A method and system for aging recursion and collaborative management for antarctic energy systems

Abstract

This invention provides a method and system for aging recursion and collaborative management of Antarctic energy systems, relating to the technical field of microgrid energy management and equipment health status assessment. The method specifically includes: acquiring multi-dimensional operational physical quantities in the polar region; uniformly mapping the operational losses of heterogeneous equipment within the system to absolute energy loss values; obtaining the real-time health status of equipment through a multi-source equipment coupled aging forward recursion model; calculating the expected remaining physical lifetime based on the real-time health status of the equipment, comparing it with the polar summer flood logistics replenishment time window to construct a survival margin index, and dynamically generating real-time penalty weight coefficients; using an improved particle swarm optimization algorithm to solve for the optimal power allocation ratio that minimizes the total energy loss that can be processed by the system; converting the optimal power allocation ratio into absolute active power scheduling instructions and issuing them to the underlying physical hardware of the microgrid for execution.

Description

Technical Field

[0001] This invention relates to the field of microgrid energy management and equipment health status assessment technology, and in particular to an aging regression and collaborative management method and system for Antarctic energy systems. Background Technology

[0002] The energy supply for polar (such as Antarctica) research stations and related facilities typically relies on independent microgrid systems comprising wind power, photovoltaic power, diesel generators, and multi-source heterogeneous energy storage (such as lithium batteries, supercapacitors, fuel cells, and electrolyzers). The Antarctic environment is extremely unique, characterized by perennial extreme cold, alternating periods of polar night and day, and frequent extreme blizzards. These harsh geographical and climatic conditions result in highly random, intermittent, and drastic high-frequency fluctuations in the output of renewable energy within the system. Furthermore, due to their remote location, the supply of materials and maintenance of polar microgrids are subject to strict seasonal logistical windows, often with only a brief summer flood season for equipment replacement and fuel replenishment. Therefore, the system operates unattended under high stress for extended periods, placing extremely high demands on the absolute reliability of the power supply system and the long-term sustainability of equipment operation.

[0003] Under extreme operating conditions, the health status of various heterogeneous devices within a microgrid faces severe challenges. Extremely cold environments not only directly increase the physical internal resistance of electrochemical energy storage devices such as batteries, but also lead to a surge in cold-start losses for power equipment such as electrolyzers. Furthermore, the high-frequency and drastic fluctuations in wind power generation cause the microgrid system to frequently perform high-rate charging, discharging, and start-stop operations, resulting in a rapid acceleration of the aging of multi-source devices. Current energy management technologies for microgrids typically aim at instantaneous power balance or minimizing single-fuel consumption. When it comes to equipment lifespan management, isolated evaluation models are often used to calculate the state of health (SOH) decay of individual components, ignoring the cross-constraints between devices responding to different frequency bands during the physical aging process. Simultaneously, existing lifespan quantification methods largely rely on static economic cost calculations or empirical parameters, lacking unified engineering physical measurement standards. This makes it difficult to accurately coordinate the physical fatigue of multiple heterogeneous devices and the energy flow of the entire network within a single control model, and thus fails to accurately reflect the system's true dispatchability.

[0004] On the other hand, existing microgrid collaborative scheduling optimization algorithms exhibit significant limitations when facing the specific needs of polar regions. Conventional swarm intelligence optimization algorithms are prone to gradient information distortion due to sudden data abrupt extremes when processing meteorological and power input data in Antarctica, leading to local optima or computational divergence. More critically, traditional scheduling strategies typically assign static constraint weights to all equipment during the optimization process, lacking survival intervention mechanisms specific to polar scenarios. Because the temporal correlation between equipment health degradation rates and polar logistics resupply time windows is ignored, existing management methods cannot adaptively trigger defensive scheduling or weight shifts when a core component in the system becomes critically ill before the next resupply window arrives. The system continues to operate along the established conventional optimization path, which can easily lead to the complete failure of critical components before resupply arrives, resulting in system-wide power supply collapse and failing to meet the high-safety-factor stationary operation requirements of polar energy systems. Summary of the Invention

[0005] This invention provides an aging regression and collaborative management method for Antarctic energy systems. The method includes the following steps: obtaining multidimensional operational physical quantities in the polar region, and uniformly mapping the operational losses of each heterogeneous device in the system to the absolute loss value of the energy that can be processed. By combining the global temperature penalty benchmark and the physical cross-constraint relationship between heterogeneous devices, the real-time health status of the devices is obtained through a multi-source device coupled aging forward recursive model. The expected remaining physical life is calculated based on the real-time health status of the equipment, and then compared with the time window for summer logistics replenishment in polar regions to construct a survival margin index, and a real-time penalty weight coefficient is dynamically generated. The system's real-time net power is nonlinearly smoothed and an adaptive inertial weight is constructed by combining the survival margin index. An improved particle swarm optimization algorithm is used to solve for the optimal power allocation ratio that minimizes the total energy loss of the system's throughput. The optimal power allocation ratio is then converted into an absolute active power scheduling command and sent to the underlying physical hardware of the microgrid for execution.

[0006] The operating losses of each heterogeneous device within the system are uniformly mapped to the absolute loss value of the energy throughput. Specifically, this includes the cumulative absolute loss value of the energy throughput for electrochemical and hydrogen energy conversion devices in the multi-source heterogeneous energy storage module. The mapping formula is: ; in, For equipment In time Real-time health status indicators; For equipment The energy mapping reference constant; These correspond to lithium batteries, supercapacitors, fuel cells, and electrolyzers, respectively. For the diesel generator set in the foundation power supply module with the shortfall, its cumulative equivalent energy loss value The formula for the continuous recursive mapping is: ; in, For historical iteration variables; For diesel generator sets at time steps The volume of fuel consumed in a single step; Density of special low-temperature diesel fuel for polar environments; The lower heating value of special low-temperature diesel oil; This refers to the energy conversion efficiency of the diesel generator set.

[0007] Combining the global temperature penalty benchmark and the physical constraints between heterogeneous devices, a multi-source device coupled aging forward recursive model is used to obtain the real-time health status of the devices and construct the extreme cold temperature penalty coefficient under continuous time steps. : ; in, This is the standard operating reference temperature for multi-source heterogeneous energy storage modules; This is a coefficient characterizing temperature sensitivity; Real-time polar ambient temperature; Real-time health status of supercapacitor array The calculation is performed using a single-step forward recursive approach: ; in, This refers to the health status of the supercapacitor at the previous time step. This refers to the actual high-frequency charging and discharging power undertaken by the supercapacitor at the current time step; This represents the voltage stress deviation ratio of the current supercapacitor terminal voltage relative to the rated voltage. and These are the reference attenuation coefficients driven by power stress and voltage stress, respectively; The sampling period duration; This is the maximum permissible attenuation safety limit set for single-step operation under polar conditions.

[0008] The physical constraints between heterogeneous devices include the cross-physical constraint mechanism of supercapacitors on lithium batteries. Specifically, based on the health status of the supercapacitor in the previous time step, the upper limit of the allowable power change rate of the lithium battery in the current time step is dynamically narrowed. : ; in, This refers to the rated maximum power change rate of the lithium battery. This is the lower limit threshold for the permissible safe lifespan of equipment in a microgrid system. To limit the smoothing factor; Perform an aging regression analysis of the lithium battery energy storage unit to determine its health status. The recurrence relation is: ; in, This represents the actual charging and discharging power allocated to the lithium battery at the current time step, and the single-step change in this power is constrained by the upper limit of the allowable power change rate. ; This represents the increment of the discharge depth at the current time step. and These are the cycle decay coefficients for power load and depth of discharge, respectively.

[0009] The physical constraints between heterogeneous devices also include dynamic wake-up constraints for hydrogen energy devices based on the health status of lithium batteries. The system dynamically adjusts the dead zone threshold of the fuel cell's start-up power according to the real-time status of the lithium batteries. : ; in, This is the rated reference start-up threshold for the fuel cell; Under the dynamic wake-up constraint of hydrogen energy equipment, proton exchange membrane fuel cells and electrolyzers respectively execute their own health state recursive formulas: ; ; in, This refers to the actual output power of the fuel cell; This refers to the rated output power of the fuel cell; This represents the fuel cell load attenuation coefficient. The fixed life loss equivalent caused by a single ignition start-stop cycle; This is a Boolean start / stop event indicator variable; This represents the current actual hydrogen production capacity absorbed by the electrolyzer. This is the rated absorption power of the electrolytic cell; This is the reference power load attenuation coefficient for the electrolytic cell.

[0010] The expected remaining physical lifespan is calculated based on the real-time health status of the equipment. This lifespan is then compared with the polar summer flood logistics resupply time window to construct a survival margin index. Real-time penalty weighting coefficients are dynamically generated. Specifically, the smoothed decay rate of the equipment's true physical degradation trend is extracted. : ; in, A sliding time window for historical observation; The device's historical health status at the start of the sliding time window; The sampling period duration; Calculate the expected remaining physical life of each device. And construct a survival margin index : ; ; in, This is the lower limit threshold for the permissible safe lifespan of equipment in a microgrid system. Establish a time window for logistics resupply during the polar summer flood season; dynamically generate real-time penalty weight coefficients for each piece of equipment based on survivability margin indicators. : ; in, For equipment The base energy loss weight under the initial safe state; This is the sensitivity constant for endangered species punishment.

[0011] A nonlinear smoothing mapping is performed on the real-time net power of the system to generate a compression net power index. Its mathematical expression is: in, It is a symbolic function; This refers to the system's real-time net power. The polar weather fluctuation compression factor; This represents the absolute maximum net power peak value of the polar microgrid. When using the improved particle swarm optimization algorithm, a multi-objective fitness function is constructed for the spatiotemporal boundary. : ; in, For the first In the nth iteration A set of power allocation ratio coefficients for each particle; This is the real-time penalty weighting coefficient; and These are the absolute loss value of single-step equivalent energy and the cumulative equivalent energy loss value calculated under the current particle allocation scheme, respectively. Static rejection weight; This is a cross-penalty item.

[0012] Cross-penalty items The calculation formula for the cross-physical constraint setting of supercapacitors on lithium batteries is as follows: ; in, This is a severe punishment for the multiplier; For the first In the nth iteration The lithium battery power allocation ratio coefficient for each particle. This represents the actual power output of the lithium battery at the previous time step. This is the upper limit of the allowable rate of change of power, which is dynamically narrowed.

[0013] An adaptive inertia weight is constructed by combining the survival margin index, and the solution is obtained using an improved particle swarm optimization algorithm. The specific process is as follows: constructing an adaptive inertia weight that is dimension-independent for heterogeneous devices. : ; in, and These are the upper and lower bound constants of the inertia weight, respectively; This represents the current iteration number. This represents the maximum number of iterations. A survival margin sensitive factor; As an indicator of survival margin; the first Individual particles in the device dimension speed With position The iterative update expression is: ; ; in, and To learn step size; and for Random numbers that are uniformly distributed between them; For the first Particles in dimension The individual's historical best position; For the entire population in dimension The globally optimal position.

[0014] This invention provides an aging regression and collaborative management system for Antarctic energy systems, characterized in that the system comprises: Acquisition Unit: Acquires multidimensional operational physical quantities in the polar region and maps the operational losses of various heterogeneous devices within the system into absolute energy loss values ​​that can be processed. Forward recursive model generation unit: Combining the global temperature penalty benchmark and the physical cross-constraint relationship between heterogeneous devices, the real-time health status of the equipment is obtained through the multi-source device coupled aging forward recursive model. Weighting calculation unit: Calculates the expected remaining physical life based on the real-time health status of the equipment, compares it with the polar summer flood logistics replenishment time window to construct a survival margin index, and dynamically generates real-time penalty weight coefficients; The optimization unit performs nonlinear smooth mapping on the real-time net power of the system, constructs adaptive inertial weights in conjunction with the survival margin index, and uses an improved particle swarm optimization algorithm to solve for the optimal power allocation ratio that minimizes the total energy loss of the system's throughput. The optimal power allocation ratio is then converted into an absolute active power scheduling command and sent to the underlying physical hardware of the microgrid for execution.

[0015] This invention provides a method and system for aging regression and collaborative management of Antarctic energy systems. The invention constructs a forward regression model for multi-source equipment under extreme cold and blizzard conditions. This model not only quantifies the global penalty effect of the polar cold environment on the nonlinear surge in the physical internal resistance of electrochemical equipment, but also breaks through the barriers of traditional isolated component impact analysis, establishing a dynamic physical constraint logic where high-frequency energy storage equipment constrains medium-frequency energy storage equipment, and medium-frequency energy storage equipment guides low-frequency conversion equipment. When frequent transient high-frequency disturbances in the polar region cause degradation of front-end filtering equipment, the system can automatically narrow the response depth and allowable rate of change of back-end core equipment based on this cross-coupling relationship, effectively blocking the accelerated aging phenomenon of multi-source equipment cascading caused by power stress transfer, and significantly improving the overall operating life and fault tolerance of the microgrid system under extreme high-dynamic shocks.

[0016] Meanwhile, this invention uses the polar summer flood logistics supply time window as the highest-level absolute spatiotemporal constraint, employing a non-uniform adaptive weight configuration and intervention mechanism based on polar survival principles. By predicting the remaining physical lifespan of each key component in real time and comparing it with the remaining absolute time before the next icebreaker arrives at the research station, when a core device is detected to be in a precarious state of premature decommissioning, it can spontaneously break the conventional equal optimization path, generating extreme penalty weights that exhibit a non-linear surge. This mechanism forcibly guides the microgrid to evolve in a direction that protects endangered equipment, ensuring that all core components survive the polar winter lockdown period and preventing a catastrophic system-wide power outage caused by the premature collapse of a single critical node.

[0017] This invention improves the global optimization algorithm to address the extreme and abrupt changes in Antarctic meteorological data. By introducing a resource signal smoothing and dimensionality reduction mechanism, it effectively avoids the malicious interference of high-frequency spikes from blizzards on the particle search interval, ensuring the gradient stability and convergence continuity of the algorithm model under extreme perturbations. Simultaneously, considering the survival crisis gradient of equipment, for equipment dimensions in high-risk situations, it automatically reduces the search step size for conservative and refined optimization, avoiding secondary physical impacts on fragile hardware caused by trial-and-error high-power commands. This ensures that complex energy optimization strategies can be transformed into precise electrical execution actions at the underlying level without loss, guaranteeing the absolute survivability and long-term power supply stability of the Antarctic independent energy system in an unattended environment. Attached Figure Description

[0018] Figure 1 This is a flowchart illustrating the collaborative management of Antarctic energy systems according to the present invention; Figure 2 A comparison chart of power curves allocated to each module of the present invention; Figure 3 This is a comparison chart of the SOH state and cumulative energy loss of each module in this invention. Detailed Implementation

[0019] This embodiment provides a collaborative management system for Antarctic energy systems. This system serves as the hardware foundation for executing subsequent multi-source equipment coupling aging recursion and collaborative scheduling methods, and is deployed in the extremely cold, wind-solar fluctuating, and strictly seasonal logistics resupply window constraints of polar research stations. The collaborative management system is built upon an AC / DC hybrid microgrid topology and specifically includes: a renewable energy generation module, a multi-source heterogeneous energy storage module, a gap-based basic power supply module, and a data acquisition and collaborative control module. Each module is interconnected via AC / DC buses and interacts with the data acquisition and collaborative control module via an industrial communication bus.

[0020] Specifically, renewable energy generation modules, serving as the primary energy source for the polar microgrid, include wind power generation units and photovoltaic (PV) power generation units. For polar regions, the wind power generation units preferably employ icing-resistant wind turbines equipped with actively heated de-icing blades; the PV power generation units preferably utilize double-sided glass PV modules adapted to the high albedo environment of the polar regions. The wind power generation units and PV power generation units are connected to the system bus via corresponding AC / DC converters and DC / DC converters, respectively.

[0021] The multi-source heterogeneous energy storage module is designed to address power fluctuations and energy transfer needs at different time scales. The module consists of three physical devices connected in parallel with different response frequency bands, specifically including a supercapacitor energy storage unit, a lithium battery energy storage unit, and a hydrogen energy conversion and storage unit.

[0022] The supercapacitor energy storage unit is used to suppress high-frequency power surges ranging from seconds to minutes caused by polar blizzards. A low-temperature resistant, wide-temperature-range supercapacitor array is preferred to withstand the high-frequency pulsating overstress of the system. The lithium-ion battery energy storage unit is used to handle short-to-medium-term energy balance and tracking of the microgrid on an hourly basis. A lithium iron phosphate battery cluster equipped with a self-contained thermal management system is preferred to avoid irreversible lithium plating damage to the cells caused by extreme cold environments.

[0023] The hydrogen energy conversion and storage unit is used to handle the inter-day or long-cycle energy transfer caused by the alternation of polar night and polar day. This unit includes a proton exchange membrane electrolyzer, a high-pressure hydrogen storage tank, and a proton exchange membrane fuel cell. The electrolyzer and fuel cell preferably employ proton exchange membrane technology to meet the requirements of rapid cold start and a wide power regulation range under polar conditions.

[0024] The gap-deficit basic power supply module is configured as the last guarantee for the system's power supply, including a low-temperature start-up diesel generator set and its matching oil storage tank. This module is only connected as the highest priority rigid support power source when the multi-source heterogeneous energy storage module is completely depleted or when critical energy storage equipment reaches its lifespan constraint boundary. The data acquisition and collaborative control module is the data hub and scheduling execution carrier of the entire energy system, including an environmental meteorological monitoring station, distributed low-level state acquisition devices, and a central energy management host. The environmental meteorological monitoring station is installed outside the research station and includes an ultrasonic anemometer, a light transmitter, and a wide-range environmental thermometer for real-time acquisition of polar environmental temperature and wind and solar meteorological data. The distributed low-level state acquisition devices are deployed at the low-level interfaces of various power generation and energy storage devices, including power sensors for acquiring real-time voltage and current signals of the bus and each branch, and monitoring terminals for acquiring the physical status of equipment such as lithium battery state of charge, supercapacitor terminal voltage, fuel cell and electrolyzer start-stop count, and diesel engine oil level. The central energy management host contains a high-performance industrial control computer or microprocessor. Its internal storage unit is pre-programmed with parameters for the absolute logistics resupply time window between the polar research station and the arrival of the next summer supply ship, as well as the baseline values ​​for the rated total throughput of various devices throughout their entire life cycle. The central energy management host receives real-time data uploaded from environmental meteorological monitoring stations and distributed low-level status acquisition devices via fieldbus. It executes heterogeneous device coupled aging recursive calculations and global optimization strategies locally, and issues specific pulse width modulation or power command control instructions to the power converters of each device, thereby realizing the physical control of the entire Antarctic microgrid system.

[0025] Next, based on the provided collaborative management system architecture, this embodiment explains the polar-specific operating condition data acquisition and physical dimension unification in the collaborative management method for Antarctic energy systems. In extreme isolated operating environments such as polar regions, the economic assessment method for equipment aging based on conventional monetary systems suffers from severe physical distortion. Because polar microgrids are physically isolated from the outside world during the long polar winter, equipment wear and tear cannot be compensated for by purchasing spare parts at any time. Therefore, this embodiment starts from engineering thermodynamics and physical fatigue properties, and simultaneously completes the physical dimension unification of all heterogeneous equipment in the system during the data acquisition phase, strictly mapping the operating losses of all equipment to the absolute energy loss value that characterizes the system's extreme survivability, using this as the benchmark for subsequent multi-source collaborative scheduling.

[0026] The data acquisition and collaborative control module uses a fixed sampling period. Real-time reading of underlying sensor data from the system. The discrete time step is defined as... At any time step The system acquires the following polar-specific multidimensional operational physical quantities: real-time net power of the system. Its mathematical expression is: in, For renewable energy generation modules at the current time step The collected real-time power generation includes the sum of wind and solar power output. For polar scientific research stations at the current time step Real-time load power. Under the disturbance of polar blizzards... The high-frequency, large-amplitude fluctuations directly determine the direction of real-time power allocation commands for various energy storage and power supply units in the system.

[0027] Obtain real-time polar ambient temperature Extreme cold temperatures are a disturbance factor that distinguishes Antarctic microgrids from conventional microgrids. A sudden drop in ambient temperature will cause a nonlinear increase in battery internal resistance and cold start damage to the electrolyzer. This parameter is collected in real time and used as a global penalty benchmark for subsequent equipment coupling aging recursion.

[0028] Obtain the real-time remaining fuel volume of the diesel generator set , This indicates the ultimate bottom line for ensuring the supply of microgrid systems.

[0029] Obtaining the time window for polar summer flood logistics and supply . This represents the remaining absolute physical time (in hours) until the next supply shipment from the polar icebreaker arrives at the research station. In polar regions, the remaining physical lifetime of all microgrid core equipment must be strictly greater than [a certain value]. Otherwise, there will be a fatal risk of a complete system power outage. The physical remaining lifetime is the absolute spatiotemporal constraint boundary of the adaptive priority of this application.

[0030] To enable devices with different electrochemical response frequency bands and different mechanical wear mechanisms to be constrained and optimized under the same objective function, the data acquisition and collaborative control module uniformly converts the health degradation of various devices into the absolute value of cumulative throughput energy loss. The equipment set number is These correspond to lithium batteries, supercapacitors, fuel cells, electrolyzers, and diesel generator sets, respectively.

[0031] For electrochemical and hydrogen energy conversion devices in multi-source heterogeneous energy storage modules ( Its dimension-unified mapping formula is constructed as follows: in, For equipment At time step The real-time health status indicator ranges from 1 to the lower limit threshold of the device's lifespan. For equipment The energy mapping reference constant. For the heterogeneous device properties in the polar regions, the energy mapping physical reference constant... It is further precisely defined as: For lithium battery packs: in, The rated total throughput of the lithium battery energy storage unit over its entire life cycle, calibrated under polar standard operating conditions (unit: ).

[0032] For supercapacitor arrays: in, The rated total throughput of the supercapacitor energy storage unit over its entire life cycle, as physically calibrated at the factory (unit: ).

[0033] For proton exchange membrane fuel cells: in, The rated total operating hours over the entire lifecycle of the fuel cell; This represents the rated output power of the fuel cell. The product of these two values ​​represents the absolute upper limit of the total physical electrical energy that the fuel cell can output until it is scrapped.

[0034] For proton exchange membrane electrolyzers: in, This refers to the rated total operating hours throughout the entire life cycle of the electrolytic cell; This represents the rated absorption power of the electrolyzer. The product of these two values ​​represents the total excess physical electrical energy that the electrolyzer can absorb before the end of its lifespan.

[0035] For the diesel generator set in the foundation power supply module of the gap ( It lacks an electrochemical SOH index; its physical equivalent energy loss is directly determined by the irreversible fossil fuel consumption under extreme Antarctic conditions, and its cumulative equivalent energy loss value... The formula for the continuous recursive mapping is: in, From the initial time to the current time step Historical iteration variables; For diesel generator sets at time steps The calculation relationship for the generated single-step fuel consumption volume is as follows: ; This is for the density of special low-temperature diesel fuel configured for polar environments; This refers to the lower heating value of this special low-temperature diesel oil; This refers to the energy conversion efficiency of the diesel generator set.

[0036] Through data collection and unified processing, this embodiment quantifies all hardware losses in polar microgrids, including wind, solar, diesel, energy storage, and hydrogen, into permanent losses of physical energy that can be guaranteed for future dispatch at Antarctic research stations, and converts the units to kilowatt-hours.

[0037] Next, based on the completion of polar-specific operating condition data collection and dimensional unification, this embodiment describes the construction of a forward recursive model of state of health (SOH) based on the cross-coupling of polar temperature and physical state in the collaborative management method for Antarctic energy systems.

[0038] In traditional microgrid energy management, the lifespan degradation of various devices is usually considered as an independent, univariate evolution process. However, in the extreme isolated conditions of Antarctica, microgrid systems face global physical intervention from the frigid environment. Furthermore, under frequent blizzard disturbances, the health degradation of a particular frequency band regulating device will inevitably lead to the transfer of power stress to other devices, thereby triggering cascaded accelerated aging of multi-source devices. Therefore, this embodiment incorporates the nonlinear penalty effect of extreme cold temperatures and the physical cross-constraint relationships between response devices of different frequency bands into the recursive process, constructing a forward recursive model for the coupled aging of multi-source devices in polar regions.

[0039] The specific implementation steps are as follows: A global temperature penalty benchmark based on the extreme cold characteristics of Antarctica is constructed. Extreme cold temperatures directly increase the viscosity of the electrolyte and decrease the ionic conductivity of electrochemical devices, thereby triggering a nonlinear increase in physical internal resistance. This embodiment considers the real-time polar ambient temperature. Construct the extreme cold temperature penalty coefficient under continuous time steps : in, The standard operating reference temperature (in Kelvin) for multi-source heterogeneous energy storage modules during factory calibration. ); This is a characterization coefficient for temperature sensitivity determined through field bench experiments in polar regions. When polar blizzards cause... During a sudden drop, It increases exponentially and rapidly. This parameter will act as a global intervention factor in the subsequent aging recursive equations of temperature-sensitive lithium batteries and electrolyzers.

[0040] Secondly, the aging process of the high-frequency power suppression equipment and its cross-constraints on the medium-frequency energy storage equipment are implemented. When the power of the Antarctic wind source experiences high-frequency, drastic fluctuations, the supercapacitor energy storage unit absorbs pulsating overstress on a second- to minute-level scale. The real-time health status of the supercapacitor array is also assessed. The calculation is performed using a single-step forward recursive approach: in, This refers to the health status of the supercapacitor at the previous time step. This refers to the actual high-frequency charging and discharging power undertaken by the supercapacitor at the current time step; This represents the voltage stress deviation ratio of the current supercapacitor terminal voltage relative to the rated voltage. and These are the reference attenuation coefficients driven by power stress and voltage stress, respectively; This is a maximum permissible attenuation safety limit set for single-step operation under polar conditions to prevent numerical overflow.

[0041] With the long-term operation of polar microgrids, supercapacitors suffer from high dynamic overstress. As the current gradually decreases, its high-frequency filtering capability deteriorates. At this point, high-frequency power pulsations will penetrate the supercapacitor and directly impact the slower-responding lithium battery energy storage unit. To prevent physical breakdown of the lithium battery caused by large-rate sudden current surges, this embodiment constructs a cross-physical constraint mechanism between the supercapacitor and the lithium battery. Based on the health status of the supercapacitor in the previous time step, the system dynamically narrows the upper limit of the allowable power change rate of the lithium battery in the current time step. : in, The rated maximum power change rate specified by the manufacturer for lithium batteries; This is the lower limit threshold for the permissible safe lifespan of equipment in a microgrid system. To limit the smoothing factor.

[0042] When the supercapacitor ages severely, the system is forced to reduce the high-frequency scheduling depth of the lithium battery, sacrificing some transient power tracking accuracy to ensure the safety of the lithium battery.

[0043] Based on these cross-constraints, the aging process of the lithium battery energy storage units is further recursively implemented. Lithium batteries are responsible for short-term energy balance in the microgrid, and their aging is driven by a combination of power load, depth of discharge, and penalties from extremely cold temperatures. Their health status... The recurrence relation is: in, This represents the actual charging and discharging power allocated to the lithium battery at the current time step, and the single-step change in this power is strictly constrained by the aforementioned derived upper limit of the dynamic change rate: ; This represents the increment of the discharge depth at the current time step. and These are the cycle decay coefficients for power load and depth of discharge, respectively.

[0044] The lithium battery performs cross-guidance of the long-cycle hydrogen energy conversion equipment and completes the aging process of the hydrogen energy equipment. During long-cycle operation with alternating polar nights and days, if the lithium battery frequently undergoes deep charging and discharging, its... The lifespan of the fuel cell will be rapidly approaching its lower limit. To alleviate the pressure on lithium battery supply, this embodiment constructs a dynamic wake-up constraint for hydrogen energy equipment based on the health status of the lithium battery. The system dynamically adjusts the start-up power dead zone threshold of the fuel cell according to the real-time status of the lithium battery. : in, This is the rated baseline start-up threshold for the fuel cell. When the system's real-time net power... At that time, the fuel cell is forcibly ignited and started. This cross-guidance mechanism indicates that when the lithium battery ages severely, the start-up threshold of the fuel cell decreases accordingly, and the system will mobilize long-cycle hydrogen energy in advance to bear the load, thereby preventing the vicious degradation of the lithium battery.

[0045] Under the aforementioned guidance mechanism, the proton exchange membrane fuel cell and the electrolyzer each perform their own health state recursion. The aging of the fuel cell is mainly affected by mechanical stress damage from operating power dwell time and frequent start-stop cycles, and its recursive formula is as follows: in, This refers to the actual output power of the fuel cell; This represents the fuel cell load attenuation coefficient. The fixed life loss equivalent caused by a single ignition start-stop cycle; This is a Boolean start / stop event indicator variable, which is used at time step... The value is 1 when the startup action is triggered, and 0 otherwise.

[0046] Considering that the proton exchange membrane inside the electrolyzer is prone to freeze-hardening in extremely cold environments, and frequent power injection would exacerbate the risk of mechanical tearing of the membrane, an extreme cold temperature penalty coefficient is introduced into the aging recursive process of the electrolyzer, and its expression is as follows: in, This represents the current actual hydrogen production capacity absorbed by the electrolyzer. This is the reference power load attenuation coefficient for the electrolytic cell.

[0047] Based on this, this embodiment constructs a recursive model of Antarctic temperature characteristics, where high frequency (supercapacitor) constrains mid-frequency (lithium battery), and mid-frequency (lithium battery) guides low frequency (fuel cell / electrolyzer).

[0048] This embodiment, based on the completed forward recursive model of multi-source equipment coupling aging, illustrates the adaptive optimization weight configuration based on the resupply window period in the collaborative management method for Antarctic energy systems.

[0049] In conventional microgrid optimization scheduling, multi-objective optimization algorithms typically assign quantitative static weights to the losses of various equipment and solve them jointly. However, in the extreme isolation of Antarctica, microgrid systems face blockades lasting for months, and the physical wear and tear of equipment cannot be compensated for by purchasing spare parts on the fly. If the remaining lifespan of a core energy storage or power supply component in the system is exhausted before the arrival of the polar summer supply ship, it will trigger a fatal survival crisis of system-wide power outage. Therefore, this embodiment uses the polar summer supply time window... As a spatiotemporal constraint, a non-equal adaptive weight penalty mechanism based on the polar survival law was constructed through survival prediction verification, providing dynamic emphasis guidance for global optimization.

[0050] Calculate various electrochemical and hydrogen energy conversion devices in multi-source heterogeneous energy storage modules ( The real-time health degradation rate and expected remaining physical lifetime of the device are calculated. Considering the significant noise in single-step health degradation values ​​caused by the drastic transient power jumps resulting from frequent Antarctic blizzards, this embodiment introduces a sliding time window based on historical observations during polar meteorological shortwave periods. To filter high-frequency disturbances, a smooth decay rate is extracted to reveal the true physical degradation trend of the equipment. : in, For the current time step Real-time health status of the device obtained through a forward recursive model; The device's historical health status at the start of the sliding time window; The sampling period is the duration. Based on the extracted smooth decay rate, the threshold threshold for each device to reach the allowable safe lifespan of the microgrid system under the current extreme operating conditions is calculated backward. Expected remaining physical lifetime (Unit: hour): By cross-referencing the expected remaining physical lifetime of the equipment with the polar resupply boundary, a survival margin index specific to polar microgrids is constructed. This metric is defined as the ratio of the equipment's own exhaustion time to the arrival time of external rescue: This survival margin index This directly indicates the level of crisis faced by each piece of equipment during its Antarctic deployment mission. When This indicates that the equipment has sufficient lifespan to support the arrival of the polar summer supply ship, and the system is in a relatively safe state; when This indicates that the current aging trajectory of the equipment will lead to its premature scrapping before resupply arrives, triggering a polar survival endangerment warning.

[0051] A non-uniform adaptive optimization weight allocation model is constructed based on the survival margin index. The system dynamically generates real-time penalty weight coefficients for each device. Its segmented model is defined as follows: in, For equipment The basic energy loss weight under the initial safe state is the inherent preference of the system for the operating loss of the equipment under normal operating conditions. This is the sensitivity constant for endangered species punishment; adjusting this parameter can control the severity of crisis intervention.

[0052] In this step, when a device (such as a lithium battery) gets stuck... When the situation is endangered, because the denominator is less than 1 and the superposition is... The exponential amplification effect, and its corresponding real-time penalty weighting coefficient The loss will increase exponentially and non-linearly. The penalty weighting coefficient for this surge will be directly output to the global optimization objective function in the next stage. The system defines the physical loss of this critically ill device as the current maximum cost, forcibly guiding subsequent optimization algorithms to prioritize sacrificing the cycle life of other healthy devices, even forcibly activating the basic power supply module, in order to significantly reduce the power scheduling instructions allocated to this critically ill device and ensure its physical lifespan survives the extreme winter until... The moment has arrived.

[0053] This embodiment, based on the non-uniform adaptive optimization weight configuration based on the resupply window period, illustrates the improved multi-objective cooperative optimization of particle swarm in the cooperative management method for Antarctic energy systems.

[0054] After obtaining the adaptive penalty weights containing the survival crisis gradient, the system needs to solve for the optimal power allocation ratio of each heterogeneous device at the current time step using a global optimization algorithm. However, frequent blizzards in Antarctica can cause extreme abrupt peaks in the system's net power demand. Conventional swarm intelligence optimization algorithms are prone to distorting particle gradient information due to abrupt changes in the search space when dealing with such data containing extremely large singular values, leading to local optima or numerical divergence. Furthermore, the search step size in all dimensions of the conventional PSO algorithm is usually controlled by a uniform inertia weight, failing to reflect the unique device-level survival differences in the polar regions. Therefore, this embodiment constructs an improved adaptive PSO algorithm by incorporating a survival margin index.

[0055] To eliminate the detrimental effect of high-frequency power spikes caused by Antarctic blizzards on the search range of the optimization algorithm, this embodiment uses the real-time net power of the system collected in the first stage. Perform nonlinear smoothing mapping to generate net compression power index : in, This is a sign function used to represent the current overall charging and discharging direction of the microgrid; This is a polar weather fluctuation compression factor used to control the smoothing intensity of the maxima; This represents the absolute maximum net power peak recorded by the polar microgrid in the historical weather database. Through this logarithmic dimensionality reduction process, the power including extreme spikes is mapped to... The smooth interval provides stable boundary constraints for the particle swarm, preventing numerical explosion in the early stages of optimization.

[0056] Define the particle search space and objective function for multi-source cooperative optimization. Define the population size of the improved PSO algorithm as follows. The maximum number of iterations is In the first In the nth iteration, the 1st The position vector of each particle is defined as the set of power allocation ratio coefficients for heterogeneous devices. These correspond to lithium batteries, supercapacitors, fuel cells, and electrolyzers, respectively. The actual power allocation for each device... The physical mapping relationship with the allocation ratio coefficient is as follows: Furthermore, the physical constraints of instantaneous power balance in microgrids must be strictly met: .

[0057] in The gap base allocation ratio for diesel generator sets is only activated when the energy storage module reaches its physical limit.

[0058] Constructing a multi-objective fitness function for spatiotemporal boundaries The core of this function is to minimize the total throughput energy loss of the system. in, Based on logistics replenishment time window Dynamically generated real-time penalty weight coefficients; The absolute energy loss of the device in a single step is calculated by the second-stage cross-coupling forward recursive model under the current particle allocation scheme. It is a static rejection weight of extremely high order of magnitude, used to strongly suppress the consumption of fossil fuels under non-extreme power shortage conditions; This is a cross-penalty item.

[0059] Regarding the cross-physical constraints of supercapacitors on lithium batteries derived in the second stage... Cross-penalty items for: in, This is a severe punishment for the multiplier; This represents the actual power output of the lithium battery at the previous time step. This penalty ensures that, as the supercapacitor ages, it will never generate drastic power surges that could compromise the physical safety of the lithium battery.

[0060] The algorithm employs particle-based adaptive iterative updates guided by polar survivability margins. In the conventional PSO algorithm, particle velocity updates determine the exploration and development capabilities for optimization. To reflect the physical requirement of "whoever is endangered, should receive refined protection" in polar microgrids, this embodiment constructs adaptive inertial weights independent of the dimensions of heterogeneous devices. This parameter is directly controlled by the survival margin index extracted in the third stage. : in, and These are the upper and lower bound constants of the inertia weight, respectively; It is a survival margin sensitive factor.

[0061] No. Individual particles in the device dimension speed With position The iterative update expression is: in, and To learn step size; and for Random numbers that are uniformly distributed between them; For the first Particles in dimension The individual's best historical position experienced above; For the entire population in dimension The globally optimal position.

[0062] When the equipment corresponding to a certain dimension (such as a fuel cell) is in a critical state ( )hour, This will be forcibly pulled down. This forces all particles to explore the power distribution ratio of the fuel cell ( In this dimension, it completely loses the ability to make large leaps and instead focuses on the currently known safe optimal solution. and This involves a conservative fine-tuning process with minimal step sizes. This dimensionally decoupled, non-uniform adaptive convergence effectively avoids secondary impacts on critical components caused by trial-and-error high-power allocation during the optimization process, thus ensuring the physical survival limit of the Antarctic heterogeneous energy equipment during resupply vacuum periods. When the number of iterations reaches... The iteration terminates when the global optimal fitness value of the population is continuously stable, and the current global optimal position vector is output. This represents the optimal power allocation ratio for the system to cope with multi-source fluctuations in the polar region at the current time step.

[0063] Based on multi-objective collaborative optimization, this embodiment provides the final execution step in the collaborative management method for Antarctic energy systems: the issuance of microgrid underlying physical control commands.

[0064] After the power allocation ratio of the multi-source devices in the polar region is determined, the system needs to transform the optimal solution at the algorithm level into physical electrical commands that can drive the operation of the underlying hardware of the microgrid, thereby completing closed-loop physical control within a complete sampling cycle.

[0065] The central energy management host in the data acquisition and collaborative control module extracts the globally optimal power allocation ratio vector output by the improved particle swarm optimization algorithm after convergence. The system multiplies this set of ratio parameters by the actual net power required by the system at the current time step, directly restoring it into an absolute active power scheduling command for each type of heterogeneous equipment.

[0066] Subsequently, the central energy management host accurately sends the aforementioned absolute active power dispatching instructions to the distributed status acquisition devices, power converters, and dedicated electrical controllers at the underlying level of each physical device through the field industrial communication bus, according to the preset communication protocol specifications.

[0067] During the actual execution of the underlying physical hardware, each module strictly follows the coordinated scheduling instructions to carry out its work: Upon receiving high-frequency dispatch commands, the supercapacitor energy storage unit in the multi-source heterogeneous energy storage module immediately performs rapid charging and discharging operations through its underlying bidirectional DC-DC converter, absorbing or compensating for the transient pulse energy caused by polar blizzards on-site. Upon receiving medium-to-short-term power commands, the lithium battery energy storage unit performs smooth base load tracking to maintain the stability of the microgrid system's bus voltage and frequency. To address the long-term energy transfer demands across days and nights caused by polar nights, the electrolyzers or fuel cells in the hydrogen energy conversion and storage unit perform corresponding ignition start-stop and smooth power ramp-up operations according to commands. Meanwhile, the foundation power supply module only starts the diesel generator set and consumes the configured low-temperature special diesel fuel for foundation power generation when it receives the highest-level rigid support command.

[0068] Specifically, under the unique survival intervention mechanism of the polar regions, if the system assesses in the early stages that a critical piece of equipment (such as a lithium battery) faces the risk of premature scrapping before the arrival of summer logistics supplies, the actual dispatch power command received by this endangered equipment in this embodiment will be forcibly and significantly reduced due to the extremely high attenuation penalty weight it was given during the optimization phase. At the underlying level, the equipment will automatically enter a light-load operation or standby hibernation mode, and its original power load will be smoothly transferred to other equipment or diesel generator sets within the system with sufficient health margins.

[0069] Once all the power converters of the underlying devices have completed their command responses and physical power outputs, the collaborative management process for the current discrete time step ends. The data acquisition and collaborative control module then enters the next sampling cycle, continuing to read new polar meteorological data and device physical status, repeating this cycle. Through the precise issuance and execution of these underlying physical commands, this embodiment fundamentally ensures the absolute survivability and long-term power supply of the Antarctic independent microgrid system under extreme conditions and during prolonged logistical blockades.

[0070] Those skilled in the art will understand that all or part of the processes in the methods of the above embodiments can be implemented by a computer program instructing related hardware. The computer program can be stored in a non-volatile computer-readable storage medium, and when executed, it can include the processes of the embodiments of the above methods. Any references to memory, storage, databases, or other media used in the embodiments provided in this application can include non-volatile and / or volatile memory. Non-volatile memory can include read-only memory (ROM), programmable ROM (PROM), electrically programmable ROM (EPROM), electrically erasable programmable ROM (EEPROM), or flash memory. Volatile memory can include random access memory (RAM) or external cache memory. By way of illustration and not limitation, RAM is available in various forms, such as static RAM (SRAM), dynamic RAM (DRAM), synchronous DRAM (SDRAM), dual data rate SDRAM (DDRSDRAM), enhanced SDRAM (ESDRAM), synchronous link DRAM (SLDRAM), Rambus direct RAM (RDRAM), direct memory bus dynamic RAM (DRDRAM), and memory bus dynamic RAM (RDRAM), etc.

[0071] In this specification, the same or similar parts between the various embodiments can be referred to mutually. Each embodiment focuses on describing the differences from other embodiments. In particular, the descriptions of the embodiments described later are relatively simple, and relevant parts can be referred to the descriptions of the foregoing embodiments.

[0072] The above description is merely a specific embodiment of this application, but the scope of protection of this application is not limited thereto. Any variations or substitutions that can be easily conceived by those skilled in the art within the technical scope disclosed in this application should be included within the scope of protection of this application. Therefore, the scope of protection of this application should be determined by the scope of the claims.

Claims

1. A method for aging regression and collaborative management of Antarctic energy systems, characterized in that, Includes the following steps: Obtain multidimensional operational physical quantities in the polar regions and uniformly map the operational losses of various heterogeneous devices within the system to the absolute loss value of the energy that can be transmitted. By combining the global temperature penalty benchmark and the physical cross-constraint relationship between heterogeneous devices, the real-time health status of the devices is obtained through a multi-source device coupled aging forward recursive model. The expected remaining physical life is calculated based on the real-time health status of the equipment, and then compared with the time window for summer logistics replenishment in polar regions to construct a survival margin index, and a real-time penalty weight coefficient is dynamically generated. The real-time net power of the system is nonlinearly smoothed and an adaptive inertial weight is constructed in combination with the survival margin index. An improved particle swarm algorithm is used to solve for the optimal power allocation ratio that minimizes the total throughput energy loss of the system. The optimal power allocation ratio is converted into an absolute active power scheduling command and sent to the underlying physical hardware of the microgrid for execution.

2. The aging regression and collaborative management method for Antarctic energy systems according to claim 1, characterized in that, The operating losses of each heterogeneous device within the system are uniformly mapped to the absolute loss value of the energy throughput. Specifically, this includes the cumulative absolute loss value of the energy throughput for electrochemical and hydrogen energy conversion devices in the multi-source heterogeneous energy storage module. The mapping formula is: ； in, For equipment In time Real-time health status indicators; For equipment The energy mapping reference constant; These correspond to lithium batteries, supercapacitors, fuel cells, and electrolyzers, respectively; for the diesel generator set in the gap-base power supply module, its cumulative equivalent energy loss value... The formula for the continuous recursive mapping is: ； in, For historical iteration variables; For diesel generator sets at time steps The volume of fuel consumed in a single step; Density of special low-temperature diesel fuel for polar environments; The lower heating value of special low-temperature diesel oil; This refers to the energy conversion efficiency of the diesel generator set.

3. The aging regression and collaborative management method for Antarctic energy systems according to claim 1, characterized in that, Combining the global temperature penalty benchmark and the physical constraints between heterogeneous devices, a multi-source device coupled aging forward recursive model is used to obtain the real-time health status of the devices and construct the extreme cold temperature penalty coefficient under continuous time steps. : ； in, This is the standard operating reference temperature for multi-source heterogeneous energy storage modules; This is a coefficient characterizing temperature sensitivity; Real-time polar ambient temperature; real-time health status of the supercapacitor array. The calculation is performed using a single-step forward recursive approach: ； in, This refers to the health status of the supercapacitor at the previous time step. This refers to the actual high-frequency charging and discharging power undertaken by the supercapacitor at the current time step; This represents the voltage stress deviation ratio of the current supercapacitor terminal voltage relative to the rated voltage. and These are the reference attenuation coefficients driven by power stress and voltage stress, respectively; The sampling period duration; This is the maximum permissible attenuation safety limit set for single-step operation under polar conditions.

4. The aging regression and collaborative management method for Antarctic energy systems according to claim 3, characterized in that, The physical constraints between heterogeneous devices include the cross-physical constraint mechanism of supercapacitors on lithium batteries. Specifically, based on the health status of the supercapacitor in the previous time step, the upper limit of the allowable power change rate of the lithium battery in the current time step is dynamically narrowed. : ； in, This refers to the rated maximum power change rate of the lithium battery. This is the lower limit threshold for the permissible safe lifespan of equipment in a microgrid system. To limit the smoothing factor; Perform an aging regression analysis of the lithium battery energy storage unit to determine its health status. The recurrence relation is: ； in, This represents the actual charging and discharging power allocated to the lithium battery at the current time step, and the single-step change in this power is constrained by the upper limit of the allowable power change rate. ; This represents the increment of the discharge depth at the current time step. and These are the cycle decay coefficients for power load and depth of discharge, respectively.

5. The aging regression and collaborative management method for Antarctic energy systems according to claim 4, characterized in that, The physical constraints between heterogeneous devices also include dynamic wake-up constraints for hydrogen energy devices based on the health status of lithium batteries. The system dynamically adjusts the dead zone threshold of the fuel cell's start-up power according to the real-time status of the lithium batteries. : ； in, This is the rated reference start-up threshold for the fuel cell; Under the dynamic wake-up constraint of hydrogen energy equipment, proton exchange membrane fuel cells and electrolyzers respectively execute their own health state recursive formulas: ； ； in, This refers to the actual output power of the fuel cell; This refers to the rated output power of the fuel cell; This represents the fuel cell load attenuation coefficient. The fixed life loss equivalent caused by a single ignition start-stop cycle; This is a Boolean start / stop event indicator variable; This represents the current actual hydrogen production capacity absorbed by the electrolyzer. This is the rated absorption power of the electrolytic cell; This is the reference power load attenuation coefficient for the electrolytic cell.

6. The aging regression and collaborative management method for Antarctic energy systems according to claim 1, characterized in that, The expected remaining physical lifespan is calculated based on the real-time health status of the equipment. This lifespan is then compared with the polar summer flood logistics resupply time window to construct a survival margin index. Real-time penalty weighting coefficients are dynamically generated. Specifically, the smoothed decay rate of the equipment's true physical degradation trend is extracted. : ； in, A sliding time window for historical observation; The device's historical health status at the start of the sliding time window; The sampling period duration; Calculate the expected remaining physical life of each device. And construct a survival margin index : ； ； in, This is the lower limit threshold for the permissible safe lifespan of equipment in a microgrid system. Establish a time window for logistics resupply during the polar summer flood season; dynamically generate real-time penalty weight coefficients for each piece of equipment based on survivability margin indicators. : ； in, For equipment The base energy loss weight under the initial safe state; This is the sensitivity constant for endangered species punishment.

7. The aging regression and collaborative management method for Antarctic energy systems according to claim 1, characterized in that, A nonlinear smoothing mapping is performed on the real-time net power of the system to generate a compression net power index. Its mathematical expression is: ； in, It is a symbolic function; This refers to the system's real-time net power. The polar weather fluctuation compression factor; The absolute maximum net power peak value of the polar microgrid; when solving using the improved particle swarm optimization algorithm, a multi-objective fitness function is constructed for the spatiotemporal boundary. : ； in, For the first In the nth iteration A set of power allocation ratio coefficients for each particle; This is the real-time penalty weighting coefficient; and These are the absolute loss value of single-step equivalent energy and the cumulative equivalent energy loss value calculated under the current particle allocation scheme, respectively. Static rejection weight; This is a cross-penalty item.

8. The aging regression and collaborative management method for Antarctic energy systems according to claim 7, characterized in that, Cross-penalty items The calculation formula for the cross-physical constraint setting of supercapacitors on lithium batteries is as follows: ； in, This is a severe punishment for the multiplier; For the first In the nth iteration The lithium battery power allocation ratio coefficient for each particle. This represents the actual power output of the lithium battery at the previous time step. This is the upper limit of the allowable rate of change of power, which is dynamically narrowed.

9. The aging regression and collaborative management method for Antarctic energy systems according to claim 7, characterized in that, An adaptive inertia weight is constructed by combining the survival margin index, and the solution is obtained using an improved particle swarm optimization algorithm. The specific process is as follows: constructing an adaptive inertia weight that is dimension-independent for heterogeneous devices. : ； in, and These are the upper and lower bound constants of the inertia weight, respectively; This represents the current iteration number. This represents the maximum number of iterations. A survival margin sensitive factor; As an indicator of survival margin; the first Individual particles in the device dimension speed With position The iterative update expression is: ； ； in, and To learn step size; and for Random numbers that are uniformly distributed between them; For the first Particles in dimension The individual's historical best position; For the entire population in dimension The globally optimal position.

10. An aging recursion and collaborative management system for Antarctic energy systems, characterized in that, The system includes: Acquisition Unit: Acquires multidimensional operational physical quantities in the polar region and maps the operational losses of various heterogeneous devices within the system into absolute energy loss values ​​that can be processed. Forward recursive model generation unit: Combining the global temperature penalty benchmark and the physical cross-constraint relationship between heterogeneous devices, the real-time health status of the equipment is obtained through the multi-source device coupled aging forward recursive model. Weighting calculation unit: Calculates the expected remaining physical life based on the real-time health status of the equipment, compares it with the polar summer flood logistics replenishment time window to construct a survival margin index, and dynamically generates real-time penalty weight coefficients; The optimization unit performs nonlinear smooth mapping on the real-time net power of the system, constructs adaptive inertial weights in conjunction with the survival margin index, and uses an improved particle swarm optimization algorithm to solve for the optimal power allocation ratio that minimizes the total energy loss of the system's throughput. The optimal power allocation ratio is then converted into an absolute active power scheduling command and sent to the underlying physical hardware of the microgrid for execution.

Images