An outdoor workstation emergency energy intelligent scheduling method and system

Abstract

The application discloses an outdoor workstation emergency energy intelligent scheduling method and system, generates an energy state parameter set through collecting multi-source energy data and load operation data, executes a margin interval division to identify a power supply margin level and performs gradient analysis to generate a load response coefficient according to the parameter set, and constructs a power supply switching threshold table; performs coupling analysis on the energy state parameter set to determine an energy complementary coefficient, identifies a composite power supply feature, and dynamically adjusts the threshold table to generate a scheduling compensation parameter; performs priority check on the scheduling compensation parameter to identify a power supply constraint condition, determines an effective switching action, and generates a switching scheduling sequence; and determines a scheduling priority weight according to an emergency output parameter and a constraint condition, and outputs an energy scheduling instruction. The application improves the accuracy and response efficiency of power supply scheduling, enhances the adaptability of multi-energy collaborative power supply, and improves the reliability of emergency scheduling.

Description

Technical Field

[0001] This invention relates to the field of energy management technology, and in particular to an intelligent emergency energy dispatching method and system for outdoor workstations. Background Technology

[0002] Outdoor workstations are typically equipped with multiple power sources, including solar power, wind power, energy storage, and mains power, to meet operational needs. Traditional power monitoring methods mainly focus on the output status of a single power source, lacking a comprehensive analysis of the dynamic balance between multiple power sources and electrical equipment. When external conditions change or load fluctuates, it is difficult to assess the supply and demand situation and power reserve level in a timely manner. At the same time, although different power sources have time-complementary characteristics, they are often controlled independently, failing to identify the synergistic relationship between power sources, resulting in a lack of specificity in the combined power supply scheme.

[0003] Load management in emergency scenarios also has shortcomings. When power supply capacity is limited, a simple sequential power outage strategy is often adopted, which does not fully consider the differences in equipment importance and state limitations, and may lead to power outages or frequent operation of critical equipment. Therefore, an intelligent scheduling method is needed to solve at least one of the above problems. Summary of the Invention

[0004] This invention discloses an intelligent emergency energy dispatching method and system for outdoor workstations. It aims to comprehensively analyze the dynamic relationship between multiple energy sources and load operation, and integrate power supply coordination characteristics and load priority rules to comprehensively and accurately identify and map key dispatching elements such as power supply margin, margin gradient, and energy complementarity relationship. This reveals the coordination rules between power sources and the constraint relationship between loads, and ultimately forms a multi-level and differentiated dispatching decision output, providing accurate and reliable dispatching solutions for application scenarios such as emergency power supply guarantee, energy optimization configuration, and intelligent load management of outdoor workstations.

[0005] The first aspect of this invention proposes an intelligent emergency energy dispatching method for outdoor workstations, comprising the following steps: Collect multi-source energy data and load operation data from outdoor workstations, and perform time-series alignment and feature extraction processing on the multi-source energy data and load operation data to generate an energy status parameter set; Based on the energy state parameter set, a margin interval is divided to identify the power margin level. Margin gradient analysis is performed along the power margin level to generate a load response coefficient. A power supply switching threshold table is constructed by mapping the load response coefficient. Energy coupling analysis is performed on the energy state parameter set to determine the energy complementarity coefficients of different energy sources. The composite power supply characteristics are identified through the energy complementarity coefficients. Based on the composite power supply characteristics, the power supply switching threshold table is dynamically adjusted to generate scheduling compensation parameters. The load priority verification is performed on the scheduling compensation parameters to identify power supply constraints. Based on the power supply constraints, an emergency margin determination is performed to determine valid switching actions. Priority parameters are extracted from the valid switching actions to generate a switching scheduling sequence. The power supply switching threshold table is associated and matched with the scheduling compensation parameters to generate emergency output parameters. The scheduling priority weight is determined based on the emergency output parameters and the power supply constraints. The energy scheduling command is output using the scheduling priority weight in combination with the switching scheduling sequence.

[0006] A second aspect of this invention provides an intelligent emergency energy dispatching system for outdoor workstations, comprising: The data acquisition module is used to collect multi-source energy data and load operation data from the outdoor workstation, and to perform time-series alignment and feature extraction processing on the multi-source energy data and the load operation data to generate an energy status parameter set. The status assessment module is used to perform margin interval division to identify power margin levels based on the energy status parameter set, perform margin gradient analysis along the power margin levels to generate load response coefficients, and construct a power supply switching threshold table by mapping the load response coefficients. The coupling analysis module is used to perform energy coupling analysis on the energy state parameter set to determine the energy complementarity coefficient of different energy sources, identify composite power supply characteristics through the energy complementarity coefficient, and dynamically adjust the power supply switching threshold table based on the composite power supply characteristics to generate scheduling compensation parameters. The timing scheduling module is used to perform load priority verification on the scheduling compensation parameters to identify power supply constraints, perform emergency margin determination based on the power supply constraints to determine effective switching actions, and extract priority parameters from the effective switching actions to generate a switching scheduling sequence. The instruction generation module is used to associate and match the power supply switching threshold table with the scheduling compensation parameters to generate emergency output parameters, determine the scheduling priority weight based on the emergency output parameters and the power supply constraints, and output energy scheduling instructions using the scheduling priority weight in combination with the switching scheduling sequence.

[0007] The beneficial effects of this invention are reflected in the following points: 1. A dynamic correlation mechanism between power supply margin and load demand is established. By implementing gradient mutation detection on the margin change curve to identify power supply risk periods, the characteristics of margin mutation are correlated with load patterns to obtain load type weights and map them to load response coefficients. A differentiated power supply switching threshold table is constructed, realizing the transformation from passive monitoring to proactive early warning scheduling mode, improving the accuracy and response efficiency of power supply scheduling. 2. A method for quantifying energy complementarity coefficients is proposed. Energy types are identified and switching efficiency is evaluated by extracting time-frequency domain characteristics. Based on time-series complementarity characteristics and switching losses, topology path analysis is performed to quantify the synergistic relationship between different energy sources. The switching threshold table is dynamically adjusted according to the composite power supply characteristics, breaking through the limitations of traditional independent multi-energy control and enhancing the adaptability of multi-energy coordinated power supply. 3. An emergency dispatch mechanism integrating load priority and margin determination was constructed. Conflicting actions were identified and combined by analyzing the task status of load-related nodes. Effective switching actions were screened based on the over-limit identification of the power transfer margin critical point. A systematic priority control system that considers the differences in equipment importance and status limitations was formed, avoiding the risk of power loss of critical equipment caused by simple sequential power outages and improving the reliability of emergency dispatch. Attached Figure Description

[0008] The accompanying drawings illustrate specific examples of the technical solutions described in this invention and, together with the detailed embodiments, form part of the specification, serving to explain the technical solutions, principles, and effects of this invention.

[0009] Figure 1 This is a flowchart illustrating an intelligent emergency energy dispatching method for outdoor workstations according to the present invention.

[0010] Figure 2 This is a structural block diagram of an outdoor workstation emergency energy intelligent dispatching system according to the present invention. Detailed Implementation

[0011] In the following description, specific details such as particular system architectures and techniques are set forth for illustrative purposes and not for limitation, in order to provide a thorough understanding of the embodiments of this application. However, those skilled in the art will understand that this application may also be implemented in other embodiments without these specific details. In other instances, detailed descriptions of well-known systems, apparatuses, circuits, and methods have been omitted so as not to obscure the description of this application with unnecessary detail.

[0012] It should be understood that, when used in this application specification and the appended claims, the term "comprising" indicates the presence of the described features, integrals, steps, operations, elements and / or components, but does not exclude the presence or addition of one or more other features, integrals, steps, operations, elements, components and / or a collection thereof.

[0013] References to "one embodiment" or "some embodiments" as described in this specification mean that one or more embodiments of this application include a specific feature, structure, or characteristic described in connection with that embodiment. Therefore, the phrases "in one embodiment," "in some embodiments," "in other embodiments," "in still other embodiments," etc., appearing in different parts of this specification do not necessarily refer to the same embodiment, but rather mean "one or more, but not all, embodiments," unless otherwise specifically emphasized. The terms "comprising," "including," "having," and variations thereof mean "including but not limited to," unless otherwise specifically emphasized.

[0014] The technical solutions of the embodiments of this application will be described below.

[0015] like Figure 1 As shown, this embodiment of the invention provides an intelligent emergency energy dispatching method for outdoor workstations, including the following steps S110-S150: Step S110: Collect multi-source energy data and load operation data from the outdoor workstation, and perform time-series alignment and feature extraction processing on the multi-source energy data and load operation data to generate an energy state parameter set.

[0016] Specifically, multi-source energy data and load operation data are collected from the outdoor workstation. Multi-source energy data includes photovoltaic (PV) power generation data, wind power generation data, energy storage battery data, and mains power connection data. Each type of data records four parameters: voltage, current, power, and timestamp. The PV power generation data acquisition cycle is 1 second, with an accuracy of voltage ±0.1V and current ±0.01A. In clear weather, the outdoor workstation's PV array outputs a voltage of 48V and a current fluctuating between 5A and 12A. Wind power generation data is obtained from the wind turbine controller via the Modbus protocol, with an acquisition cycle of 1 second. Energy storage battery data includes SOC (State of Charge) and SOH (State of Health) parameters, with an acquisition cycle of 2 seconds. SOC is used to calculate remaining capacity, and SOH is used to assess the degree of performance degradation. Mains power connection data is measured using a three-phase electricity meter, with an acquisition cycle of 5 seconds. Load operation data records the power consumption status of four types of equipment: lighting load, air conditioning load, communication load, and charging load. The lighting load power ranges from 200W to 800W, adjusted according to the lighting conditions. The air conditioning load consumes up to 3000W during high-temperature periods. The communication load has a steady-state power consumption of 500W, and the charging load power is dynamically adjusted from 1000W to 5000W. Multi-source energy data and load operation data are aggregated to the edge computing gateway via a fieldbus, with the data transmission latency of the fieldbus controlled within 50 milliseconds.

[0017] A set of energy state parameters is generated by time-series alignment and feature extraction of multi-source energy data and load operation data. The four types of energy data from the multi-source sources have different acquisition cycles: photovoltaic and wind power data have a 1-second cycle, energy storage battery data has a 2-second cycle, and grid connection data has a 5-second cycle. Time-series alignment uses a unified 10-second reference cycle to align the multi-source energy data with different cycles to the same time grid. The four types of load operation data have an acquisition time delay of 100 to 150 milliseconds compared to the multi-source energy data. Time-series alignment compensates for this delay to a unified reference through timestamp calibration. In the multi-source energy data, photovoltaic power contains 10 sampling points within a 10-second window. Time-series alignment uses a window averaging method to obtain a representative power value of 520W for this window. Feature extraction calculates the fluctuation coefficient CV = σ / μ from the aligned multi-source energy data to quantify energy output stability, where σ is the standard deviation of the power values ​​within the 10-second window, and μ is the arithmetic mean of the power values ​​within the window. The fluctuation coefficient is used for subsequent power margin stability assessment. Feature extraction extracts the load demand change rate from load operation data to reflect the power consumption fluctuation trend. The load demand change rate is used for subsequent margin change prediction. The energy status parameter set integrates seven parameters: energy type identifier, real-time power value, remaining capacity, fluctuation coefficient, various load demand values, load demand change rate, and timestamp.

[0018] Step S120: Based on the energy state parameter set, perform margin interval division to identify power margin levels, perform margin gradient analysis along the power margin levels to generate load response coefficients, and construct a power supply switching threshold table by mapping the load response coefficients.

[0019] Specifically, the power margin level is identified by dividing the margin range based on the energy state parameter set. The energy state parameter set shows that the current total power supply capacity is 13820W, the total load demand is 4200W, and the power supply margin is 9620W. The margin range is divided into four ranges according to the percentage of load demand: sufficient, normal, strained, and critical. The sufficient range corresponds to a margin greater than 150% of the load demand, the normal range corresponds to 100% to 150%, the strained range corresponds to 50% to 100%, and the critical range corresponds to less than 50%. The power supply margin of 9620W in the energy state parameter set accounts for 229% of the load demand, and at this moment it is classified as sufficient. The sufficient range triggers the power margin level assessment as high margin. The power margin level is determined jointly by the margin range and the remaining capacity of the energy state parameter set. The remaining capacity reflects the sustainable power supply capacity of the energy storage. The energy state parameter set shows that the remaining energy storage capacity is 32.5kWh, and the sustainable power supply duration calculated based on the current load demand is 7.7 hours. A sustainable power supply duration greater than 6 hours is defined as high margin. Power margin levels are divided into four categories: high margin, medium margin, low margin, and zero margin. High margin corresponds to a sufficient or normal range with a duration greater than 6 hours; medium margin corresponds to a normal or strained range with a duration of 3 to 6 hours; low margin corresponds to a strained or critical range with a duration of 1 to 3 hours; and zero margin corresponds to a critical range with a duration of less than 1 hour. The power margin level is combined with the fluctuation coefficient of the energy state parameter set for stability verification. A fluctuation coefficient greater than 0.4 indicates unstable energy output, and the required energy storage reserve capacity is increased to 50% during power margin level assessment to cope with rapid fluctuations.

[0020] In some embodiments, the step of generating load response coefficients by performing margin gradient analysis along the power margin level includes: extracting cumulative margin records based on the power margin level to construct a margin change curve; performing gradient abrupt change detection on the margin change curve to generate a margin abrupt change early warning identifier; performing load mode association based on the margin abrupt change early warning identifier to obtain load type weights; and using the load type weights to perform response intensity mapping on the margin abrupt change early warning identifier to generate load response coefficients.

[0021] Based on the power margin level, cumulative margin records were extracted to construct a margin change curve. The margin values ​​from the power margin level records were compiled into a time series over a continuous 24-hour period, containing 8640 margin values ​​in 10-second windows. The cumulative margin records statistically analyzed the duration and average margin value of each margin level: high margin level lasted 18 hours, medium margin lasted 4 hours, and low margin lasted 2 hours. The margin change curve plotted the trend of margin change over time with time as the horizontal axis and margin value as the vertical axis. The power margin level increased from 80% (medium margin) to 180% (high margin) at 6:00 AM, with a positive slope indicating rapid margin growth. The margin change curve reached its peak of 250% at 12:00 PM, when both photovoltaic and wind power generation were at high output. The margin change curve for outdoor workstations showed an upward trend during the daytime, with photovoltaic output power gradually increasing as sunlight intensified. The margin change curve shows a downward inflection point at 18:00 in the evening, with the margin value dropping from 200% to 80%. This inflection point corresponds to the period when photovoltaic power is shut down and evening load demand increases. The power margin level drops to its lowest value of 50% at 22:00, entering the tense zone. The margin change curve exhibits a clear daily cyclical fluctuation pattern.

[0022] Gradient abrupt change detection is implemented on the margin change curve to generate a margin abrupt change warning indicator. The gradient of the margin change curve is defined as the difference between the margin values ​​at adjacent time points divided by the time interval. The unit of the gradient value is %margin / 10 seconds, and the gradient value reflects the rate of margin change. A positive gradient at 6:00 AM indicates an increase in margin, and a negative gradient at 6:00 PM indicates a decrease in margin. Gradient abrupt change detection identifies the location of gradient sign reversal or numerical jump by comparing adjacent gradient values. Abrupt change detection is triggered when the gradient changes from positive to negative or when the absolute value of the gradient increases by more than 5%margin / 10 seconds. When the gradient of the margin change curve changes from positive to negative at 6:00 PM, and the gradient change exceeds the threshold, this location is marked as a gradient abrupt change point, and the margin abrupt change warning indicator records the time position of this abrupt change point as 6:00 PM. Outdoor workstations experience rapid degradation of photovoltaic power generation during the evening hours, while construction lighting and charging equipment are simultaneously activated. This causes the power supply margin to drop from ample to strained in a short period, with the gradient abrupt change exceeding the detection threshold. The margin abrupt change warning indicator records the magnitude and direction of this abrupt change as decreasing. The margin change curve identified 12 gradient abrupt change points within 24 hours, including 4 upward abrupt changes and 8 downward abrupt changes. The higher number of downward abrupt changes reflects the frequent occurrence of power supply risks. Gradient abrupt change detection sets graded thresholds for abrupt change magnitude: mild, moderate, and severe abrupt changes correspond to gradient change ranges of 5-10, 10-20, and >20% margin / 10 seconds, respectively. The margin abrupt change warning indicator includes the time location, magnitude, and direction of the gradient abrupt change point.

[0023] Load type weights are obtained by associating load patterns based on margin mutation early warning indicators. The 12 mutation points of the margin mutation early warning indicators are time-aligned with the load demand values ​​in the energy state parameter set to identify the load operation mode at the time of the mutation point. The load demand corresponding to the mutation point at 18:00 in the evening jumps from 3000W to 6200W, with the 3200W increase mainly coming from the simultaneous start-up of air conditioning and charging loads. Load pattern association analysis shows the power consumption changes of the four types of loads within a 10-minute window before and after the mutation point: air conditioning load increases from 2000W to 3000W contributing a 1000W increase; charging load increases from 0W to 2000W contributing a 2000W increase; lighting load increases from 500W to 700W contributing a 200W increase; and communication load remains unchanged at 500W. Load type weight is defined as the proportion of the power consumption increment of that type of load to the total load increment. The highest weight is for charging load (2000 / 3200 = 0.625), followed by air conditioning load (1000 / 3200 = 0.313), then lighting load (200 / 3200 = 0.063), and communication load has a weight of zero. The load structure of outdoor workstations shifts from being dominated by power loads during the day to being dominated by lighting and charging loads at night during the evening. The time-period differences in load type weights reflect the diurnal variation of the work pattern. The load type weights for the 12 abrupt change warning points are calculated for each corresponding time point, and the weight distribution reveals the contribution of different loads to the margin change. Load pattern correlation statistics show the frequency of margin changes triggered by various load types within 24 hours: charging load triggered 6 times, air conditioning load triggered 4 times, lighting load triggered 1 time, and communication load triggered 0 times. The load type weights are weighted averaged according to the trigger frequency, with charging load having the highest average weight, followed by air conditioning load, and then lighting and communication loads.

[0024] Load response coefficients are generated by mapping the response intensity of margin mutation warning indicators using load type weights. The response intensity mapping combines the load type weights with the gradient amplitude of the margin mutation warning indicator. The response intensity calculation formula is R = W × |g|, where W is the load type weight (dimensionless), g is the gradient amplitude (in % margin / 10 seconds), and R is the response intensity (in % margin / 10 seconds). Outdoor workstations experience rapid photovoltaic degradation in the evening, leading to a sharp drop in power supply margin. The high weight of 0.625 and the large gradient amplitude of 15% margin / 10 seconds for charging loads together determine the response intensity of this load as 9.4% margin / 10 seconds. The response intensity of the 12 mutation points of the margin mutation warning indicator is calculated for four types of loads. Charging loads have the highest average response intensity over 24 hours, followed by air conditioning loads, while lighting and communication loads have lower average responses. The load response coefficient is defined as the normalized response intensity, which is achieved by dividing the response intensity of each type of load by the sum of the response intensities of the four types of loads. The load response coefficient is highest for charging loads, followed by air conditioning loads, while lighting and communication loads have lower response coefficients. The load response coefficient is used to determine the priority of load adjustments in power supply switching decisions; loads with higher response coefficients are prioritized for switching or load shedding. The time-of-day characteristics of the load response coefficient are influenced by both photovoltaic fluctuations and the remaining capacity of energy storage. The load response coefficient is trend-corrected by incorporating the load demand change rate from the energy state parameter set; when the load demand change rate is positive and greater than 20% / minute, the load response coefficient is increased by 10% to anticipate rapid load increases.

[0025] In some embodiments, constructing a power supply switching threshold table using the load response coefficient mapping includes: dividing the load response coefficient into time periods according to energy cycle characteristics to generate a time period response coefficient set; performing load state analysis on the time period response coefficient set to dynamically identify peak and valley time period feature identifiers; setting peak and valley time period switching boundaries based on the peak and valley time period feature identifiers and matching differentiated threshold benchmarks; and integrating the peak and valley time period switching boundaries to form a power supply switching threshold table.

[0026] Load response coefficients are divided into time periods based on energy cycle characteristics to generate time period response coefficient sets. Energy cycle characteristics are represented by the daily cycle of photovoltaic (PV) power generation and the random cycle of wind power generation. PV power generation is effective from 6:00 AM to 6:00 PM, and zero power generation is from 6:00 PM to 6:00 AM the next day. The 24-hour sequence of load response coefficients is divided into daytime and nighttime periods according to the PV power generation cycle. The load response coefficient during the daytime period is affected by PV fluctuations, while the load response coefficient during the nighttime period is constrained by energy storage capacity. The time period division further subdivides the daytime period into three sub-periods: a rising period from 6:00 AM to 12:00 PM, a peak period from 12:00 PM to 3:00 PM, and a falling period from 3:00 PM to 6:00 PM. The charging load coefficient gradually increases during the rising period, remains at a high level during the peak period, and begins to decline during the falling period. Outdoor workstations begin to start various devices during the rising period as sunlight intensifies, and the charging load response coefficient gradually increases from a low level. The time period response coefficient set statistically analyzes the average and standard deviation of the load response coefficients within each time period. The average response coefficient and standard deviation of the charging load during the rising period reflect the response characteristics of that period. Nighttime hours are divided into a high-capacity period from 6 PM to midnight (12 AM) and a low-capacity period from midnight to 6 AM (12 AM) based on remaining energy storage capacity. During the high-capacity period, the energy storage SOC (State of Charge) is above 50%, while during the low-capacity period, the SOC is between 30% and 50%. The time-period response coefficient set covers four types of load response coefficients across five time periods. The charging load coefficient of the time-period response coefficient set reaches its highest value during the peak period, when photovoltaic power is abundant, operational activities are concentrated, and charging demand is high.

[0027] For example, the step of performing load state analysis to dynamically identify peak and valley time period characteristics for the time period response coefficient set includes: performing coefficient distribution analysis on the time period response coefficient set to locate the coefficient distribution critical point; performing load state analysis on the coefficient distribution critical point to generate load state parameters; performing threshold response analysis on the load state parameters to generate threshold response coefficients; and performing time period calibration based on the threshold response coefficients to generate peak and valley time period characteristics.

[0028] Coefficient distribution analysis was performed on the time-period response coefficient set to locate the critical point of the coefficient distribution. The charging load coefficients of the time-period response coefficient set formed a numerical sequence across five time periods. Coefficient distribution analysis calculated the quantiles of the sequence to identify the central tendency and dispersion of the coefficients. The critical point of the coefficient distribution was defined as the midpoint between the 75th and 25th quantiles. Periods above the critical point were classified as high-response periods, and periods below the critical point were classified as low-response periods. The periods above the critical point in the time-period response coefficient set included peak periods, declining periods, and high-capacity periods, during which the charging load response coefficients were relatively high. Outdoor workstations had the most concentrated operational activities during peak periods, and their charging load response coefficients were significantly higher than those during low-response periods. The coefficient distribution critical point accurately delineated the boundary between peak and stable operational periods. The calculation method for the coefficient distribution critical point of air conditioning load was the same. Periods above the critical point were the rising period, peak period, declining period, and high-capacity period. During these periods, outdoor workstations were subject to strong temperature control requirements due to direct sunlight and equipment heat dissipation. Coefficient distribution analysis calculated the distribution critical point for each of the four load types. The critical point value reflected the average response level of that load type. The lowest critical point for communication load indicated that its response coefficient fluctuated the least across time periods.

[0029] Load state parameters are generated through load state analysis targeting the critical point of coefficient distribution. The critical point divides the charging load into five time periods: high-response and low-response periods. The high-response period includes peak, declining, and high-capacity periods. Load state analysis extracts the time-period response coefficient set data for the high-response period. Parameters such as the average charging load coefficient, average load demand, and average energy storage SOC reflect the characteristics of this period. The core indicator of the load state parameters is the ratio of load demand during the high-response period to that during the low-response period; a ratio greater than 1 indicates that the load demand during the high-response period is significantly higher than that during the low-response period. For outdoor workstations, the load demand during the high-response period is a multiple of that during the low-response period; the larger the ratio, the more significant the peak-valley difference, requiring more flexible power supply switching strategies. The high-response period for air conditioning load includes rising, peak, declining, and high-capacity periods. During the high-temperature days of the day, the air conditioning at outdoor workstations operates at full load, resulting in a significantly higher average load demand during the high-response period compared to the low-response period. Load status analysis calculates the standard deviation of load demand during high-response periods. The standard deviation reflects the degree of load fluctuation during this period. The standard deviation of charging load at outdoor workstations is larger due to the uncertainty of construction vehicle arrival times. Load status parameters integrate three dimensions: load demand ratio, standard deviation, and average SOC.

[0030] Threshold response analysis is performed on load state parameters to generate threshold response coefficients. The load demand ratio of charging load in the load state parameters represents the ratio of demand during high-response periods to demand during low-response periods. A larger ratio indicates a more significant peak-valley load difference. Threshold response analysis transforms this ratio into an adjustment coefficient for the switching threshold. The threshold response coefficient is defined as the logarithmically normalized value of the load state parameters. The logarithmic transformation maps the linear demand ratio to a non-linear response coefficient, making the impact of peak-valley differences smoother in threshold setting. For outdoor workstations with significant peak-valley load differences, a larger threshold response coefficient is used to guide the system to employ stricter margin monitoring during peak periods. The load state parameters of air conditioning load correspond to threshold response coefficients, and the degree of peak-valley difference in air conditioning load is reflected by the coefficient value. Threshold response analysis considers the standard deviation of the load state parameters. When the standard deviation is large, the threshold response coefficient is multiplied by a correction factor to reflect high volatility characteristics. For outdoor workstations, the standard deviation is large during periods of frequent start-stop of construction equipment, triggering correction and increasing the threshold response coefficient. When the average SOC of the load state parameters is low, the threshold response coefficient is multiplied by a decay factor to reduce the switching threshold during low-power periods, ensuring the system responds early when energy storage capacity is insufficient. The threshold response coefficient exhibits different values ​​at different times, with the coefficients differing for peak periods, declining periods, and high-capacity periods.

[0031] Peak and valley period characteristic identifiers are generated based on threshold response coefficients. The charging load column values ​​in the threshold response coefficient matrix are arranged by period. Periods with values ​​greater than a set threshold are identified as peak charging load periods, including peak periods, declining periods, and high-capacity periods. The period calibration divides the five periods into peak and valley periods according to the threshold response coefficients. Higher coefficients for peak periods indicate that the load is more sensitive to power supply changes, while lower coefficients for valley periods indicate lower power supply pressure. The peak and valley period characteristic identifiers record the peak and valley attributes of each period. Outdoor workstations face greater power supply pressure during peak periods, and the period calibration results trigger the system to adopt stricter power supply monitoring and faster switching responses. The periods with higher threshold response coefficients for air conditioning loads are the rising period, peak period, declining period, and high-capacity period. Outdoor workstations are most sensitive to power supply changes during the sustained high temperatures of the day and the evening when residual heat dissipates; these four periods are identified as peak air conditioning load periods. The time period labeling identifies time periods simultaneously marked as peak periods by multiple load types. Decreasing periods and high-capacity periods are simultaneously marked as peak periods by both charging and air conditioning loads. Outdoor workstations experience the greatest power supply pressure during the evening when photovoltaic degradation and charging peaks overlap; these periods are defined as comprehensive peak periods. Peak-valley time period characteristic identifiers include the peak-valley attributes of each time period and the corresponding threshold response coefficient.

[0032] The peak-valley time period switching boundaries are set based on the peak-valley time period characteristic identifiers and differentiated threshold benchmarks. Peak periods corresponding to peak periods in the peak-valley time period characteristic identifiers have higher threshold response coefficients. The differentiated threshold benchmarks set the power supply switching threshold for peak periods to be lower than that for valley periods, making it easier to trigger power supply switching during peak periods. Charging loads have higher threshold response coefficients during peak periods; the threshold benchmark is set to trigger switching when the margin is lower than 120% of load demand, while the valley period threshold benchmark is set to a margin lower than 150%. Outdoor workstations have concentrated work activities during peak periods; a lower switching threshold ensures that energy storage or mains power support is prioritized during peak charging periods. The peak-valley time period switching boundary is defined as the margin threshold value for triggering power supply switching. The peak period switching boundary is lower than the valley period switching boundary, and the boundary difference reflects the difference in switching sensitivity between peak and valley periods. The peak and valley period switching boundaries for air conditioning loads are set by comprehensively considering load importance and energy availability. The differentiated threshold benchmarks construct a threshold matrix according to load type and time period type, with matrix elements representing the percentage of the switching boundary. The switching boundary for the comprehensive peak period in the peak-valley time period characteristic identifiers is taken as the minimum value of the peak period boundaries for various load types, ensuring the highest switching sensitivity for the comprehensive peak period. In practical applications, the peak-valley switching boundary is dynamically adjusted based on the remaining capacity of the energy state parameter set. When the energy storage SOC is low, all switching boundaries are lowered to trigger power supply switching in advance.

[0033] The peak-valley switching boundaries are integrated to form a power supply switching threshold table. This table uses a matrix structure of 4 load categories × 5 time periods. The matrix rows represent charging loads, air conditioning loads, lighting loads, and communication loads. The matrix columns represent five time periods: rising period, peak period, falling period, high-capacity period, and low-capacity period. Matrix elements represent the percentage of the switching boundary for each load in each time period. For charging loads, the switching boundary is set higher during the rising period when photovoltaic power is sufficient to withstand larger load fluctuations. During the peak and falling periods, when operations are concentrated, the switching boundary is appropriately lowered to accelerate response. During the low-capacity period, when energy storage is limited, the switching boundary is lowered to the minimum to trigger power supply switching earlier. For communication loads, the switching boundary is set to the minimum in all time periods to ensure that communication equipment receives priority under any power supply conditions. The power supply switching threshold table includes an energy storage SOC correction parameter. When the SOC is high, the original boundary is maintained; when the SOC is at a medium level, all switching boundaries are reduced proportionally; when the SOC is below the safety threshold, the switching boundary is further reduced to trigger power supply switching earlier. The peak-valley switching boundary integration process performs consistency checks on the table data to ensure that the peak period boundary is lower than the valley period boundary. The power supply switching threshold table is updated every 24 hours based on the latest load response coefficient.

[0034] Step S130: Perform energy coupling analysis on the energy state parameter set to determine the energy complementarity coefficients of different energy sources, identify composite power supply characteristics through the energy complementarity coefficients, and dynamically adjust the power supply switching threshold table based on the composite power supply characteristics to generate scheduling compensation parameters.

[0035] In some embodiments, the step of performing energy coupling analysis on the energy state parameter set to determine the energy complementarity coefficients of different energy sources includes: extracting energy spectrum features based on the energy state parameter set in the time and frequency domain; performing pattern matching on the energy spectrum features to identify energy types; performing switching efficiency evaluation on the energy types to determine energy switching efficiency loss values; and performing topology path analysis based on the energy types and the energy switching efficiency loss values ​​to obtain the energy complementarity coefficients of different energy sources.

[0036] Energy spectral characteristics are obtained by extracting time-frequency domain features based on the energy state parameter set. The real-time photovoltaic (PV) power of the energy state parameter set forms a time series over 24 hours, containing 8640 power sampling points with 10-second windows. Time-frequency domain feature extraction performs a Fast Fourier Transform (FFT) on the power time series; the frequency domain amplitude of the transform reflects the power fluctuation intensity of each frequency component. The PV power spectrum of the energy state parameter set shows a main peak at the diurnal frequency, and the amplitude of the main peak reflects the diurnal fluctuation intensity of PV power. Energy spectral characteristics record the amplitude and frequency position of each frequency component. The PV spectral characteristics include three characteristic frequency bands: the diurnal main frequency, high-frequency fluctuation components, and low-frequency trend components. The PV power at the outdoor workstation is affected by cloud cover, resulting in multiple small peaks in the high-frequency band. The high-frequency components correspond to rapid power fluctuations, and the amplitude proportion of the high-frequency components is used to assess PV stability. The wind power spectral characteristics of the energy state parameter set are significantly different from those of PV. The wind spectrum distributes power components at multiple frequency positions, exhibiting a wideband characteristic, which is used to distinguish between wind and PV. The energy concentration of the power spectrum is extracted and calculated using time-frequency domain characteristics. Energy concentration is defined as the proportion of the dominant frequency amplitude to the total spectral energy. A photovoltaic energy concentration of 0.68 indicates that power is mainly concentrated at the daily cycle frequency. Energy concentration is used to quantify the strength of energy periodicity. The energy storage power spectrum characteristics of the energy state parameter set are affected by the charge-discharge control strategy, with characteristic peaks appearing at the charge-discharge switching frequency. Energy spectrum characteristics are characterized by three parameters: dominant frequency position, dominant frequency amplitude, and energy concentration. The combination of these three parameters constitutes the energy's frequency domain fingerprint for subsequent pattern matching.

[0037] For example, the step of performing pattern matching to identify energy types based on the energy spectrum features includes: performing frequency domain energy distribution analysis to obtain energy distribution features through the energy spectrum features; performing peak identification on the energy distribution features to generate a peak feature set and an energy performance degradation identifier; using the peak feature set and the energy performance degradation identifier to perform feature mapping to construct a distribution attribute set; and performing pattern library association matching to identify energy types based on the distribution attribute set.

[0038] Energy distribution characteristics are obtained through frequency domain energy distribution analysis using energy spectrum features. The frequency domain amplitude sequence of energy spectrum features contains multiple frequency sampling points within the frequency range of 0 to 0.5 Hz, with each sampling point corresponding to the amplitude magnitude at a specific frequency. Frequency domain energy distribution analysis calculates the energy value at each frequency point; the energy value is the square of the amplitude, and the total frequency domain energy is the sum of the energy values ​​at all frequency points. For photovoltaic (PV) energy, the energy proportion near the daily cycle frequency reaches 68%, with the remaining 32% distributed in the high-frequency and low-frequency regions. The energy distribution features record the energy proportion in different frequency bands, which are divided into low-frequency, mid-frequency, and high-frequency bands. The low-frequency band corresponds to slow changes, the mid-frequency band to periodic fluctuations, and the high-frequency band to rapid fluctuations. For outdoor workstations, PV energy has the highest proportion in the low-frequency band, corresponding to slow changes in sunlight, followed by the mid-frequency band, corresponding to power fluctuations caused by cloud cover. For wind energy, the energy distribution across the three frequency bands is relatively uniform, with each band's energy proportion ranging from 25% to 35%. This uniform distribution reflects the random fluctuation characteristics of wind power. Energy distribution characteristics are represented by a frequency band energy proportion vector, with a vector dimension of 3 corresponding to the three frequency bands, and the sum of the vector elements is 1. The energy storage type with energy spectrum characteristics has a higher energy proportion in the low-frequency and high-frequency bands. The low-frequency band corresponds to a long-term charge-discharge trend, while the high-frequency band corresponds to a power regulation response.

[0039] Peak identification is implemented based on energy distribution characteristics, generating peak feature sets and energy performance degradation indicators. The frequency band energy proportion curve of the energy distribution characteristics exhibits local maxima at certain frequency locations; these local maxima are defined as peak points. The peak identification algorithm traverses the energy proportion curve, identifying points where the energy is greater than adjacent frequency points and exceeds a threshold of 10% as peak points. For photovoltaic (PV) energy distribution characteristics, a main peak is identified near the daily cycle frequency, with the main peak energy proportion at 68%. The peak width reflects the stability of the daily cycle. The peak feature set records three parameters for each peak: frequency location, energy proportion, and peak width. The peak width is defined as the frequency range where the peak energy drops to half. In historical operating data, the main peak energy proportion of outdoor PV stations has decreased year by year; the current main peak of 68% is lower than the initial 75%, reflecting performance degradation of the PV modules. Energy performance degradation indicators are generated by comparing the current peak feature set with a baseline peak feature set. A difference in peak energy proportion exceeding 5% is marked as degradation, and this degradation indicator is used for capacity correction. Wind energy distribution characteristics contain multiple peak points; a large number of peaks indicates the multi-frequency fluctuation characteristics of wind power. The energy performance degradation indicator records the degree and type of degradation. The degree of degradation is divided into three levels: mild, moderate, and severe. The degradation types include three categories: output attenuation, increased fluctuation, and response delay. For energy storage types with energy distribution characteristics, an increase in the peak energy ratio in the high-frequency band indicates frequent charge-discharge switching, and frequent switching triggers the performance degradation indicator.

[0040] A distribution attribute set is constructed using peak feature sets and energy performance degradation identifiers for characteristic mapping. The main peak frequency position of the peak feature set maps to the energy's periodic characteristics, the daily periodic frequency maps to the diurnal cycle attribute, and the hourly periodic frequency maps to the short-cycle fluctuation attribute. The degradation degree of the energy performance degradation identifier maps to the reliability attribute; mild degradation corresponds to high reliability, and severe degradation corresponds to low reliability. The reliability attribute is used for energy dispatch priority assessment. The distribution attribute set integrates three dimensions: periodic characteristics, energy concentration characteristics, and reliability characteristics, providing a comprehensive three-dimensional description of the energy's frequency domain characteristics. The photovoltaic energy distribution attribute set of the outdoor workstation displays diurnal cycle attributes, high energy concentration characteristics, and medium reliability. The number of peaks in the peak feature set maps to the fluctuation complexity attribute; 1 to 2 peaks correspond to simple fluctuations, and more than 5 peaks correspond to complex fluctuations. The degradation type of the energy performance degradation identifier maps to the degradation mode attribute; output attenuation corresponds to capacity degradation, increased fluctuation corresponds to control degradation, and response delay corresponds to inertial degradation. The distribution attribute set is represented by an attribute vector with a dimension of 6, containing six attribute fields: period type, energy concentration, number of peaks, reliability level, degree of degradation, and degradation mode. The peak feature set and energy performance degradation identifier are converted into the distribution attribute set through a mapping rule base, which stores the correspondence between frequency domain feature parameters and attribute categories.

[0041] Energy types are identified through pattern library association matching based on a distribution attribute set. The six-dimensional attribute vector of the distribution attribute set is compared with the standard attribute vectors of the four energy types in the pattern library using a weighted Euclidean distance method. The photovoltaic standard attribute vector includes attributes such as diurnal cycle, energy concentration of 0.7, number of peaks of 1, and high reliability. When the weighted Euclidean distance between the distribution attribute set and the photovoltaic standard attribute vector is minimized, the pattern library association matching identifies the energy type as photovoltaic. An outdoor workstation's energy distribution attribute set shows diurnal cycle, energy concentration of 0.68, number of peaks of 1, and medium reliability. Its distance from the photovoltaic standard attribute vector is 0.15, and its matching confidence score is 0.92, confirming it as photovoltaic energy. The energy type identification result includes a type label and matching confidence score. Pattern library association matching calculates the similarity for each of the four energy types, selecting the type with the highest similarity as the identification result. Multi-type labeling is triggered when the similarity difference is less than a threshold of 0.1. The wind type in the distribution attribute set corresponds to an attribute combination with no obvious periodicity, dispersed energy, and a large number of peaks. The model library is updated regularly with standard attribute vectors based on actual operating data, with an update cycle of quarterly, to ensure that the model library reflects the actual operating characteristics of energy.

[0042] Energy switching efficiency loss values ​​are determined through energy type-based switching efficiency assessment. Energy type identification results show four energy types: solar, wind, energy storage, and grid power. Switching between different energy types involves circuit topology changes and power transmission path adjustments. The switching efficiency assessment calculates the power loss when switching from energy A to energy B, which includes switching device losses, line transmission losses, and converter losses. When an outdoor workstation switches from solar to energy storage, the solar inverter shuts off, and the bidirectional energy storage converter starts up. The switching process lasts approximately 200 milliseconds, during which power output decreases by 15%. The energy switching efficiency loss value is defined as the percentage of average power loss during the switching process relative to the power before switching; the switching efficiency loss value from solar to energy storage is 15%. The energy type switching efficiency loss value is related to the switching direction; the 15% loss value from solar to energy storage differs from the 12% loss value from energy storage to solar, due to the different starting characteristics of the converters. Switching from solar to wind power requires dual conversion by both the inverter and rectifier, resulting in a switching efficiency loss value of 18%. A 4×4 switching loss matrix is ​​constructed for switching efficiency assessment. The rows of the matrix represent the energy type before switching, the columns represent the energy type after switching, and the matrix elements represent the energy switching efficiency loss values. The switching loss value from mains power to energy storage is relatively low at 8%, as both mains power and energy storage are supplied by DC buses, and the switching only involves circuit breaker operation. The switching efficiency loss value is affected by the load power; the higher the load power, the greater the absolute value of the switching loss. The assessment process measures the loss value under rated load conditions. A diagonal element of 0 in the switching loss matrix indicates no switching loss between energy sources.

[0043] Based on energy type and energy switching efficiency loss value, topology path analysis is performed to obtain the energy complementarity coefficient of different energy sources. Topology path analysis constructs an energy supply topology graph, with nodes representing four energy types, edges representing switching paths between energy sources, and edge weights representing energy switching efficiency loss values. The energy complementarity coefficient M is calculated considering the temporal complementarity characteristics and switching efficiency of the two energy types. The formula is M=C×(1-L), where C is the absolute value of the temporal negative correlation coefficient (the higher the negative correlation, the stronger the complementarity), and L is the energy switching efficiency loss value. For the outdoor workstation, the temporal negative correlation coefficient between photovoltaic and wind power is -0.62, taking the absolute value of 0.62, with a switching efficiency loss value of 18%, and the energy complementarity coefficient is 0.62×0.82 equal to 0.51, indicating that photovoltaic and wind power have moderate complementarity. For the energy type energy storage, the temporal negative correlation coefficient with grid power is -0.35, taking the absolute value of 0.35, with a switching efficiency loss value of 8%, and the energy complementarity coefficient is 0.32, indicating that the complementarity between energy storage and grid power is relatively weak. Topology path analysis identifies energy pairs with high complementarity coefficients. The photovoltaic-wind power complementarity coefficient (0.51) is higher than the photovoltaic-grid complementarity coefficient (0.28), and the photovoltaic-wind power combination is prioritized for inclusion in the composite power supply scheme. The complementarity coefficients of other energy pairs are calculated using the same method based on their respective time-series negative correlation characteristics and switching efficiency loss values. The wind power-energy storage complementarity coefficient (0.38) and the wind power-grid complementarity coefficient (0.25) reflect the synergistic capabilities of wind power with energy storage and grid power, respectively. An energy complementarity coefficient matrix records the pairwise complementarity coefficients between the four energy types. The matrix is ​​symmetric, and the matrix elements range from 0 to 1. The impact of energy switching efficiency loss values ​​on the complementarity coefficients is reflected by the loss reduction factor (1-L). The larger the loss value, the smaller the reduction factor, and the lower the complementarity coefficient. The energy complementarity coefficients are used to guide energy dispatch strategies. Energy combinations with high complementarity coefficients are prioritized for switching when power supply is insufficient, while energy combinations with low complementarity coefficients are prevented from frequent switching.

[0044] Composite power supply characteristics are identified through energy complementarity coefficients. The energy complementarity coefficient matrix shows that the complementarity coefficients are as follows: PV-wind power 0.51, PV-energy storage 0.42, PV-grid 0.28, wind power-energy storage 0.38, wind power-grid 0.25, and energy storage-grid 0.32. Composite power supply characteristics are defined as a combination mode of supplying power from multiple energy sources simultaneously. The selection of the combination mode is based on the energy complementarity coefficients and current power demand. Energy pairs with energy complementarity coefficients greater than 0.4 are marked as strong complementarity pairs, including PV-wind power and PV-energy storage. The PV-wind power combination utilizes the inverse correlation between the power outputs of the two energy sources to maintain a stable power supply, while the PV-energy storage combination utilizes the rapid response of energy storage to mitigate PV fluctuations. These two energy combinations are preferentially used to construct power supply modes with composite power supply characteristics. The outdoor workstation employs a photovoltaic-energy storage hybrid power supply during daytime hours, with photovoltaic power handling 70% of the load and energy storage handling 30%. The load allocation ratio of the hybrid power supply is dynamically adjusted based on the real-time power output and energy complementarity coefficient of each energy source, with higher complementarity coefficients receiving higher load allocation weights. The maximum row and column values ​​of the energy complementarity coefficient matrix identify the optimal complementary partner for each energy type; for photovoltaics, the optimal complementary partner is wind power, and for energy storage, it is photovoltaics. These optimal complementary partners are used for priority combinations. The hybrid power supply feature includes three attributes: energy combination type, load allocation ratio, and switching priority. The energy combination type is the set of energy types participating in the power supply. The switching priority of the hybrid power supply feature is sorted from highest to lowest complementarity coefficient, with the photovoltaic-wind power hybrid mode having the highest priority and the energy storage-grid hybrid mode having the lowest priority. The switching priority is used to select the optimal hybrid power supply combination when a single energy source is insufficient, with higher priority combinations being activated first.

[0045] In some embodiments, the step of dynamically adjusting the power supply switching threshold table based on the composite power supply characteristics to generate scheduling compensation parameters includes: performing capacity distribution analysis on the composite power supply characteristics to generate a buffer standby identifier; performing compensation characteristic matching based on the buffer standby identifier to generate compensation characteristic parameters; determining adjustment configuration parameters using the compensation characteristic parameters; and determining scheduling compensation parameters based on the adjustment configuration parameters and the power supply switching threshold table.

[0046] Capacity distribution analysis is performed to generate buffer reserve indicators for composite power supply characteristics. In the morning session, the photovoltaic-energy storage combination with composite power supply characteristics has an actual photovoltaic output of 520W and an actual energy storage output of 1500W, with a total combined output of 2020W, meeting the load demand of 1800W during this period. Capacity distribution analysis calculates the reserve capacity of each energy source, which is the maximum output capacity minus the actual output power. In the composite power supply characteristic, the photovoltaic maximum output capacity is 800W, the actual output is 520W, and the reserve capacity is 280W; the energy storage maximum output capacity is 5000W, the actual output is 1500W, and the reserve capacity is 3500W. The buffer reserve indicator is defined as the percentage of reserve capacity to maximum output capacity; the photovoltaic buffer reserve indicator is 35%, and the energy storage buffer reserve indicator is 70%. When the outdoor workstation is in a photovoltaic-energy storage composite power supply situation, the high buffer reserve indicator of the energy storage ensures that the energy storage can quickly compensate for sudden drops in photovoltaic power. The sum of the buffer reserve indicators for the composite power supply characteristics is the percentage of the sum of the reserve capacity of the two energy types to the sum of the combined maximum output capacity. For the photovoltaic-energy storage combination, the sum of the buffer reserve indicators is 65%. Capacity distribution analysis identifies energy sources with insufficient buffer capacity. When the buffer reserve indicator is below 30%, it is marked as a low reserve state, triggering an energy switching warning. The load allocation ratio of the composite power supply characteristics is dynamically adjusted based on the buffer reserve indicators. The load allocation ratio of low-reserve energy sources is reduced to retain more reserve capacity. This ratio adjustment ensures the system's ability to cope with sudden load fluctuations. The buffer reserve indicator varies at different times of day; the photovoltaic buffer reserve indicator is higher during the day and drops to zero at night. These time-related changes in the buffer reserve indicator guide the time-of-day switching of the composite power supply characteristics.

[0047] Compensation characteristic parameters are generated based on the buffer reserve identifier. The 35% photovoltaic (PV) and 70% energy storage (ESD) levels of the buffer reserve identifier are matched with a compensation characteristic database, which stores compensation characteristic templates corresponding to different reserve levels. Each compensation characteristic template defines three parameters: compensation response time, compensation power limit, and compensation duration. Higher reserve levels correspond to faster response and higher power limits. The 70% ESD level of the buffer reserve identifier is matched to a high reserve template with parameters of 50 milliseconds response time, 3500W power limit, and 30 minutes duration. When the PV power at an outdoor workstation suddenly drops by 300W due to cloud cover, the ESD system, based on the compensation characteristic parameters, increases the output by 300W within 50 milliseconds to compensate for the power shortfall. The response time of the compensation characteristic parameters reflects the dynamic response capability of the energy source; the 50 millisecond response time of ESD is faster than the 200 milliseconds of mains power, making ESD more suitable for handling rapid power fluctuations. The 35% PV level of the buffer reserve identifier is matched to a medium reserve template with parameters of 500 milliseconds response time, 280W power limit, and unlimited duration. The upper limit of the compensation characteristic parameter is equal to the reserve capacity, and the upper limit of the power determines the maximum compensation capability of the energy source. Compensation characteristic matching generates compensation characteristic parameters for each type of energy source with composite power supply characteristics. The compensation characteristic parameter set for the photovoltaic-energy storage combination includes parameters for both types of energy sources. When the buffer reserve indicator is below 20%, it matches to a low reserve template. The low reserve template has a longer response time and a lower power limit, making this energy source unsuitable as a primary compensation power source.

[0048] The adjustment configuration parameters are determined using compensation characteristic parameters. The energy storage response time of 50 milliseconds and the power limit of 3500W are used to calculate the threshold adjustment amount in the power supply switching threshold table. The adjustment amount is defined as a correction coefficient of the baseline threshold. The correction coefficient of the adjustment configuration parameters is positively correlated with the compensation capability; the stronger the compensation capability, the larger the correction coefficient, and the lower the switching threshold in the power supply switching threshold table can be set. The high compensation capability of the outdoor workstation's energy storage allows the switching threshold of the charging load to be reduced from the baseline value of 120% to 110%. The reduced threshold triggers power supply switching earlier, fully utilizing the rapid compensation capability of energy storage. The response time of the compensation characteristic parameters determines the response speed coefficient of the threshold adjustment; the shorter the response time, the larger the speed coefficient, and the more timely the threshold adjustment. The adjustment configuration parameters include three dimensions: correction coefficient, response speed coefficient, and duration coefficient. These three-dimensional parameters comprehensively characterize the threshold adjustment characteristics. The power limit of the compensation characteristic parameters determines the upper limit of the threshold adjustment amplitude. The power limit of 3500W corresponds to a maximum correction coefficient of 0.9, indicating that the threshold can be reduced by a maximum of 10%. Adjusting the duration coefficient of the configuration parameters reflects the sustainability of the compensation capability. A duration coefficient of 0.5 corresponds to a storage duration of 30 minutes, indicating that the compensation capability is effective within 30 minutes. The upper limit of 280W for photovoltaic power in the compensation characteristic parameters is relatively low, corresponding to a correction coefficient of 0.95, indicating that the adjustment range of photovoltaic to the threshold is relatively small. Adjusting the configuration parameters integrates the compensation capabilities of energy storage and photovoltaic, allocating correction coefficients according to the response coefficients and priorities of various loads: 0.9 for charging loads, 0.92 for air conditioning loads, 0.94 for lighting loads, and 0.96 for communication loads. The higher the load priority, the smaller the correction coefficient and the greater the threshold reduction.

[0049] The scheduling compensation parameters are determined based on the adjusted configuration parameters and the power supply switching threshold table. The charging load correction factor of 0.9 in the adjusted configuration parameters applies to the peak-period switching boundary of 120% in the power supply switching threshold table, resulting in a switching boundary of 108%. The charging load valley-period switching boundary of 150% in the power supply switching threshold table is adjusted to 135% after the correction factor of 0.9. The adjusted threshold table is used for actual power supply scheduling decisions. The scheduling compensation parameters record the adjusted switching boundary values ​​and the corresponding compensation energy identifiers. The peak-period scheduling compensation parameters for charging load are a switching boundary of 108%, compensation energy storage, and a compensation power limit of 3500W. When the power supply margin of the outdoor workstation drops to 110%, the power supply switching of the charging load is triggered, and the energy storage increases the output compensation power gap within 50 milliseconds. The response speed factor in the adjusted configuration parameters is used to set the response delay of the scheduling compensation parameters; a larger response speed factor results in a shorter response delay. The switching boundaries of the four types of loads in the five time periods are corrected to form a new threshold matrix with a matrix size of 4×5. The scheduling compensation parameters now include trigger conditions for compensation actions. These conditions include a margin threshold, the type of energy source for compensation, the amount of compensation power, and the duration of compensation. Adjusting the duration coefficient of the configuration parameters sets the validity period of the scheduling compensation parameters; a storage duration coefficient of 0.5 corresponds to a 30-minute validity period for the compensation action. The selection of compensation energy sources for the scheduling compensation parameters is based on the energy complementarity coefficient and compensation characteristic parameters, prioritizing energy sources with high complementarity coefficients and strong compensation capabilities as compensation power sources.

[0050] Step S140: Perform load priority verification on the scheduling compensation parameters to identify power supply constraints, perform emergency margin determination based on power supply constraints to determine valid switching actions, and extract priority parameters from valid switching actions to generate a switching scheduling sequence.

[0051] In some embodiments, performing load priority verification and identifying power supply constraints on the scheduling compensation parameters includes: extracting load-associated nodes for each action from the scheduling compensation parameters; performing task status analysis on the load-associated nodes to generate priority adjustment identifiers; identifying conflict action groups through the priority adjustment identifiers; and generating power supply constraints by combining and arranging the conflict action groups.

[0052] The load-related nodes for each action are extracted from the dispatch compensation parameters. The charging load switching action in the dispatch compensation parameters is associated with the charging branch circuit breaker CB1 in the distribution cabinet. The switching action requires operating CB1 to open or close. Load-related nodes are defined as the electrical equipment and control nodes involved in the switching action, including four types of nodes: circuit breakers, contactors, converters, and load equipment. The air conditioning load switching action in the dispatch compensation parameters is associated with the air conditioning contactor KC1 and the air conditioning compressor load. The switching action requires coordinating the opening and closing of KC1 and the starting and stopping of the compressor. The lighting load switching action at the outdoor workstation is associated with the lighting controller LC1 and the LED light group load. Controller LC1 adjusts the brightness or on / off status of the light group according to the switching command. Load-related nodes record three attributes: node identifier, node type, and node status. The node identifier for the charging branch circuit breaker CB1 is CB1, the node type is circuit breaker, and the node status is closed. The communication load switching action in the dispatch compensation parameters is associated with the UPS uninterruptible power supply and the communication equipment load. The UPS, as an associated node, automatically switches to battery power when the mains power is abnormal. The energy storage switching action of the scheduling compensation parameters is associated with the bidirectional energy storage converter DC1. DC1 switches between charging and discharging modes according to the switching command. A total of 12 load-related nodes are extracted from the four types of loads in the scheduling compensation parameters. The nodes are distributed at three levels: distribution cabinet, control box, and load equipment.

[0053] Task status analysis is performed on load-related nodes to generate priority adjustment flags. The original load priority order of the scheduling compensation parameters is communication level 1, lighting level 2, air conditioning level 3, and charging level 4, with lower priority values ​​indicating higher priority. The charging branch circuit breaker CB1 of the load-related node is currently closed and carrying a charging load of 2000W. Task status analysis determines whether CB1 is in a switchable state. Task status analysis checks the operating status, fault status, and maintenance status of the load-related node. Nodes in normal operating status, without faults, and not in maintenance status are marked as switchable. The air conditioning contactor KC1 of the load-related node is in operating status but has detected an overheating alarm. Task status analysis marks KC1 as a restricted switchable state. The priority adjustment flag adjusts the switching priority of the load based on the task status analysis results. Loads in the restricted switchable state have lower priority to avoid equipment damage. The charging load carried by the charging branch circuit breaker CB1 of the load-related node includes two charging piles, one of which is charging an emergency vehicle. Task status analysis identifies a critical charging task and generates a priority adjustment flag to temporarily increase the priority of the charging load. The priority adjustment flag includes two parameters: adjustment direction and adjustment magnitude. The adjustment direction is either to increase or decrease, and the adjustment magnitude is the amount of change in priority level. If the communication load associated with a load-related node is in a normal power supply state, but the UPS battery capacity is below 20%, the task status analysis generates a priority adjustment flag that further increases the communication load priority. The adjustment result of the priority adjustment flag forms a dynamic priority table. The dynamic priority table overlays the temporary task status adjustment on top of the original priority and is used for subsequent conflict identification.

[0054] Conflict action groups are identified through priority adjustment flags. The priority adjustment flag indicates that the charging load's priority has been temporarily increased from level 4 to level 2 due to a critical charging task, creating a priority conflict with the original level 2 lighting load. Conflict identification compares the adjusted dynamic priority table, identifying load pairs with the same or similar priorities. Load pairs with a priority difference of less than 1 level are marked as conflict pairs. The charging load's priority is now level 2 after adjustment, and the lighting load's original priority was level 2; the priority difference between the two is 0, triggering conflict identification. When the outdoor workstation has insufficient power margin, it needs to simultaneously switch between charging and lighting loads, but since both types of loads have the same priority, the switching order cannot be determined, resulting in a switching decision conflict. A conflict action group is defined as a set of load switching actions with priority conflicts; the charging load switching action and the lighting load switching action form conflict action group A. The priority adjustment flag indicates that the air conditioning load's priority has been reduced from level 3 to level 4 due to contactor limitations; its priority difference with the communication load is large, so there is no conflict. The conflict identification process calculates the conflict intensity of conflict action groups. Conflict intensity is defined as the ratio of the sum of the power of conflicting loads to the total load. Conflict action group A has a total power of 2700W (2000W for charging and 700W for lighting), with a conflict intensity of 64%. Conflict action groups include conflicting load types and conflict intensities.

[0055] Power supply constraints are generated through a combination of conflict action groups. In conflict action group A, the charging load and lighting load have the same priority. The combination arrangement performs a secondary prioritization based on load characteristics, with critical charging tasks being more important than general lighting, thus placing the charging load before the lighting load. The priority ranking rule for power supply constraints adds a secondary discrimination condition: when priorities are the same, they are compared sequentially by task criticality, load power, and switching response time. The combination arrangement result for conflict action group A is that the charging load has a higher priority than the lighting load. The switching order after arrangement prioritizes ensuring charging power supply, and the lighting load is reduced if necessary. When the power supply margin of an outdoor workstation drops to a critical level, the combination arrangement result prioritizes maintaining charging power supply for emergency vehicles, and the lighting load is reduced to 50% brightness to save power. The switching timing constraints of the power supply constraints stipulate that switching actions in conflict action groups cannot be executed simultaneously, and a minimum time window is required between two actions. Due to contactor limitations, the priority of the air conditioning load is reduced to level 4, ranking after charging and lighting in the power supply constraints. Communication equipment must not be powered down under any circumstances. The power supply constraints generated by the combination arrangement include three categories: priority constraints, timing constraints, and capacity constraints. The capacity constraint of the power supply conditions limits the total power of the charging load and lighting load to no more than 3500W of energy storage backup capacity when they are switched at the same time. The total power of 2700W, which is 2000W for charging and 700W for lighting, meets the capacity constraint.

[0056] In some embodiments, the step of determining a valid switching action based on the power supply constraints includes: parsing the transfer margin critical point of the associated action based on the power supply constraints; performing margin over-limit identification for the transfer margin critical point to generate an over-limit feature set; using the over-limit feature set to filter actions that meet the transfer conditions to obtain a valid action configuration table; and determining a valid switching action based on the valid action configuration table.

[0057] The critical point of power transfer margin is analyzed based on the power supply constraints and related actions. The charging load switching action under the power supply constraints is associated with an energy storage power transfer scheme. This scheme requires the energy storage output power to be increased from 1500W to 3500W to handle a charging load of 2000W. The critical point of power transfer margin is defined as the minimum power supply margin required when performing the power transfer action. The critical point of charging load power transfer margin is when the remaining energy storage capacity can provide continuous power for 30 minutes. Under the power supply constraints, the remaining energy storage capacity is 32.5kWh, which, based on a total power of 3500W after the charging load transfer, can provide continuous power for 9.3 hours, meeting the minimum requirement of 30 minutes for the critical point of power transfer margin. When the outdoor workstation performs charging load transfer during the daytime, photovoltaic power generation is still continuously inputting power. The critical point of power transfer margin decreases to 15 minutes of continuous power supply from the energy storage, as the photovoltaic system can recover some power capacity within 15 minutes. The calculation of the critical point of power transfer margin considers three factors: the duration of power transfer, the power transfer output, and the energy recovery time. The longer the duration, the higher the critical point requirement. The switching action of lighting loads under power supply constraints is associated with the grid-to-grid power transfer scheme. The critical point for grid-to-grid power transfer margin is that the available grid capacity is greater than the lighting load power by 700W. The critical point for the associated action's critical point is categorized based on the type of energy source being transferred: energy storage critical point, grid critical point, and photovoltaic critical point. The criteria for determining the critical point differ for each type of energy source. The switching action of air conditioning loads under power supply constraints is associated with dual-path grid-to-grid power transfer, consisting of energy storage and grid power. The critical point for dual-path grid-to-grid power transfer margin is that the sum of the energy storage and grid power reserve capacities is greater than the air conditioning load power by 3000W. The critical point for the critical point includes the critical margin value, the corresponding energy source being transferred, and the duration of the transfer. For charging loads, the critical point for the critical point is 9.3 hours of energy storage, energy storage for the transferred energy source, and the duration of the transfer until photovoltaic power is restored.

[0058] A margin exceedance identification feature set is generated for the critical point of power transfer margin. Margin exceedance identification compares the actual power supply margin with the critical point of power transfer margin. If the actual margin is less than the critical point, it is marked as a severe exceedance; if the actual margin is between 1 and 3 times the critical point, it is marked as a critical exceedance; if the actual margin is between 3 and 5 times the critical point, it is marked as close to exceedance; and if the actual margin is greater than 5 times, it is marked as not exceedance. The charging load at the critical point of power transfer margin requires energy storage to provide continuous power for 30 minutes. Currently, the actual continuous power supply of energy storage is 9.3 hours, far exceeding the critical point by 5 times, therefore the charging load exceedance status is not exceeded. The lighting load transfer requirement under power supply constraints is a mains power available capacity greater than 700W. The current mains power access capacity of 8000W far exceeds the requirement, therefore the lighting load transfer margin is not exceeded. The outdoor workstation's remaining energy storage capacity drops to 5kWh during the nighttime period. Based on a total power of 3500W after power transfer, it can only provide power for 1.4 hours. The actual margin is 2.8 times the critical 30-minute limit, falling within the 1-3 times range. This margin exceedance is marked as critical. The air conditioning load at the power transfer margin critical point requires the sum of energy storage and mains backup capacity to be greater than 3000W. Currently, the energy storage backup of 3500W and the mains backup of 8000W total 11500W, indicating sufficient margin for air conditioning load power transfer without exceeding the limit. The margin exceedance identification feature set includes an exceedance load list and a non-exceedance load list. Exceedance loads do not meet the power transfer conditions under current conditions, while non-exceedance loads can be transferred. For communication load power transfer under power supply constraints, the UPS battery capacity must be greater than 10%. The current UPS battery capacity is 18%, 1.8 times the critical 10%, falling within the 1-3 times range. This communication load power transfer margin is marked as critical. The over-limit feature set includes the over-limit status, over-limit level and corresponding transfer margin information of each load. Loads that are severely over-limit are prohibited from being transferred, loads that are critically over-limit are restricted from being transferred, and loads that are close to being over-limit are allowed to be transferred but need to be monitored.

[0059] Using the over-limit feature set, actions meeting the transfer conditions are filtered to obtain a valid action configuration table. The charging load in the over-limit feature set has an actual margin of 9.3 hours, which is greater than the critical point of 30 minutes. The over-limit status is not over-limit, and the filtering result indicates that the charging load meets the transfer conditions. The valid action configuration table records the load switching actions meeting the transfer conditions. The charging load transfer action is included in the valid action configuration table, with action parameters including energy storage transfer, transfer power of 2000W, and transfer duration of 9.3 hours. The lighting load in the over-limit feature set has an actual margin of 8000W, which is greater than the critical point of 700W. The lighting load transfer action is included in the valid action configuration table, with the energy source being mains power and the transfer power being 700W. The over-limit feature set of the outdoor workstation shows that the air conditioning load, lighting load, and charging load are not over-limit. All transfer actions for these three types of loads are included in the valid action configuration table. The communication load maintains UPS power supply. The UPS automatically switches when the mains power is abnormal. When the UPS battery is in a critical over-limit state, capacity monitoring is strengthened. The screening criteria include three conditions: no limit violation status, non-severe limit violation level, and available energy for transfer. Actions that meet all conditions pass the screening. If the remaining energy storage capacity of the limit violation feature set is insufficient for a certain period, and the limit violation status of the charging load transfer action is severe limit violation, the screening result is that the charging load does not meet the transfer conditions, and this action is excluded from the effective action configuration table. The number of actions in the effective action configuration table is equal to the number of loads that have not exceeded the limit. The effective action configuration table for the current period includes transfer actions for three types of loads: charging, lighting, and air conditioning. The screening process for the limit violation feature set prioritizes the transfer actions of high-priority loads. The loads that need to be transferred are dynamically prioritized as follows: charging > lighting > air conditioning. High-priority loads are still retained for transfer actions when their margin is close to the limit. The effective action configuration table records the execution conditions for each transfer action. The execution conditions for the charging load transfer action are that the power supply margin is less than 108% and the remaining energy storage capacity is greater than the critical point.

[0060] Effective switching actions are determined based on the effective action configuration table. The table lists three actions: charging load transfer, lighting load transfer, and air conditioning load transfer. The actual effective switching actions to be executed are determined based on the current power supply margin and switching boundaries. The current power supply margin is 105%. The charging load switching boundaries of 108% and the air conditioning load switching boundary of 115% have both been triggered, while the lighting load switching boundary of 94% has not been triggered. The determination of effective switching actions follows the principle of minimizing switching, prioritizing loads with high response coefficients for transfer. The charging load with the highest response coefficient is prioritized as the effective switching action. When the power supply margin at the outdoor workstation drops to 105%, the charging load is transferred from photovoltaic to energy storage, the air conditioning load is transferred from photovoltaic to a dual-circuit power supply of energy storage and mains power, and the lighting load maintains its original power supply mode. Effective switching actions record three attributes: switching time, switched load, and energy source supplied. For charging loads, the effective switching action record is: switching time 18:30, switching load charging 2000W, energy source supplied to storage. For air conditioning loads, the effective switching action record is: switching time 18:31, switching load air conditioning 3000W, energy source supplied to storage and mains power. When the power supply margin continuously decreases to 93%, the lighting load switching boundary is triggered at 94%, and the lighting load switching action is determined as a new effective switching action. The system executes the switching of lighting loads from photovoltaic power to mains power. When multiple loads trigger effective switching actions simultaneously, they are executed sequentially according to the priority of power supply constraints, with higher priority loads taking precedence. Actions not triggered in the effective action configuration table are retained in the configuration table as backup switching schemes to be executed when the power supply margin further decreases.

[0061] Priority parameters are extracted from effective switching actions to generate a switching scheduling sequence. The priority parameters for the charging load transfer action of an effective switching action are priority level 2, switching time 18:30, switching duration 200 milliseconds, and transfer power 2000W. The priority parameters include four dimensions: priority level, switching time, switching duration, and transfer power, which describe the scheduling characteristics of the switching action. The switching scheduling sequence is arranged from earliest to latest according to the switching time of the priority parameters. The power supply margin drops to 105% at 18:30, triggering charging switching; it continues to decrease at 18:31, triggering air conditioning switching; and it further decreases to 93% at 18:33, triggering lighting switching. The switching scheduling sequence is as follows: charging load 18:30, air conditioning load 18:31, and lighting load 18:33. As the power supply margin of the outdoor workstation continues to decrease, the switching scheduling sequence guides the system to execute the transfer of charging, air conditioning, and lighting in sequence, avoiding power supply shock caused by the simultaneous switching of multiple loads. The time interval of the handover scheduling sequence is determined by the timing constraints of the power supply constraints. The time interval between adjacent handover actions is at least the handover duration of the previous action plus a 50-millisecond safety margin. After the handover scheduling sequence is generated, a handover timetable is formed. The timetable records the scheduled execution time and completion time of each handover action. The system automatically executes the handover operation according to the timetable. The dynamic adjustment mechanism of the handover scheduling sequence inserts or deletes handover actions based on real-time changes in power supply margin. When the power supply margin suddenly drops, an emergency handover action is inserted. When the power supply margin recovers, sorted but unexecuted handover actions are canceled.

[0062] Step S150: The power supply switching threshold table and the scheduling compensation parameters are correlated and matched to generate emergency output parameters. The scheduling priority weight is determined based on the emergency output parameters and power supply constraints. The energy scheduling command is output by combining the scheduling priority weight with the switching scheduling sequence.

[0063] Specifically, the power supply switching threshold table and scheduling compensation parameters are correlated and matched to generate emergency output parameters. The charging load peak-hour switching boundary of 120% in the power supply switching threshold table is correlated and matched with the charging load adjustment boundary of 108% in the scheduling compensation parameters, with the threshold adjustment range identified as 10%. The correlation and matching process compares the original switching boundary of the power supply switching threshold table with the adjustment boundary of the scheduling compensation parameters, calculating the difference and percentage difference. The lighting load peak-hour switching boundary of 100% in the power supply switching threshold table is matched with the lighting load adjustment boundary of 94% in the scheduling compensation parameters, with a threshold adjustment range of 6%, reflecting the degree of impact of compensation capability on the switching boundary. The emergency output parameters record the final switching boundary and corresponding compensation scheme after correlation and matching. The emergency output parameters for the charging load are a switching boundary of 108%, compensation energy storage, and compensation power of 3500W. When the power supply margin of the outdoor workstation drops to 108%, the emergency output parameters trigger the power supply switching of the charging load, and the energy storage increases the output compensation power gap within 50 milliseconds. The five time-period switching boundaries of the power supply switching threshold table are matched with the five time-period adjustment boundaries of the scheduling compensation parameters to generate 20 emergency output parameters for 4 types of loads × 5 time periods. The emergency output parameters include four fields: load type, switching boundary, compensation energy source, and compensation power. The correlation matching process generates a priority ranking of the emergency output parameters, ranking each load based on its lowest switching boundary during low-capacity periods: communication 77%, lighting 94%, charging 108%, and air conditioning 115%. This ranking result is used to determine the switching execution order.

[0064] The dispatch priority weight is determined based on emergency output parameters and power supply constraints. A minimum communication load switching boundary of 77% in the emergency output parameters indicates that communication loads receive the highest priority. A communication load priority level of 1 in the power supply constraints confirms communication as the highest priority load. The dispatch priority weight is calculated by combining the switching boundary of the emergency output parameters and the priority level of the power supply constraints, using the formula Wd = P × (1 - T / 200), where Wd is the dispatch priority weight (dimensionless), P is the priority level coefficient (dimensionless), and T is the switching boundary percentage (in %). The priority level coefficient is allocated as follows: communication 1.0, lighting 0.8, air conditioning 0.7, and charging 0.6. The communication load has the lowest switching boundary and the highest priority level coefficient, resulting in a dispatch priority weight of 0.615, the highest among the four load categories. When the power supply margin of the outdoor workstation continues to decrease, the dispatch priority weight guides the system to prioritize maintaining power to high-weight loads; the highest-weighted communication load will not be disconnected under any circumstances. The lighting load has a switching boundary of 94% combined with a priority level coefficient of 0.8, resulting in a dispatch priority weight of 0.424, ranking second. The switching boundary for air conditioning loads is 115%, combined with a priority level coefficient of 0.7, resulting in a scheduling priority weight of 0.298. The switching boundary for charging loads is 108%, combined with a priority level coefficient of 0.6, resulting in a scheduling priority weight of 0.276, the lowest among the four load categories. The scheduling priority weight ranges from 0 to 1, with higher values ​​indicating higher load priority. The weight ranking is: Communication 0.615 > Lighting 0.424 > Air Conditioning 0.298 > Charging 0.276. The priority adjustment flag for power supply constraints dynamically corrects the scheduling priority weight. When a critical charging task triggers a priority increase, the charging load weight temporarily increases by 0.1. After correction, the charging weight may exceed the air conditioning weight, triggering a priority reversal. The availability of compensating energy in emergency output parameters affects the scheduling priority weight. When compensating energy fails or its capacity is insufficient, the corresponding load weight decreases to ensure the feasibility of switching actions. The scheduling priority weight is used to prioritize energy dispatch commands. Loads with higher weights are given priority in ensuring power supply when there is a power shortage, while loads with lower weights are given priority for reduction or switching.

[0065] The energy dispatch command is output using a combination of dispatch priority weights and switching dispatch sequences. The communication load, with the highest dispatch priority weight of 0.615, has no switching action in the switching dispatch sequence, and the energy dispatch command is to maintain UPS power supply for the communication load without switching. The charging load transfer action in the switching dispatch sequence is executed first at 18:30, with a dispatch priority weight of 0.276 confirming that charging is a switchable load, and the energy dispatch command is to switch the charging load from photovoltaic to energy storage. The outdoor workstation executes the energy dispatch command at 18:30; the charging branch circuit breaker CB1 disconnects the photovoltaic power supply, the energy storage bidirectional converter DC1 starts outputting 2000W, and the charging pile completes the power switch within 200 milliseconds. The energy dispatch command includes five parameters: switching time, switching load, source energy, target energy, and switching method. The air conditioning load transfer action in the switching dispatch sequence is executed second at 18:31, with a dispatch priority weight of 0.298 confirming that air conditioning is a switchable load, and the energy dispatch command is to switch the air conditioning load from photovoltaic to dual power supply from energy storage and mains power. The lighting load transfer action in the switching sequence was executed third at 18:33, with a scheduling priority weight of 0.424, confirming lighting as a medium-priority load. The energy dispatch command was to switch the lighting load from photovoltaic power to grid power. The switching methods for energy dispatch commands are divided into hard switching and soft switching. Hard switching involves disconnecting first and then closing, resulting in a brief power outage, while soft switching involves closing first and then disconnecting, achieving seamless switching. High-priority loads use soft switching. Energy dispatch commands are issued to each load-related node via the fieldbus, and circuit breakers, contactors, and converters execute opening and closing operations according to the command parameters. The scheduling priority weight is used for conflict arbitration of energy dispatch commands; when multiple switching actions are requested simultaneously, they are executed sequentially from highest to lowest weight.

[0066] To implement the above-described method embodiments, a smart emergency energy dispatching method for outdoor workstations is proposed to achieve the corresponding functions and technical effects. See also... Figure 2 , Figure 2 This application provides a structural block diagram of an outdoor workstation emergency energy intelligent dispatching system 200, which includes: Data acquisition module 201 is used to collect multi-source energy data and load operation data of outdoor workstations, and to perform time-series alignment and feature extraction processing on the multi-source energy data and load operation data to generate an energy status parameter set; The status assessment module 202 is used to perform margin interval division to identify power margin levels based on the energy status parameter set, perform margin gradient analysis along the power margin levels to generate load response coefficients, and construct a power supply switching threshold table by mapping the load response coefficients. The coupling analysis module 203 is used to perform energy coupling analysis on the energy state parameter set to determine the energy complementarity coefficient of different energy sources, identify composite power supply characteristics through the energy complementarity coefficient, and dynamically adjust the power supply switching threshold table based on the composite power supply characteristics to generate scheduling compensation parameters. The timing scheduling module 204 is used to perform load priority verification and identify power supply constraints on the scheduling compensation parameters, perform emergency margin determination based on the power supply constraints to determine effective switching actions, and extract priority parameters from the effective switching actions to generate a switching scheduling sequence. The instruction generation module 205 is used to perform association matching processing between the power supply switching threshold table and the scheduling compensation parameters to generate emergency output parameters, determine the scheduling priority weight based on the emergency output parameters and the power supply constraints, and output energy scheduling instructions using the scheduling priority weight in combination with the switching scheduling sequence.

[0067] The aforementioned intelligent emergency energy dispatching system 200 for outdoor workstations can implement the intelligent emergency energy dispatching method for outdoor workstations described in the above-described method embodiments. The options described in the above method embodiments are also applicable to this embodiment and will not be detailed here. The remaining contents of this application's embodiments can be referred to the contents of the above method embodiments, and will not be repeated in this embodiment.

[0068] The above embodiments are not an exhaustive list based on the present invention, and there may be many other embodiments not listed. Any substitutions and improvements made without departing from the concept of the present invention are within the protection scope of the present invention.

Claims

1. A method for intelligent emergency energy dispatching for outdoor workstations, characterized in that, include: Collect multi-source energy data and load operation data from outdoor workstations, and perform time-series alignment and feature extraction processing on the multi-source energy data and load operation data to generate an energy status parameter set; Based on the energy state parameter set, a margin interval is divided to identify the power margin level. Margin gradient analysis is performed along the power margin level to generate a load response coefficient. A power supply switching threshold table is constructed by mapping the load response coefficient. Energy coupling analysis is performed on the energy state parameter set to determine the energy complementarity coefficients of different energy sources. The composite power supply characteristics are identified through the energy complementarity coefficients. Based on the composite power supply characteristics, the power supply switching threshold table is dynamically adjusted to generate scheduling compensation parameters. The load priority verification is performed on the scheduling compensation parameters to identify power supply constraints. Based on the power supply constraints, an emergency margin determination is performed to determine valid switching actions. Priority parameters are extracted from the valid switching actions to generate a switching scheduling sequence. The power supply switching threshold table is associated and matched with the scheduling compensation parameters to generate emergency output parameters. The scheduling priority weight is determined based on the emergency output parameters and the power supply constraints. The energy scheduling command is output using the scheduling priority weight in combination with the switching scheduling sequence.

2. The method according to claim 1, characterized in that, The step of generating load response coefficients by performing margin gradient analysis along the power margin level includes: Based on the power margin level, the cumulative margin record is extracted to construct the margin change curve; Gradient mutation detection is performed on the margin change curve to generate a margin mutation early warning indicator; Based on the margin mutation early warning identifier, load type weights are obtained by associating load patterns. The load type weights are used to perform response intensity mapping on the margin mutation warning flag to generate load response coefficients.

3. The method according to claim 1, characterized in that, The construction of the power supply switching threshold table using the load response coefficient mapping includes: The load response coefficients are divided into time periods according to energy cycle characteristics to generate a time period response coefficient set; For the aforementioned time period response coefficient set, load status analysis is performed to dynamically identify peak and valley time period characteristics; The peak-valley time period switching boundary is set based on the peak-valley time period feature identifier matching differential threshold benchmark; The peak-valley time period switching boundaries are integrated to form a power supply switching threshold table.

4. The method according to claim 1, characterized in that, The step of performing energy coupling analysis on the energy state parameter set to determine the energy complementarity coefficients of different energy sources includes: Energy spectrum features are obtained by extracting time-frequency domain characteristics based on the energy state parameter set. The energy spectrum characteristics are used to perform pattern matching to identify energy types; The energy switching efficiency loss value is determined by performing a switching efficiency assessment based on the energy type. Based on the energy type and the energy switching efficiency loss value, topology path analysis is performed to obtain the energy complementarity coefficients of different energy sources.

5. The method according to claim 1, characterized in that, The step of dynamically adjusting the power supply switching threshold table based on the composite power supply characteristics to generate scheduling compensation parameters includes: Capacity distribution analysis is performed on the composite power supply characteristics to generate buffer standby identifiers; Compensation characteristic parameters are generated by matching the compensation characteristics based on the buffer backup identifier; The adjustment configuration parameters are determined using the aforementioned compensation characteristic parameters; The scheduling compensation parameters are determined based on the adjusted configuration parameters and the power supply switching threshold table.

6. The method according to claim 1, characterized in that, The step of performing load priority verification and identifying power supply constraints on the scheduling compensation parameters includes: Extract the load-related nodes for each action from the scheduling compensation parameters; Perform task status analysis on the load-associated nodes to generate priority adjustment identifiers; The conflict action group is determined by using the priority adjustment flag; Power supply constraints are generated by combining and arranging the conflict action groups.

7. The method according to claim 1, characterized in that, The step of determining a valid switching action based on the power supply constraints includes: Based on the power supply constraints, the critical point of the power transfer margin for related actions is analyzed. For the aforementioned supply margin critical point, margin over-limit identification is performed to generate an over-limit feature set; Using the aforementioned over-limit feature set, actions that meet the transfer conditions are filtered to obtain a valid action configuration table; Valid switching actions are determined based on the valid action configuration table.

8. The method according to claim 3, characterized in that, The process of dynamically identifying peak and valley time period characteristics through load state analysis on the set of time period response coefficients includes: Perform coefficient distribution analysis on the time period response coefficient set to locate the critical point of coefficient distribution; Load state parameters are generated by performing load state analysis on the critical point of the coefficient distribution. Threshold response coefficients are generated by performing threshold response analysis on the load state parameters. Based on the threshold response coefficient, time period calibration is performed to generate peak and valley time period feature identifiers.

9. The method according to claim 4, characterized in that, The process of performing pattern matching to identify energy types based on the energy spectrum characteristics includes: Energy distribution characteristics are obtained by performing frequency domain energy distribution analysis based on the energy spectrum characteristics. Peak identification is performed based on the energy distribution characteristics to generate a peak feature set and an energy performance degradation identifier; A distribution attribute set is constructed by using the peak feature set and the energy performance degradation identifier for characteristic mapping; Based on the distributed attribute set, the energy type is identified by performing association matching based on the pattern library.

10. An intelligent emergency energy dispatching system for outdoor workstations, characterized in that, include: The data acquisition module is used to collect multi-source energy data and load operation data from the outdoor workstation, and to perform time-series alignment and feature extraction processing on the multi-source energy data and the load operation data to generate an energy status parameter set. The status assessment module is used to perform margin interval division to identify power margin levels based on the energy status parameter set, perform margin gradient analysis along the power margin levels to generate load response coefficients, and construct a power supply switching threshold table by mapping the load response coefficients. The coupling analysis module is used to perform energy coupling analysis on the energy state parameter set to determine the energy complementarity coefficient of different energy sources, identify composite power supply characteristics through the energy complementarity coefficient, and dynamically adjust the power supply switching threshold table based on the composite power supply characteristics to generate scheduling compensation parameters. The timing scheduling module is used to perform load priority verification on the scheduling compensation parameters to identify power supply constraints, perform emergency margin determination based on the power supply constraints to determine effective switching actions, and extract priority parameters from the effective switching actions to generate a switching scheduling sequence. The instruction generation module is used to associate and match the power supply switching threshold table with the scheduling compensation parameters to generate emergency output parameters, determine the scheduling priority weight based on the emergency output parameters and the power supply constraints, and output energy scheduling instructions using the scheduling priority weight in combination with the switching scheduling sequence.

Images