained.

[0036] When performing multi-level decomposition of the power supply aging dynamic map, the topology nodes, topology edges, and aging risk values ​​in the power supply aging dynamic map are first read. Among them, the topology nodes correspond to multiple power distribution devices, multiple electrical devices, and multiple protection devices. The topology edges correspond to the electrical connection relationships, upstream and downstream power supply relationships, and protection coordination relationships between devices. The aging risk values ​​correspond to the aging status of each device. Then, the power supply aging dynamic map is layered according to the device function type of the topology nodes. The topology nodes, topology edges, and aging risk values ​​corresponding to transformers, distribution cabinets, busbars, and distribution lines are assigned to the power distribution aging dynamic layer. The topology nodes, topology edges, and aging risk values ​​corresponding to motors, air conditioning units, lighting circuits, and production electrical equipment are assigned to the electrical aging dynamic layer. The topology nodes, topology edges, and aging risk values ​​corresponding to circuit breakers, fuses, relay protection devices, and leakage protection devices are assigned to the protection aging dynamic layer.

[0037] Next, when performing capacity expansion analysis and reconstruction of the current power distribution selection scheme based on the load change trend sequence and the power distribution aging dynamic layer, the inflection point of stage load growth and the peak load migration trajectory are extracted based on the load change trend sequence. Then, the current power distribution selection scheme is expanded and compensated based on the inflection point of stage load growth and the peak load migration trajectory to obtain the basic power distribution regulation strategy. Then, the power distribution aging dynamic layer is sensitively analyzed based on the aging risk threshold to obtain the power distribution aging sensitive topology. Finally, the power distribution regulation basic strategy is compensated and reconstructed based on the power distribution aging sensitive topology to generate the power distribution selection optimization strategy.

[0038] Subsequently, when reconstructing the current electrical selection scheme for overload analysis based on the load change trend sequence and the electrical aging dynamic layer, the system first reads the electrical equipment model, rated power, rated current, starting method, and connection location from the current electrical selection scheme. Then, it extracts the predicted operating power, predicted operating current, and predicted start / stop status of each electrical device within the future window from the load change trend sequence. The predicted operating power and rated power of the same electrical device are compared with its rated power, and the predicted operating current is compared with its rated current to obtain the load rate and overload period for each electrical device within the future window. The load rate is the ratio of predicted operating power to rated power, and the overload period is the time interval during which the predicted operating power exceeds the rated power or the predicted operating current exceeds the rated current. Finally, the system reads the aging data of the corresponding electrical equipment from the electrical aging dynamic layer. The system calculates risk values ​​and maps aging risk values ​​to overload periods. When an electrical device simultaneously has an aging risk value that reaches a preset risk range and an overload period within a future window, the device is identified as a device to be reconfigured. Subsequently, based on the load rate, overload duration, predicted operating power, predicted operating current, and aging risk value of the device to be reconfigured, the rated power, rated current, starting method, and connection location of the devices in the current electrical selection scheme are adjusted. Devices with insufficient rated power or rated current are replaced with devices that meet the predicted operating power and predicted operating current. Devices with large starting impacts are adjusted to soft start or frequency converter start. Devices whose connection location causes local overload are adjusted to power supply branches with remaining capacity that meet the connection requirements. This results in the output of an electrical selection optimization strategy.

[0039] Finally, when reconstructing the protection trigger association of the current protection selection scheme based on the load change trend sequence and the protection aging dynamic layer, the protection trigger is first fitted according to the load change trend sequence and the current protection selection scheme to obtain protection trigger fitting data; then, anomaly identification is performed based on the protection trigger fitting data to obtain protection anomaly pre-simulation features, including protection sensitivity decay characteristics, tripping delay drift characteristics, and upper and lower level protection coordination offset characteristics; then, the current protection selection scheme is correlated and compensated according to the protection anomaly pre-simulation features to obtain the basic protection adjustment strategy, and the protection aging dynamic layer is sensitively analyzed according to the aging risk threshold to obtain the protection aging sensitive topology; finally, the basic protection adjustment strategy is corrected according to the protection aging sensitive topology to generate the protection selection optimization strategy.

[0040] Furthermore, in the method provided in the application embodiment, the current power distribution selection scheme is expanded and reconstructed based on the load change trend sequence and the power distribution aging dynamic layer, and the power distribution selection optimization strategy is output, which further includes: Extract the inflection point of phased load growth and the peak load migration trajectory based on the load change trend sequence; perform capacity expansion compensation adjustment on the current power distribution selection scheme based on the inflection point of phased load growth and the peak load migration trajectory to obtain the basic power distribution regulation strategy; perform sensitivity analysis on the power distribution aging dynamic layer based on the aging risk threshold to obtain the power distribution aging sensitive topology; perform compensation reconstruction on the basic power distribution regulation strategy based on the power distribution aging sensitive topology to generate the power distribution selection optimization strategy.

[0041] In this embodiment, when extracting the inflection point of stage load growth and the peak load migration trajectory based on the load change trend sequence, the predicted load value of each distribution branch within the future window is read in the order of prediction time. The predicted load value of the previous prediction time point is subtracted from the predicted load value of the later prediction time point to obtain the load growth amount between adjacent prediction time points. The load growth amount is divided by the predicted load value of the previous prediction time point to obtain the load growth rate. When the load growth rate is greater than a preset growth rate threshold, and the predicted load values ​​of multiple consecutive prediction time points after the prediction time point are all greater than the predicted load values ​​before the prediction time point, the prediction time point is determined as the stage load growth inflection point. At the same time, the predicted load values ​​of each distribution branch are compared at each prediction time point. The predicted load value with the largest value is determined as the peak load, and the distribution branch corresponding to the peak load is determined as the branch where the peak load is located. The process of the branch where the peak load is located moving from one distribution branch to another is recorded in the order of prediction time to form the peak load migration trajectory.

[0042] Next, when adjusting the current power distribution scheme for capacity expansion compensation based on the inflection point of phased load growth and the peak load migration trajectory, the transformer capacity, rated current of the distribution cabinet, rated current of the bus, and current carrying capacity of the distribution line in the current power distribution scheme are read. The maximum predicted load value and maximum predicted current of each distribution branch after the inflection point of phased load growth are also read. For the transformer, the maximum predicted load values ​​of each distribution branch powered by the transformer after the inflection point of phased load growth are added together to obtain the total predicted load of the transformer. The total predicted load of the transformer is divided by the load safety factor to obtain the required transformer capacity, where the load safety factor is a capacity reserve factor less than 1, for example, 0.8. When the required transformer capacity is greater than the current transformer capacity, the transformer capacity is adjusted to be greater than or equal to the required transformer capacity. For distribution cabinets, busbars, and distribution lines, the maximum predicted current of the distribution branch to which the peak load migration trajectory points is divided by the load safety factor to obtain the required rated current of the distribution cabinet, the required rated current of the busbar, and the required current carrying capacity of the distribution line, respectively. When the required rated current of the distribution cabinet is greater than the current rated current of the current distribution cabinet, the rated current of the distribution cabinet is adjusted to be greater than or equal to the equipment rated current of the required rated current of the distribution cabinet. When the required rated current of the busbar is greater than the current rated current of the busbar, the rated current of the busbar is adjusted to be greater than or equal to the equipment rated current of the busbar. When the required current carrying capacity of the distribution line is greater than the current current carrying capacity of the distribution line, the current carrying capacity of the distribution line is adjusted to be greater than or equal to the line current carrying capacity of the required current carrying capacity of the distribution line. Thus, a basic distribution regulation strategy including the expansion target, expansion location, and the parameters of the distribution equipment after expansion is obtained.

[0043] Subsequently, when performing sensitivity analysis on the dynamic layer of power distribution aging based on the aging risk threshold, the aging risk value of each power distribution equipment node in the dynamic layer of power distribution aging is read, as well as the electrical connection relationship and upstream and downstream power supply relationship between each power distribution equipment node; the aging risk value of each power distribution equipment node is compared with the aging risk threshold one by one. When the aging risk value of a power distribution equipment node is greater than or equal to the aging risk threshold, the power distribution equipment node is identified as a sensitive node; the upstream power supply direction of the sensitive node is used to find the upstream power distribution equipment node directly connected to it, and the downstream power supply direction of the sensitive node is used to find the downstream power distribution equipment node directly connected to it. The found upstream and downstream power distribution equipment nodes are identified as associated nodes; the sensitive nodes, associated nodes, and the electrical connection relationship and upstream and downstream power supply relationship between sensitive nodes and associated nodes are combined to obtain the power distribution aging sensitive topology.

[0044] Finally, when compensating and reconstructing the basic distribution regulation strategy based on the distribution aging sensitive topology, the expansion objects and expansion locations in the basic distribution regulation strategy are matched with the sensitive nodes and associated nodes in the distribution aging sensitive topology. When the expansion object belongs to a sensitive node, it is not used as an expansion carrier device. Instead, distribution equipment with an aging risk value less than the aging risk threshold is searched in the same or adjacent power supply paths. The carrying capacity of this distribution equipment is calculated. The carrying capacity is the result of multiplying the rated capacity of the distribution equipment by the load safety factor and then subtracting the current predicted load value. When the carrying capacity is greater than or equal to the load value to be transferred, the load value to be transferred is connected to the power supply path corresponding to the distribution equipment. When the expansion location belongs to a sensitive node, the expansion object is not used as an expansion carrier device. When associating nodes, the required capacity, required rated current, or required current carrying capacity in the basic distribution regulation strategy is multiplied by an aging compensation coefficient to obtain the compensated demand parameters. The aging compensation coefficient is a capacity amplification factor greater than 1, such as 1.1. The distribution equipment parameters are then adjusted to have a capacity, current, or current carrying capacity greater than or equal to the compensated demand parameters. When the peak load migration trajectory passes through a sensitive node, the allowable load value of the power supply branch where the sensitive node is located is calculated. The allowable load value is the result of multiplying the rated capacity of the branch by the load safety factor. The portion of the predicted load value exceeding the allowable load value is taken as the load value to be transferred. The load value to be transferred is then allocated to power supply branches where the aging risk value is less than the aging risk threshold and the carrying capacity margin meets the requirements. The adjusted expansion targets, expansion locations, distribution equipment parameters, and load transfer results are integrated to generate a distribution selection optimization strategy.

[0045] By using the above-mentioned capacity expansion compensation and adjustment methods based on load change trends and power distribution aging dynamic maps, high-risk equipment can be identified in advance and loads can be reasonably allocated, ensuring that capacity expansion decisions take into account both equipment aging status and power supply safety; at the same time, the capacity utilization rate of power distribution equipment can be optimized, and the operational stability and reliability can be improved, providing a basis for dynamic adjustment and long-term power supply planning.

[0046] Furthermore, in the method provided in the application embodiment, the protection trigger association reconstruction of the current protection selection scheme is performed based on the load change trend sequence and the protection aging dynamic layer, and the protection selection optimization strategy is output, which further includes: Based on the load change trend sequence and the current protection selection scheme, protection trigger fitting is performed to obtain protection trigger fitting data; based on the protection trigger fitting data, anomaly identification is performed to obtain protection anomaly pre-simulation features, which include protection sensitivity attenuation features, tripping delay drift features, and upper and lower level protection coordination offset features; based on the protection anomaly pre-simulation features, the current protection selection scheme is correlated and compensated to obtain the protection adjustment basic strategy; based on the aging risk threshold, the protection aging dynamic layer is sensitive to analysis to obtain the protection aging sensitive topology; based on the protection aging sensitive topology, the protection adjustment basic strategy is corrected to generate the protection selection optimization strategy.

[0047] In this embodiment, when performing protection trigger fitting based on the load change trend sequence and the current protection selection scheme, the predicted operating current at each predicted time point in the load change trend sequence is read, and the setting current, operating time, and coordination relationship between upper and lower level protection devices in the current protection selection scheme are read. The operating time is the tripping time corresponding to different trigger ratio ranges pre-stored in the current protection selection scheme, and the trigger ratio is obtained by dividing the predicted operating current by the setting current. For each predicted time point, the predicted operating current is divided by the setting current of the corresponding protection device to obtain the trigger ratio. The trigger ratio is then compared one by one with each trigger ratio range pre-stored in the current protection selection scheme. When the trigger ratio is greater than or equal to the lower limit of a certain trigger ratio range and less than the upper limit of that trigger ratio range, it is determined that the trigger ratio matches the trigger ratio range, and the operating time corresponding to that trigger ratio range is read as the expected tripping time. The predicted operating current, setting current, trigger ratio, expected tripping time, corresponding protection device, and coordination relationship between upper and lower level protection are recorded in the order of predicted time to obtain protection trigger fitting data.

[0048] When identifying anomalies based on protection trigger fitting data, the system reads the predicted operating current, setting current, trigger ratio, expected trip time, and coordination relationship between upper and lower level protections from the protection trigger fitting data. When the predicted operating current is greater than or equal to the setting current, but the trigger ratio does not match any trigger ratio range pre-stored in the current protection selection scheme, the protection sensitivity attenuation feature is extracted. When the difference between the expected trip time and the action time corresponding to the same trigger ratio range in the current protection selection scheme is greater than zero, the trip delay drift feature is extracted. For the coordination offset characteristics of upper and lower level protection, firstly, determine the lower-level protection equipment and upper-level protection equipment in the same power supply path according to the coordination relationship of upper and lower level protection. Then, read the expected tripping time of the lower-level protection equipment and the expected tripping time of the upper-level protection equipment. Subtract the expected tripping time of the lower-level protection equipment from the expected tripping time of the upper-level protection equipment to obtain the actual action time difference. Next, read the coordination time difference pre-stored in the current protection selection scheme and compare the actual action time difference with the coordination time difference. When the actual action time difference is less than the coordination time difference, extract the upper and lower level protection coordination offset characteristics. Combine the protection equipment identification, anomaly occurrence time, protection sensitivity attenuation characteristics, tripping delay drift characteristics, and upper and lower level protection coordination offset characteristics in the order of prediction time to obtain the protection anomaly pre-simulation characteristics.

[0049] When adjusting the current protection selection scheme based on the characteristics of protection anomaly prediction, the following steps are taken: First, read the protection devices exhibiting sensitivity attenuation characteristics, the maximum normal operating current of the branch containing the protection device, and the predicted operating current. Subtract the maximum normal operating current from the predicted operating current to obtain the current difference. Divide the current difference by 2 and add it to the maximum normal operating current to obtain the adjusted setting current. Second, read the protection devices exhibiting tripping delay drift characteristics. Subtract the expected tripping time from the corresponding action time in the current protection selection scheme to obtain the delay time. Subtract the delay time from the original action time to obtain the adjusted action time. Third, read the protection devices exhibiting upper and lower level protection characteristics. For protection devices with offset characteristics, the estimated tripping time of the upper-level protection device is subtracted from the estimated tripping time of the lower-level protection device to obtain the actual operating time difference. Then, the actual operating time difference is subtracted from the pre-stored coordination time difference in the current protection selection scheme to obtain the time compensation value. When the time compensation value is greater than zero, the operating time of the upper-level protection device is increased by the time compensation value, so that the operating time difference after subtracting the operating time of the lower-level protection device from the operating time of the upper-level protection device is greater than or equal to the pre-stored coordination time difference in the current protection selection scheme. The adjusted setting current, operating time, and coordination relationship between the upper and lower-level protection devices are combined to obtain the basic protection adjustment strategy.

[0050] When performing sensitivity analysis on the dynamic layer of protection aging based on the aging risk threshold, the aging risk value, protection coordination relationship, and upstream and downstream power supply relationship of each protection device node in the dynamic layer of protection aging are read. The aging risk value of each protection device node is subtracted from the aging risk threshold to obtain the risk difference value. When the risk difference value is greater than or equal to zero, the corresponding protection device node is identified as a sensitive node. Then, the protection device nodes that have a direct superior-subordinate protection relationship with the sensitive node are read according to the protection coordination relationship, and the protection device nodes that are located on the same power supply path as the sensitive node are read according to the upstream and downstream power supply relationship. The sensitive nodes, the protection device nodes with direct superior-subordinate protection relationships, the protection device nodes on the same power supply path, and the corresponding protection coordination relationship and upstream and downstream power supply relationship are combined to obtain the protection aging sensitive topology.

[0051] When modifying the basic protection regulation strategy based on the aging sensitive topology, the protection device identifiers in the basic protection regulation strategy are compared one by one with the sensitive node identifiers in the aging sensitive topology. When the protection device identifier matches the sensitive node identifier, the aging risk value, setting current, and predicted operating current of other protection devices in the same power supply path are read, and the results of subtracting the aging risk value from the aging risk threshold, subtracting the maximum normal operating current from the setting current, and subtracting the setting current from the predicted operating current are calculated respectively. When the result of subtracting the aging risk value from the aging risk threshold is greater than zero, the result of subtracting the maximum normal operating current from the setting current is greater than zero, and the predicted operating current is greater than zero, the protection device identification in the basic protection regulation strategy is compared one by one with the sensitive node identifiers in the aging sensitive topology. When the result of subtracting the setting current from the current is greater than or equal to zero, the corresponding protection device is used as the replacement protection device. When the coordination relationship between upper and lower level protections in the basic protection adjustment strategy includes sensitive nodes, the action time difference obtained by subtracting the action time of the lower level protection device from the action time of the upper level protection device is recalculated. The action time of the upper level protection device is then added to the sum of the action time of the lower level protection device and the coordination time difference pre-stored in the current protection selection scheme, so that the adjusted action time difference is greater than or equal to the pre-stored coordination time difference. The replaced protection device, the corrected setting current, the corrected action time, and the corrected coordination relationship between upper and lower level protections are integrated to generate a protection selection optimization strategy.

[0052] By using the above adjustment methods based on protection anomaly prediction characteristics and aging-sensitive topologies, potential protection risks and equipment failures can be identified in advance, the configuration and operation coordination of protection equipment can be optimized, protection accuracy and response reliability can be improved, and the safety of load operation and the stability of power supply path can be ensured.

[0053] Step S400: Perform dynamic compensation for the power supply selection optimization strategy based on the startup impact feature sequence to obtain an intermediate power supply selection strategy. The dynamic compensation for startup impact includes hierarchical correlation compensation for the power supply selection optimization strategy based on the startup impact feature sequence.

[0054] In this embodiment, when performing dynamic compensation for startup impact based on the startup impact characteristic sequence of the power supply selection optimization strategy, the following parameters are read: startup instantaneous current peak value, startup duration, startup frequency, continuous startup interval, and startup impact energy. The distribution selection optimization strategy, electrical selection optimization strategy, and protection selection optimization strategy within the power supply selection optimization strategy are also read. The startup instantaneous current peak value is subtracted from the rated operating current of the corresponding equipment to obtain the startup current increment. This startup current increment is then added to the predicted operating current of the corresponding power supply branch to obtain the startup compensation current. The startup compensation current is then compared with the rated current of the distribution cabinet, the rated current of the bus, and the rated current of the distribution line, respectively. The system compares the allowable current value corresponding to the current carrying capacity. When the starting compensation current is greater than the corresponding allowable current value, it reads the pre-stored power distribution equipment selection parameter table and selects the power distribution equipment selection parameter with an allowable current value greater than or equal to the starting compensation current from the power distribution equipment selection parameter table. For transformers, the starting compensation current is multiplied by the corresponding supply voltage to obtain the starting compensation capacity. When the starting compensation capacity is greater than the current transformer capacity, it reads the pre-stored transformer capacity level table and selects the target transformer capacity greater than or equal to the starting compensation capacity from the transformer capacity level table. The selection parameter corresponding to the current transformer is then replaced with the selection parameter corresponding to the target transformer capacity.

[0055] When compensating the electrical selection optimization strategy based on the start frequency and continuous start interval, the start frequency is divided by the statistical time to obtain the number of starts per unit time. This number is then compared with the pre-stored allowable start count in the electrical selection optimization strategy. If the number of starts per unit time is greater than the allowable start count, the starting method of the corresponding electrical equipment is replaced with reduced-voltage starting. Similarly, the continuous start interval is compared with the pre-stored allowable continuous start interval in the electrical selection optimization strategy. If the continuous start interval is less than the allowable continuous start interval, the waiting time between two adjacent starts is set to be greater than or equal to the allowable continuous start interval. When compensating the protection selection optimization strategy based on the start duration and start impact energy, the protection device setting current is set to a value greater than the maximum normal operating current and less than the fault trigger current. The protection device's operating time is compared with the start duration. If the protection device's operating time is less than the start duration, the protection device with time-delay protection function in the current protection selection scheme is read, and its time-delay operating time is set to a value greater than the start duration and less than the allowable fault clearing time. Finally, the compensated distribution selection optimization strategy, electrical selection optimization strategy, and protection selection optimization strategy are combined to obtain an intermediate power supply selection strategy.

[0056] Step S500: Based on the load coordination relationship sequence, perform mutual exclusion identification and three-phase balance optimization on the intermediate power supply selection strategy to obtain the target power supply selection strategy, and execute the power supply configuration optimization deployment of the target power consumption area according to the target power supply selection strategy.

[0057] In this embodiment, mutual exclusion identification is performed on the intermediate power supply selection strategy based on the load coordination relationship sequence to obtain the mutual exclusion identification result; three-phase imbalance is deduced from the intermediate power supply selection strategy based on the load coordination relationship sequence to obtain the three-phase imbalance path; then, the intermediate power supply selection strategy is optimized and compensated based on the mutual exclusion identification result and the three-phase imbalance path to output the target power supply selection strategy, and the power supply configuration optimization deployment of the target power consumption area is executed according to the target power supply selection strategy.

[0058] Furthermore, in the method provided in the application embodiments, the process of performing mutual exclusion identification and three-phase balance optimization on the intermediate power supply selection strategy based on the load coordination relationship sequence to obtain the target power supply selection strategy further includes: The intermediate power supply selection strategy is mutually exclusive based on the load coordination relationship sequence to obtain the mutual exclusion identification result; the intermediate power supply selection strategy is three-phase imbalance deduction based on the load coordination relationship sequence to obtain the three-phase imbalance path; the intermediate power supply selection strategy is optimized and compensated based on the mutual exclusion identification result and the three-phase imbalance path to output the target power supply selection strategy.

[0059] In this embodiment, when performing mutual exclusion identification of the intermediate power supply selection strategy based on the load coordination relationship sequence, the load running time overlap characteristics, load start-stop association characteristics, and load power superposition characteristics in the load coordination relationship sequence are read, and the load access branch, load access phase, equipment start-up mode, power distribution equipment parameters, electrical equipment parameters, and protection equipment parameters in the intermediate power supply selection strategy are read. For any two loads, their running start time and running end time are read respectively. The larger of the two running start times is taken as the overlap start time, and the smaller of the two running end times is taken as the overlap end time. The overlap end time is then subtracted from the overlap start time to obtain the running time overlap duration. When the running time overlap duration is greater than zero, the predicted load values ​​of the two loads within the overlap time are added to obtain the power superposition value, and the power superposition value is compared with the allowable load value of the corresponding power supply branch. When the power superposition value is greater than the allowable load value, the corresponding load identifier, overlap time, power superposition value, allowable load value, and power supply branch are recorded to obtain the selection mutual exclusion identification result.

[0060] Next, when performing three-phase imbalance simulation for the intermediate power supply selection strategy based on the load coordination relationship sequence, the predicted load value and the phase of load connection for each load are read in the order of prediction time. The predicted load values ​​of the loads connected to phases A, B, and C are accumulated respectively to obtain the load values ​​of phase A, phase B, and phase C. The load values ​​of phase A, phase B, and phase C are added together and divided by 3 to obtain the three-phase average load value. The absolute values ​​of the differences between the load values ​​of phase A, phase B, and phase C and the three-phase average load value are calculated respectively, and the maximum value is selected as the maximum phase difference value. The maximum phase difference value is divided by the three-phase average load value to obtain the three-phase imbalance rate. When the three-phase imbalance rate is greater than the allowable three-phase imbalance rate pre-stored in the intermediate power supply selection strategy, the corresponding prediction time point, the load values ​​of phase A, phase B, phase C, the three-phase average load value, and the three-phase imbalance rate are recorded and arranged in the order of prediction time to obtain the three-phase imbalance path.

[0061] Finally, when optimizing and compensating the intermediate power supply selection strategy based on the mutual exclusion identification results and the three-phase imbalance path, the load identifier, overlap time, power superposition value, allowable load value, and power supply branch in the mutual exclusion identification results are read. The allowable load value is subtracted from the power superposition value to obtain the excess load value. Loads with predicted load values ​​greater than or equal to the excess load value are selected from the overlapping loads as loads to be adjusted. The allowable load values ​​of other power supply branches in the target power consumption area and the predicted load values ​​of the connected loads are read. The predicted load values ​​of the connected loads are subtracted from the allowable load values ​​to obtain the remaining allowable load value. When the remaining allowable load value is greater than or equal to the predicted load value of the load to be adjusted, the load to be adjusted is transferred to the corresponding power supply branch. Next, read the load values ​​of phase A, phase B, phase C, and the average load value of the three-phase imbalance path. Determine the phase with the largest load value as the transfer-out phase and the phase with the smallest load value as the transfer-in phase. Select loads whose predicted load value is not greater than the difference between the load value of the transfer-out phase and the average load value of the three phases from the loads connected to the transfer-out phase, and connect the loads to the transfer-in phase. Recalculate the load values ​​of phase A, phase B, phase C, the average load value of the three phases, and the three-phase imbalance rate until the three-phase imbalance rate is less than or equal to the allowable three-phase imbalance rate. Write the adjusted power supply branch, load connection phase, power distribution equipment parameters, electrical equipment parameters, and protection equipment parameters into the power supply selection intermediate strategy, and output the power supply selection target strategy.

[0062] By using the above optimization and compensation methods based on load coordination and three-phase imbalance analysis, load mutual exclusion conflicts and three-phase imbalance problems can be effectively identified, load can be reasonably allocated and phase optimized, the risk of overload in power supply branches can be reduced, power distribution reliability and power quality can be improved, and a basis can be provided for load access and capacity expansion.

[0063] In summary, the embodiments of this application have at least the following technical effects: This application performs time-series evolution analysis on load operation data of the target power consumption area to obtain load change trend sequences, startup impact characteristic sequences, and load coordination relationship sequences; performs aging dynamic prediction based on the power supply configuration strategy of the target power consumption area to establish a power supply aging dynamic map; performs multi-level dynamic adjustment of the power supply configuration strategy according to the load change trend sequence and the power supply aging dynamic map to obtain a power supply selection optimization strategy, which includes a distribution selection optimization strategy, an electrical selection optimization strategy, and a protection selection optimization strategy; performs startup impact dynamic compensation on the power supply selection optimization strategy according to the startup impact characteristic sequence to obtain an intermediate power supply selection strategy, which includes hierarchical correlation compensation on the power supply selection optimization strategy according to the startup impact characteristic sequence; performs mutual exclusion identification and three-phase balance optimization on the intermediate power supply selection strategy according to the load coordination relationship sequence to obtain a target power supply selection strategy, and executes the power supply configuration optimization deployment of the target power consumption area according to the target power supply selection strategy. This invention addresses the technical problem of low matching accuracy between the selection of electrical equipment and actual load requirements in the prior art. By performing dynamic analysis and selection optimization based on load operation data and power supply configuration strategies, it achieves the technical effect of improving the accuracy of electrical equipment selection and the adaptability of power supply configuration.

[0064] Example 2, based on the same inventive concept as the equipment selection optimization method based on load demand matching in the foregoing examples, such as... Figure 3 As shown, this application provides a power equipment selection optimization system based on load demand matching. The system and method embodiments in this application are based on the same inventive concept. The system includes: Evolutionary analysis module 11 is used to perform time-series evolution analysis based on the load operation data of the target power consumption area to obtain the load change trend sequence, the start-up impact characteristic sequence, and the load coordination relationship sequence; dynamic prediction module 12 is used to perform aging dynamic prediction based on the power supply configuration strategy of the target power consumption area and establish a power supply aging dynamic map; dynamic adjustment module 13 is used to perform multi-level dynamic adjustment of the power supply configuration strategy based on the load change trend sequence and the power supply aging dynamic map to obtain a power supply selection optimization strategy, which includes a distribution selection optimization strategy, an electrical selection optimization strategy, and a protection selection optimization strategy; dynamic compensation module 14 is used to perform start-up impact dynamic compensation on the power supply selection optimization strategy based on the start-up impact characteristic sequence to obtain an intermediate power supply selection strategy, which includes hierarchical correlation compensation on the power supply selection optimization strategy based on the start-up impact characteristic sequence; strategy execution module 15 is used to perform mutual exclusion identification and three-phase balance optimization on the intermediate power supply selection strategy based on the load coordination relationship sequence to obtain a power supply selection target strategy, and execute the power supply configuration optimization deployment of the target power consumption area according to the power supply selection target strategy.

[0065] Furthermore, the system is also used to implement the following functions: Based on the load operation data, construct the time-series operation curves for each load; predict the time-series evolution of each load operation curve based on the future window to obtain the load change trend sequence; extract features from the load change trend sequence based on the startup impact characteristic index to obtain the startup impact characteristic sequence; extract features from the load change trend sequence based on the load coordination characteristic index to obtain the load coordination relationship sequence.

[0066] Furthermore, the system is also used to implement the following functions: According to the power supply configuration strategy, multi-device dynamic monitoring is performed to obtain monitoring data of multiple devices; aging-related attention allocation is performed on the monitoring data of multiple devices to establish multiple device monitoring prominence vectors; machine learning is performed on the aging logs of multiple devices to generate an aging risk prediction model; the multiple device monitoring prominence vectors are input into the aging risk prediction model to obtain multiple aging risk prediction results, and topology association binding is performed on the multiple aging risk prediction results to generate the power supply aging dynamic map.

[0067] Furthermore, the system is also used to implement the following functions: The power supply configuration strategy is disassembled to obtain the current power distribution selection scheme, the current electrical selection scheme, and the current protection selection scheme. The power supply aging dynamic graph is disassembled at multiple levels to obtain the power distribution aging dynamic layer, the electrical aging dynamic layer, and the protection aging dynamic layer. Based on the load change trend sequence and the power distribution aging dynamic layer, the current power distribution selection scheme is expanded and reconstructed, and the power distribution selection optimization strategy is output. Based on the load change trend sequence and the electrical aging dynamic layer, the current electrical selection scheme is overload analyzed and reconstructed, and the electrical selection optimization strategy is output. Based on the load change trend sequence and the protection aging dynamic layer, the current protection selection scheme is reconstructed using protection trigger association, and the protection selection optimization strategy is output.

[0068] Furthermore, the system is also used to implement the following functions: Extract the inflection point of phased load growth and the peak load migration trajectory based on the load change trend sequence; perform capacity expansion compensation adjustment on the current power distribution selection scheme based on the inflection point of phased load growth and the peak load migration trajectory to obtain the basic power distribution regulation strategy; perform sensitivity analysis on the power distribution aging dynamic layer based on the aging risk threshold to obtain the power distribution aging sensitive topology; perform compensation reconstruction on the basic power distribution regulation strategy based on the power distribution aging sensitive topology to generate the power distribution selection optimization strategy.

[0069] Furthermore, the system is also used to implement the following functions: Based on the load change trend sequence and the current protection selection scheme, protection trigger fitting is performed to obtain protection trigger fitting data; based on the protection trigger fitting data, anomaly identification is performed to obtain protection anomaly pre-simulation features, which include protection sensitivity attenuation features, tripping delay drift features, and upper and lower level protection coordination offset features; based on the protection anomaly pre-simulation features, the current protection selection scheme is correlated and compensated to obtain the protection adjustment basic strategy; based on the aging risk threshold, the protection aging dynamic layer is sensitive to analysis to obtain the protection aging sensitive topology; based on the protection aging sensitive topology, the protection adjustment basic strategy is corrected to generate the protection selection optimization strategy.

[0070] Furthermore, the system is also used to implement the following functions: The intermediate power supply selection strategy is mutually exclusive based on the load coordination relationship sequence to obtain the mutual exclusion identification result; the intermediate power supply selection strategy is three-phase imbalance deduction based on the load coordination relationship sequence to obtain the three-phase imbalance path; the intermediate power supply selection strategy is optimized and compensated based on the mutual exclusion identification result and the three-phase imbalance path to output the target power supply selection strategy.

[0071] Furthermore, the system is also used to implement the following functions: The starting impact characteristic indicators include the peak value of the instantaneous starting current, the starting duration, the starting frequency, the continuous starting interval, and the starting impact energy.

[0072] Furthermore, the system is also used to implement the following functions: The load coordination characteristic indicators include load running time overlap characteristics, load start-stop correlation characteristics, and load power superposition characteristics.

[0073] It should be noted that the order of the embodiments described above is merely for descriptive purposes and does not represent the superiority or inferiority of the embodiments. Furthermore, the above description focuses on specific embodiments of this specification. The processes depicted in the accompanying drawings do not necessarily require a specific or sequential order to achieve the desired results. In some implementations, multitasking and parallel processing are possible or may be advantageous.

[0074] The above description is merely a preferred embodiment of the present invention and is not intended to limit the present invention in any way. Although the present invention has been disclosed above with reference to preferred embodiments, it is not intended to limit the present invention. Any person skilled in the art can make some modifications or alterations to the above-disclosed technical content to create equivalent embodiments without departing from the scope of the present invention. Any modifications, equivalent changes, and alterations made to the above embodiments based on the technical essence of the present invention without departing from the scope of the present invention shall still fall within the scope of the present invention.

Claims

1. A method for optimizing the selection of electrical equipment based on load demand matching, characterized in that, The method includes: Based on the load operation data of the target power consumption area, time-series evolution analysis is performed to obtain the load change trend sequence, the start-up impact characteristic sequence, and the load coordination relationship sequence; Based on the power supply configuration strategy of the target power consumption area, dynamic aging prediction is performed, and a dynamic power supply aging map is established. The power supply configuration strategy is dynamically adjusted in multiple levels based on the load change trend sequence and the power supply aging dynamic map to obtain a power supply selection optimization strategy, which includes a power distribution selection optimization strategy, an electrical selection optimization strategy and a protection selection optimization strategy. Dynamic compensation for startup impact is performed on the power supply selection optimization strategy based on the startup impact feature sequence to obtain an intermediate power supply selection strategy. The dynamic compensation for startup impact includes hierarchical correlation compensation of the power supply selection optimization strategy based on the startup impact feature sequence. Based on the load coordination relationship sequence, the intermediate power supply selection strategy is mutually exclusive and three-phase balance is optimized to obtain the target power supply selection strategy, and the power supply configuration optimization deployment of the target power consumption area is executed according to the target power supply selection strategy.

2. The method for optimizing the selection of electrical equipment based on load demand matching as described in claim 1, characterized in that, Based on the load operation data of the target power consumption area, time-series evolution analysis is performed to obtain the load change trend sequence, the start-up impact characteristic sequence, and the load coordination relationship sequence, including: Based on the load operation data, construct the time-series operation curves for each load; Based on the future window, the time-series operating curves of each load are predicted to undergo time-series evolution, thereby obtaining the load change trend sequence. The load change trend sequence is feature extracted based on the start-up impact characteristic indicators to obtain the start-up impact characteristic sequence; The load coordination relationship sequence is obtained by extracting features from the load change trend sequence based on the load coordination characteristic indicators.

3. The method for optimizing the selection of electrical equipment based on load demand matching as described in claim 1, characterized in that, Based on the power supply configuration strategy of the target power consumption area, dynamic aging prediction is performed to establish a power supply aging dynamic map, including: Perform dynamic monitoring of multiple devices according to the power supply configuration strategy, and obtain monitoring data of multiple devices; Aging-related attention allocation is performed on the monitoring data of the multiple devices to establish multiple device monitoring salience vectors; Machine learning is performed on aging logs from multiple devices to generate an aging risk prediction model. The multiple device monitoring prominence vectors are input into the aging risk prediction model to obtain multiple aging risk prediction results. Topological association binding is then performed on the multiple aging risk prediction results to generate the power supply aging dynamic map.

4. The method for optimizing the selection of electrical equipment based on load demand matching as described in claim 1, characterized in that, The power supply configuration strategy is dynamically adjusted at multiple levels based on the load change trend sequence and the power supply aging dynamic map to obtain a power supply selection optimization strategy. This power supply selection optimization strategy includes a distribution selection optimization strategy, an electrical selection optimization strategy, and a protection selection optimization strategy, including: The power supply configuration strategy is disassembled to obtain the current power distribution selection scheme, the current electrical selection scheme, and the current protection selection scheme; The power supply aging dynamic map is decomposed into multiple layers to obtain the power distribution aging dynamic layer, the electrical aging dynamic layer, and the protection aging dynamic layer. Based on the load change trend sequence and the power distribution aging dynamic layer, the current power distribution selection scheme is expanded and reconstructed, and the power distribution selection optimization strategy is output. Based on the load change trend sequence and the electrical aging dynamic layer, the current electrical selection scheme is overload analyzed and reconstructed, and the electrical selection optimization strategy is output. Based on the load change trend sequence and the protection aging dynamic layer, the current protection selection scheme is reconstructed by protection trigger association, and the protection selection optimization strategy is output.

5. The method for optimizing the selection of electrical equipment based on load demand matching as described in claim 4, characterized in that, Based on the load change trend sequence and the power distribution aging dynamic layer, the current power distribution selection scheme is expanded and reconstructed, and the power distribution selection optimization strategy is output, including: Extract the inflection point of phased load growth and the peak load migration trajectory based on the load change trend sequence; Based on the inflection point of the stage load growth and the peak load migration trajectory, the current power distribution selection scheme is expanded and compensated to obtain the basic strategy for power distribution regulation. Sensitivity analysis is performed on the dynamic layer of power distribution aging based on the aging risk threshold to obtain the power distribution aging sensitive topology. The power distribution regulation basic strategy is compensated and reconstructed based on the power distribution aging sensitive topology to generate the power distribution selection optimization strategy.

6. The method for optimizing the selection of electrical equipment based on load demand matching as described in claim 4, characterized in that, Based on the load change trend sequence and the protection aging dynamic layer, the current protection selection scheme is reconstructed using protection trigger association, and the protection selection optimization strategy is output, including: Based on the load change trend sequence and the current protection selection scheme, protection trigger fitting is performed to obtain protection trigger fitting data; Anomaly identification is performed based on the protection trigger fitting data to obtain protection anomaly pre-simulation features, which include protection sensitivity attenuation features, trip delay drift features, and upper and lower level protection coordination offset features. Based on the protection anomaly pre-simulation characteristics, the current protection selection scheme is correlated and compensated to obtain the basic protection adjustment strategy; Sensitivity analysis is performed on the protective aging dynamic layer based on the aging risk threshold to obtain the protective aging sensitive topology; Based on the protection aging-sensitive topology, the protection adjustment basic strategy is modified to generate the protection selection optimization strategy.

7. The method for optimizing the selection of electrical equipment based on load demand matching as described in claim 1, characterized in that, Based on the load coordination relationship sequence, the intermediate power supply selection strategy is subjected to mutual exclusion identification and three-phase balance optimization to obtain the target power supply selection strategy, including: Based on the load coordination relationship sequence, mutual exclusion identification is performed on the intermediate power supply selection strategy to obtain the mutual exclusion identification result. Based on the load coordination relationship sequence, the three-phase imbalance is deduced from the intermediate power supply selection strategy to obtain the three-phase imbalance path; Based on the mutual exclusion identification results and the three-phase imbalance path, the intermediate power supply selection strategy is optimized and compensated, and the target power supply selection strategy is output.

8. The method for optimizing the selection of electrical equipment based on load demand matching as described in claim 2, characterized in that, The starting impact characteristic indicators include the peak value of the instantaneous starting current, the starting duration, the starting frequency, the continuous starting interval, and the starting impact energy.

9. The method for optimizing the selection of electrical equipment based on load demand matching as described in claim 2, characterized in that, The load coordination characteristic indicators include load running time overlap characteristics, load start-stop correlation characteristics, and load power superposition characteristics.

10. A power equipment selection optimization system based on load demand matching, characterized in that, The system is used to execute the equipment selection optimization method based on load demand matching as described in any one of claims 1-9, and the system includes: The evolution analysis module is used to perform time-series evolution analysis based on the load operation data of the target power consumption area to obtain the load change trend sequence, the start-up impact characteristic sequence, and the load coordination relationship sequence. The dynamic prediction module is used to perform dynamic aging prediction based on the power supply configuration strategy of the target power consumption area and establish a dynamic power supply aging map. The dynamic adjustment module is used to perform multi-level dynamic adjustment of the power supply configuration strategy according to the load change trend sequence and the power supply aging dynamic spectrum, and obtain the power supply selection optimization strategy, which includes the power distribution selection optimization strategy, the electrical selection optimization strategy and the protection selection optimization strategy. The dynamic compensation module is used to perform dynamic compensation for the power supply selection optimization strategy based on the startup impact characteristic sequence, and to obtain the intermediate power supply selection strategy. The startup impact dynamic compensation includes hierarchical correlation compensation based on the startup impact characteristic sequence. The strategy execution module is used to perform mutual exclusion identification and three-phase balance optimization on the intermediate power supply selection strategy according to the load coordination relationship sequence, obtain the target power supply selection strategy, and execute the power supply configuration optimization deployment of the target power consumption area according to the target power supply selection strategy.

Images