A human body electric network management method and system

Abstract

This invention relates to a method and system for human body electrical grid management. The method includes: establishing a local thermal comfort model and a weight comfort model; based on the above models and specific scenarios, establishing a multi-objective optimization model and introducing a minimum protection logic for harsh scenarios and a battery derating logic for low temperatures; solving the multi-objective optimization model to determine the corresponding hardware configuration scheme; and employing a predictive energy management strategy for real-time energy management. This management strategy includes an offline strategy generation stage and an online real-time execution stage. Specifically, by solving a finite-time domain optimization problem offline, a control strategy mapping relationship based on the system state is pre-generated; and based on the collected real-time state data of the human body electrical grid management system and the control strategy mapping relationship, the optimal real-time power allocation command is obtained. This invention solves the computational bottleneck of complex control algorithms being unable to be solved in real time on embedded devices, and significantly improves the system's survivability and overall comfort in extremely variable outdoor scenarios.

Description

Technical Field

[0001] This invention relates to the field of human body grid management and energy management technology, and in particular to a human body grid management method and system. Background Technology

[0002] Personal thermal comfort systems are essential equipment designed to provide active heating for personnel working in cold environments. These systems vary in form depending on the complexity of their technical architecture and control logic: from intelligent heated clothing integrating heating elements and basic temperature control functions, to the Body Grid system, as described in this invention, which integrates distributed energy, multi-parameter sensing, and intelligent energy management algorithms. This type of equipment is widely used in outdoor sports, emergency rescue, military operations, and polar scientific expeditions.

[0003] Currently, most industrial applications of this type of system adopt a static "one-size-fits-all" design philosophy. Specifically, the core hardware configuration parameters of the system (such as the fixed capacity of the battery, the number and layout of heating elements, etc.) are predetermined at the initial product design stage, making dynamic adjustments impossible based on varying actual application scenarios and the user's individual physiological state. This rigid design pattern has significant flaws and is highly susceptible to two extreme consequences in specific scenarios: first, it can cause excessive redundancy in hardware performance ("performance overflow") in mild environments, leading to unnecessary increases in cost and weight; second, it can result in fatal "battery life shortages" in extreme or unexpected scenarios, posing a serious threat to the user's life safety.

[0004] Existing technologies mainly suffer from the following three key technological bottlenecks: (1) The comfort model lacks physiological basis and has no mechanism to protect against extreme environments.

[0005] Existing systems typically calculate heat demand based on simplified, steady-state heat conduction formulas. These models are crude and fail to adequately account for the significant differences in temperature sensitivity among different parts of the human body. More seriously, these models heavily rely on the wearer's own heat generation during periods of high metabolic activity (such as exercise). Once the user enters a resting state due to fatigue, injury, or task requirements, their metabolic heat generation (M) will decrease significantly. At this point, the battery capacity originally configured for active states will be insufficient to meet the actual heating needs, posing a significant risk of hypothermia. Furthermore, existing designs almost completely ignore the negative impact of the system's own weight (especially large-capacity batteries) or estimate fatigue under load using only a simple linear relationship. This fails to scientifically reflect the dynamic cumulative burden of weight on the wearer's mobility over time, resulting in poor comfort in practical applications.

[0006] (2) Static energy planning and heuristic control strategies lead to severe time-domain supply and demand mismatch.

[0007] In the system planning (design) phase, existing solutions typically use only a static constraint of "total energy supply exceeding total energy consumption" for battery capacity selection, which is a crude energy budgeting approach. In the operation and control phase, heuristic rules based on fixed thresholds (such as shutting off heating when the battery level drops below 20%) are commonly used. Both strategies fundamentally ignore the temporal mismatch between energy supply and demand. For example, in a photovoltaic charging system, energy harvesting is only effective during specific daytime periods and fluctuates drastically due to weather conditions; while human body heat demand peaks at night or on cloudy or rainy days. Static planning and heuristic control cannot anticipate this temporal difference. The consequence is that the system may overheat during the energy-rich daytime, wasting significant energy; while at night, when energy is scarce, it cannot provide the crucial heat supply due to depleted battery power, leading to system malfunction. Furthermore, existing technologies fail to fully consider the "derating effect" of extreme low temperatures on the actual usable battery capacity during the design phase, designing based on nominal capacity at room temperature, further exacerbating the range risk in winter environments.

[0008] (3) The fundamental contradiction between the computational complexity of advanced control algorithms and the limited computing power of embedded devices.

[0009] In theory, model-based predictive control strategies (such as rolling time control, RHC) are the optimal approach to solving the aforementioned time-domain mismatch problem. However, such algorithms (usually termed mixed-integer programming, MIP) have extremely high computational complexity, requiring large-scale, iterative online numerical solutions. This creates an irreconcilable contradiction with the stringent power consumption constraints and extremely limited computing power of microcontrollers (MCUs) faced by wearable embedded devices. Limited by the processing power of MCUs, existing systems are simply unable to support such complex algorithms to achieve high-frequency real-time decisions per minute or even per second, thus preventing the implementation of high-performance optimized control strategies in actual products and forcing a regression to low-performance, simple rule-based control.

[0010] Therefore, there is an urgent need in this field for a new technical solution that can overcome the above-mentioned defects in order to achieve precise, personalized and intelligent management of the human body's electrical grid.

[0011] It should be noted that the information disclosed in the background section above is only for understanding the background of this application, and therefore may include information that does not constitute prior art known to those skilled in the art. Summary of the Invention

[0012] The technical problem this application aims to solve is "how to provide a hardware configuration and energy management method for the human body electrical grid that combines physiological accuracy, robustness to extreme environments, and the ability to overcome the computing power limitations of embedded devices".

[0013] The technical solution adopted in this application to solve the above-mentioned technical problems is as follows.

[0014] This application provides a human body electrical grid management method, applied to a human body electrical grid management system, comprising the following steps: Modeling phase: Establish a local thermal comfort model that reflects the differences in thermal sensitivity of different parts of the human body and a weight comfort model that includes the cumulative effect over time; Configuration Phase: Based on specific task scenario information, a multi-objective optimization model is established; the local thermal comfort model and weight comfort model are incorporated into the optimization objective set of the multi-objective optimization model; the multi-objective optimization model introduces a minimum protection logic for harsh scenarios and a battery derating logic for low temperatures; by solving the multi-objective optimization model, the hardware configuration scheme of the human body power grid management system is determined. Operational phase: Based on the hardware configuration scheme, a predictive energy management strategy is adopted for real-time energy management; the predictive energy management strategy includes an offline strategy generation phase and an online real-time execution phase; In the offline strategy generation stage, the system state-based control strategy mapping relationship is pre-generated by solving the finite-time domain optimization problem offline. In the online real-time execution stage, the optimal real-time power allocation command is obtained based on the collected real-time state data of the human power grid management system and the control strategy mapping relationship.

[0015] In some embodiments, during the modeling phase, the local thermal comfort model is established based on human physiological data, which includes thermal sensitivity coefficients and comfort contribution weights of human body parts extracted from a standard dynamic thermophysiological model.

[0016] In some embodiments, the standard dynamic thermophysiological model includes the JOS-3 model, with the thermal sensitivity coefficient extracted from the thermal receptor distribution coefficient (SKINR) in the JOS-3 model, and the comfort contribution weight mapped from the perspiration / comfort distribution coefficient (SKINS) in the JOS-3 model.

[0017] In some embodiments, a local thermal comfort model is defined as follows: The calculation formulas are as follows:

[0018] in, It is a dimensionless comfort index; To extract the site derived from the distribution coefficient of heat receptors (SKINR) calibrated in human experiments in the JOS-3 dynamic thermophysiological model. Thermal sensitivity coefficient; Weights are assigned to comfort levels derived from the "Sweat / Comfort Distribution Coefficient (SKINS)" mapped from the JOS-3 model; For part The rated power of the heating element, The binary decision variable for whether or not to use the heating element. The total compensation power required to maintain thermal neutrality, For real-time ambient temperature, Choose a variable for clothing type; This is a unified set that includes the parameters mentioned above.

[0019] In some embodiments, during the modeling phase, the weight comfort model is established based on human physiological data, which includes physiological data and subjective index data collected from multiple real-world scenarios; Weight comfort model: The calculation formula is:

[0020] The weight comfort model includes a time variable. The dynamic exponential cumulative effect; among which, The fatigue attenuation coefficient under load ( To ensure its physiological accuracy, It is not a fixed empirical value; it is calculated based on normalized physiological and subjective index characteristics of real subjects collected in multiple scenarios, using a least squares method with non-negative constraints.

[0021] in, This represents the number of experimental samples. For the first The load (kg) of the test strip; For the first Duration of the test (h); For the first The observational normalized comfort labels for the test were obtained by fusing and truncating heart rate, skin temperature, and subjective fatigue scales; constraints This is to ensure that the weight comfort model conforms to the basic principle that "increased load / time should not increase comfort"; The total weight of the hardware representing the human body electrical grid management system is calculated using the following formula:

[0022] in, To fix the weight of the electronic components in the human body electrical grid management system, For the selected clothing The weight, and For batteries The weight and quantity of individual units. and For collector The individual weight and selection variables.

[0023] In some embodiments, the severe scenario fallback logic in the configuration phase includes: calculating the total compensation power required to maintain thermal neutrality. At that time, the resting metabolic rate was used as a premise, and the effects of the body's own high metabolic heat production and evaporative heat dissipation were ignored.

[0024] In some embodiments, the total compensation power The calculation formula is:

[0025] in, A constant temperature is set for the target skin to maintain thermal neutrality in the human body; For the selected clothing type The inherent thermal resistance; Thermal resistance of the external air layer; This represents the real-time ambient temperature.

[0026] In some embodiments, the battery low-temperature derating processing logic in the configuration phase includes: the actual usable capacity of the battery capacity used in the multi-objective optimization model after temperature derating based on the extreme ambient temperature of the target scenario.

[0027] In some embodiments, the decision variables of the multi-objective optimization model during the configuration phase are uniformly defined as the hardware decision variable set. Its mathematical expression and variable domain constraint formula are as follows:

[0028] in, For the first The quantity of each type of battery; The heating element and the data collector are defined as 0-1 Boolean variables respectively; The clothing type is limited to a preset clothing collection. .

[0029] In some embodiments, the multi-objective optimization model in the configuration phase must simultaneously satisfy the following constraints: system energy balance constraint, structural compatibility constraint, energy harvester quantity constraint, and power sufficiency constraint.

[0030] In some embodiments, the system energy balance constraint is:

[0031] The total capacity of all configured batteries This is the actual usable capacity after temperature derating based on the extreme ambient temperatures of the target scenario, plus the time step during the mission. Discrete total energy generated by the internal collector It must be greater than or equal to the discrete total energy consumption of the human power grid management system within the expected task time. The system energy balance constraint, as a static total constraint, is used to set the physical boundary for offline component selection during the configuration phase. The structural compatibility constraints are:

[0032] in For lightweight scenarios, identify variables. It is a collection of heavy hardware components; structural compatibility constraints are used to adapt to lightweight task scenarios; The number of energy harvesters is constrained as follows:

[0033] in, The maximum number of energy harvesters allowed to be installed, set according to specific task attributes or hardware physical limitations; used to limit the maximum number of harvesters that can be installed in the human body power grid management system; The power adequacy constraint is:

[0034] in, For part The maximum rated power of the heating element; the power adequacy constraint requires that the total maximum rated power of the selected heating element must meet the heat compensation requirements under extreme temperatures to ensure sufficient heating in extreme environments.

[0035] In some embodiments, the optimization objective set of the multi-objective optimization model includes: minimizing the total initial cost of the system. Maximize the overall comfort score Minimize the total weight of the system Overall comfort score It is a weighted sum of the local thermal comfort model and the weight comfort model; the multi-objective optimization model determines the hardware configuration scheme by solving the Pareto front of the optimization objective set.

[0036] In some embodiments, the optimization objective of the multi-objective optimization model is:

[0037] in:

[0038] in, The cost of fixed electronic components representing the human body electrical grid management system; This represents the cost of the i-th heating element; This represents the cost of the k-th battery; This represents the cost of the m-th energy harvester; The static balance weights for thermal comfort and weight comfort are set by the human body electrical grid management system based on scenario preferences and are read from the weight table during initialization.

[0039] In some embodiments, the predictive energy management strategy employs rolling time-domain control (RHC), which, during the offline strategy generation phase, targets the remaining task time domain. Solve for the combined objective function that maximizes thermal comfort and minimizes battery aging penalty:

[0040] in As a static balancing weight, it combines the functions of a unit conversion scalar and a preference adjustment factor, and is used to bridge the engineering trade-off between the dimensionless physiological comfort index and irreversible battery asset degradation in the comprehensive objective function. The formula for calculating the segmented linear cumulative degradation cost of the battery based on the weighted energy throughput method is as follows:

[0041] in, The total capital cost of the battery modules determined for the configuration phase; The total energy throughput over the entire battery life cycle is given by the following formula:

[0042] in For the expected rated cycle life, the number 2 represents bidirectional charge and discharge, denoted as 2; It is a piecewise linear cumulative function used to approximate the accelerated loss weights at the extreme state of charge (SOC) boundary; The offline solution process in the offline policy generation stage is constrained by the following SOC evolution constraints. By simulating future time series containing photovoltaic transient fluctuations, the temporal mismatch of energy during day-night or alternating cloudy / sunny periods is achieved:

[0043] in, The activation vector for the heating element in the future prediction time domain; and These are the power consumption of the heating array and the base power consumption, respectively. This represents the predicted transient power fluctuation of photovoltaic power generation; where, This is to prevent the risk of the human electrical grid management system failing due to overheating and battery depletion in the early stages of the mission.

[0044] In some embodiments, during the online real-time execution phase, explicit model predictive control (MPC) is used to ensure high-frequency real-time performance on wearable embedded microcontrollers (MCUs) with limited computing power. During the offline policy generation phase, the state space of the human body grid management system is discretized into a three-dimensional grid of remaining time, ambient temperature, and current state of charge (SOC). For each grid node, a rolling time-domain control (RHC) optimization problem based on a comprehensive objective function and SOC evolution constraints is solved offline. The activation vector of the optimal heating element at the current moment is stored as a pre-computed optimal control policy lookup table (LUT). During the online real-time execution phase, the MCU collects real-time sensor data only at set control steps and projects it onto neighboring grid nodes, achieving a computational complexity of O(n log n). The lookup operation outputs the current optimal heating element activation command.

[0045] In some embodiments, a human body electrical grid management system is also provided, comprising: The modeling module is configured to: establish a local thermal comfort model that reflects the differences in thermal sensitivity of different parts of the human body and a weight comfort model that includes the cumulative effect over time; The configuration module is configured to: establish a multi-objective optimization model based on specific task scenario information; incorporate the local thermal comfort model and weight comfort model into the optimization objective set of the multi-objective optimization model; introduce harsh scenario protection logic and battery low temperature derating processing logic into the multi-objective optimization model; and determine the hardware configuration scheme of the system by solving the multi-objective optimization model. The operation control module is configured to perform real-time energy management using a predictive energy management strategy. It is implemented using a decoupled architecture, including an offline strategy generation unit and an online real-time execution unit. The offline policy generation unit is configured to: pre-generate control policy mapping relationships based on system state by solving finite-time domain optimization problems offline; The online real-time execution unit is configured to obtain the optimal real-time power allocation command based on the collected real-time status data of the human power grid management system and the mapping relationship of the control strategy. One or more processors are used to execute the operations of the modeling module, configuration module, and operation control module; Memory is used to store control strategy mapping relationships, system hardware configuration schemes, and instructions executed by the processor.

[0046] In some embodiments, a computer-readable storage medium is also provided, on which a computer program is stored, which, when executed by a processor, implements the steps of the method of this application.

[0047] The present invention has the following beneficial effects: This invention provides a hardware configuration and energy management method for the human body power grid management system that combines physiological accuracy, robustness to extreme environments, and the ability to overcome the computing power limitations of embedded devices by performing a series of steps in the modeling, configuration, and operation phases. Specifically, this invention establishes a local thermal comfort model reflecting the differences in thermal sensitivity of different parts of the human body, providing a quantifiable comfort evaluation benchmark that aligns with real physiological perception and laying the physiological foundation for precise management. By establishing a weight comfort model incorporating the cumulative effect of time, the dynamic fatigue impact of system weight on the wearer is quantified, enabling the optimization process to simultaneously consider thermal protection and mobility. Based on specific task scenario information, a multi-objective optimization model is established, incorporating the local thermal comfort model and weight comfort model into the optimization objective set, achieving a shift from a "one-size-fits-all" approach to "on-demand customization" in hardware configuration, allowing the configuration to adapt to the needs and preferences of different scenarios. By introducing a severe scenario protection logic into the multi-objective optimization model, the thermal requirements of the system under the most unfavorable operating conditions are ensured, significantly improving the system's safety floor and survivability. By introducing battery low-temperature derating logic, the energy budget is based on the battery's actual usable capacity in extreme environments, avoiding the range risk caused by falsely advertised capacity and improving the accuracy of planning. Through the adoption of… This invention employs a predictive energy management strategy, decoupling it into an offline strategy generation stage and an online real-time execution stage. In the offline strategy generation stage, a control strategy mapping relationship based on system state is pre-generated by solving a finite-time domain optimization problem offline, thus pre-absorbing the enormous computational burden. Finally, in the online real-time execution stage, the optimal real-time power allocation command is obtained based on the collected real-time state data of the human body power grid management system and the aforementioned control strategy mapping relationship. This allows complex optimal control strategies to run in real-time on wearable embedded devices with limited computing power, achieving extremely low latency and extremely high frequency. This fundamentally resolves the contradiction between advanced algorithms and embedded computing power. In summary, this invention, by establishing a comfort model based on physiological data, employing multi-objective optimization in hardware configuration incorporating extreme minimum protection logic and battery derating processing, and combining an offline pre-computation and online lookup predictive control strategy, enables the human body power grid to achieve real-time optimal control on embedded devices with limited computing power while ensuring physiological comfort and safety in extreme environments. This significantly improves the reliability, adaptability, and practicality of the system in practical applications.

[0048] It should be noted that the solution to the existing technical problems in this application relies on the close synergy and cooperation of the aforementioned technical features: the local thermal comfort model and weight comfort model established in the modeling stage provide accurate and multi-dimensional optimization objectives for the multi-objective optimization model in the configuration stage, ensuring the physiological accuracy of the configuration results; while the harsh scenario protection logic and battery low-temperature derating processing logic in the configuration stage directly utilize the output of the modeling stage to set conservative and reliable boundary conditions for the optimization problem, jointly contributing to the system's robustness in extreme environments; finally, the hardware configuration scheme obtained from the configuration stage is passed as a key input parameter to the predictive energy management strategy in the operation stage, and the decoupled architecture, offline pre-generated control strategy mapping relationship, and online mechanism for obtaining the optimal real-time power allocation instruction based on the control strategy mapping relationship adopted by this strategy are direct technical means to overcome the computing power limitations of embedded devices. These three stages are interconnected, and the output of the previous stage is the input or foundation of the next stage. Without any feature, it will be impossible to simultaneously achieve the three core objectives of "physiological accuracy", "robustness in extreme environments", and "overcoming computing power limitations".

[0049] Preferably, this invention further defines the local thermal comfort model as being based on human physiological data, including thermal sensitivity coefficients and comfort contribution weights of human body parts extracted from standard dynamic thermophysiological models. This ensures that the model parameters originate from internationally recognized physiological models calibrated through rigorous human experiments, rather than empirical formulas, greatly enhancing the authority, scientific rigor, and reliability of the local thermal comfort model. By further clarifying that the standard dynamic thermophysiological model includes the JOS-3 model and associating the thermal sensitivity coefficients and comfort contribution weights with specific coefficients (SKINR and SKINS) in the JOS-3 model, this invention provides a concrete and feasible way to obtain model parameters, enabling the model to achieve the fidelity of a high-order dynamic thermophysiological model.

[0050] Preferably, this invention clarifies the baseline logic for severe scenarios by employing the resting metabolic rate and ignoring the body's own high metabolic heat production and evaporative heat loss when calculating the total compensation power required to maintain thermal neutrality. The specific technical meaning of "baseline" is defined, meaning that the system performs power budgeting based on the premise that the user is in their most vulnerable resting state, thereby providing the system with maximum safety redundancy.

[0051] Preferably, the present invention ensures that the battery capacity data, the cornerstone of the system energy budget, is true and reliable by clearly defining the battery low-temperature derating processing logic, including that the battery capacity used in the multi-objective optimization model is the actual usable capacity after temperature derating. This reflects the performance degradation of the battery at extreme low temperatures and fundamentally avoids planning errors caused by data distortion.

[0052] Other beneficial effects of the present invention will be further described below. Attached Figure Description

[0053] The above and / or additional aspects and advantages of this application will become apparent and readily understood from the following description of the embodiments taken in conjunction with the accompanying drawings, wherein: Figure 1 The end-to-end energy / information flow of the Human Grid Management System (BGEMS) in its three phases (modeling, configuration, and operation); Figure 2 This is a diagram illustrating the overall architecture of the Human Grid Management System (BGEMS). Figure 3 Four prototype examples of heated clothing (one-piece / two-piece, down / fleece) provided for this invention. Figure 4 The status of the Harbin commercial ski resort system changes over time; Figure 5 The system state changes over time in the extreme mountaineering scenario of Muztagh Ata. Figure 6 The state of the human-built collaborative microgrid system at the Antarctic research station changes over time. Detailed Implementation

[0054] The embodiments of the present invention will be described in detail below. It should be emphasized that the following description is merely exemplary and not intended to limit the scope and application of the present invention.

[0055] Furthermore, the terms "first" and "second" are used for descriptive purposes only and should not be construed as indicating or implying relative importance or implicitly specifying the number of technical features indicated. Thus, a feature defined as "first" or "second" may explicitly or implicitly include one or more of that feature. In the description of embodiments of the present invention, "a plurality of" means two or more, unless otherwise explicitly specified.

[0056] This invention discloses a method for managing a human body electrical grid. In some embodiments, such as... Figure 1 , 2 As shown, the human body electrical grid management method mainly includes the following three stages: (1) Modeling stage: Establishing an accurate dimensionless local thermal comfort model. This invention abandons the classic PMV-PPD model, which is only applicable to steady-state homogeneous environments, and directly extracts parameters from the internationally recognized JOS-3 high-order dynamic thermophysiological model that have been rigorously calibrated in human experiments: the "thermal receptor distribution coefficient (SKINR)" is used as the thermal sensitivity coefficient. Mapping the "Sweat / Comfort Distribution Coefficient (SKINS)" to a weighted average of comfort contribution. This inherently gives the model the fidelity of a high-order physiological model. Furthermore, to construct a scientific weight comfort model, this invention quantifies and normalizes the aggregated features of multi-source subjective scales and physiological signals. It employs least squares fitting with non-negativity constraints for multi-scenario calibration. Data includes: trial duration, load, exercise intensity, and subjective metrics (Borg 6–20 fatigue scale, ASHRAE thermal sensation, 0–10 pain score) and physiological metrics (mean heart rate, mean skin temperature, heart rate recovery amplitude). Data preprocessing follows strict gating and cleaning (e.g., numericalization, mandatory uniqueness, boundary truncation). (etc.), thereby providing a highly robust physiological benchmark for configuration and operational optimization.

[0057] (2) Configuration Phase: Establish the MOMINLP optimization model. In the calculation of compensation power requirements, the "Worst-case Conservative Safety Margin" logic is adopted. It is specifically based on the resting metabolic rate (PAR=1.25) and ignores the metabolic heat production (M) and evaporation (E) under high human exercise to ensure safety redundancy under extreme low temperature and extreme fatigue conditions. At the same time, it combines the actual degradation (derating loss) of battery capacity due to extreme low temperature as a correction input. Under this boundary, the Pareto front of "total cost, overall comfort, and system weight" is solved to determine the optimal hardware combination.

[0058] (3) Operation Phase: Abandoning traditional heuristic control, this invention introduces Rolling Time Control (RHC) to dynamically simulate future time series, utilizing real-time optimization with clear safety red lines to thoroughly solve the "time-domain mismatch" problem of day-night or intermittent energy supply and demand. To bridge the computing power gap in wearable embedded devices, this invention adopts an explicit model predictive control (Explicit MPC) architecture: The MIP problem under all possible states (time, temperature, SOC) is pre-solved offline on the PC, generating a lightweight three-dimensional lookup table (LUT); the device-side MCU only needs to project onto the nearest grid node during operation and... The complexity is achieved by performing a "fast lookup table" and immediately outputting the optimal heating array instruction, which is equivalent to a real-time solution.

[0059] It should be noted that the lookup table mechanism involved in the above-mentioned operation phase is essentially a control strategy mapping relationship. In addition to the lookup table mechanism, this invention can also be adapted to the following variant strategies: parameterized strategy functions (such as neural networks), a set of "if-then-else" rules, decision trees, etc., without specific limitations here.

[0060] In some embodiments, the local thermal comfort model Weight comfort model The calculation formulas are as follows:

[0061]

[0062] in, It is a rigorous dimensionless comfort index; The location is derived directly from the "skin receptor distribution coefficient (SKINR)" in the internationally authoritative dynamic thermophysiological model JOS-3, which has been rigorously calibrated in human experiments. Thermal sensitivity coefficient; Weights are assigned to comfort levels derived from the "Sweat / Comfort Distribution Coefficient (SKINS)" mapped from the JOS-3 model; For part The rated power of the heating element, The binary decision variable for whether or not to use the heating element. The total compensation power required to maintain thermal neutrality, For real-time ambient temperature, Choose a variable for clothing type; This is a unified set that includes the aforementioned hardware decision variables.

[0063] Weight comfort model Includes time variables The dynamic exponential cumulative effect; The fatigue attenuation coefficient under load ( To ensure its physiological accuracy, Instead of fixed empirical values, these values ​​are derived from normalized physiological and subjective indicators collected from real subjects in multiple scenarios (such as indoor treadmills, outdoor athletic tracks, and on-site weighted hiking), and are calculated using a least squares fitting method with non-negative constraints.

[0064] in, This represents the number of experimental samples. For the first The load (kg) of the test strip; For the first Duration of the test (h); For the first Normalized comfort labels from observations in the test (obtained by fusion and truncation of heart rate, skin temperature, and subjective fatigue scales); constraints This ensures the model adheres to the fundamental principle that "increased load / time should not increase comfort." It should be noted that... It is an estimate. These are the baseline parameters, where, yes The consistency estimator. When the sample size N approaches infinity and the data covers a sufficiently diverse range of scenarios, the consistency estimator obtained using the above formula is... It will converge to the true value according to probability. In actual system operation, because we can never obtain absolute perfection... Therefore, data-driven solutions must be used. , to replace , thus calculating The value of .

[0065] The total weight of the system's hardware is represented by the following formula:

[0066] in, To fix the weight of the electronic components in the system, For the selected clothing The weight, and For batteries The weight and quantity of individual units. and For collector The individual weight and selection variables.

[0067] In some embodiments, the total compensation power required to maintain thermal neutrality The calculation formula is:

[0068] in, To maintain the target skin temperature for thermal neutrality, a constant temperature between 34°C and 35°C is set. This constant implicitly assumes a low metabolic rate (i.e., resting activity rate PAR = 1.25), a relative humidity of 50%, a wind speed of 0.10 m / s, and no thermal resistance of clothing on the entire body surface (0.0 clo). Due to the inherent differences in the physiological heat transfer characteristics of different parts of the human body, and referring to the JOS-3 PMV calibration, each part of the body is assigned an independent fixed constant (e.g., head 35.13°C, chest 34.76°C, back 34.67°C, hands 34.44°C, etc.). For the selected clothing type The inherent thermal resistance was obtained by consulting standard system databases such as ISO 9920; The external air layer thermal resistance is calculated based on the ambient wind speed in a specific scenario. The formula deliberately ignores the body's own high metabolic heat production (M) and evaporative heat loss (E) in the complete human body thermal balance equation, employing a worst-case conservative safety margin engineering logic under extreme conditions. Its purpose is to ensure that the system's calculated heat deficit can meet the user's heating needs in their most vulnerable states (such as extreme fatigue, cessation of activity at rest, or a sudden drop in M ​​due to hypothermia or injury), thereby preventing the fatal risk of insufficient battery capacity due to overestimating the body's own metabolic heat generation.

[0069] In some embodiments, the decision variables of a multi-objective optimization model are uniformly defined as a set of hardware decision variables. Its mathematical expression and variable domain constraint formula are as follows:

[0070] in, For the first The quantity of each type of battery; The heating element and the data collector are defined as 0-1 Boolean variables respectively; The clothing type is limited to a preset clothing collection. .

[0071] In some embodiments, the multi-objective optimization model needs to satisfy time-series-based system energy balance constraints, structural compatibility constraints, energy harvester quantity constraints, and power sufficiency constraints:

[0072] The total capacity of all configured batteries This is the actual usable capacity after derating (effective capacity) based on the extreme ambient temperature of the target scenario (e.g., -30°C). This capacity is increased by the time steps during the mission. Discrete total energy generated by the internal collector It must be greater than or equal to the discrete total energy consumption of the system during the expected task time. Formula (7) is a static total constraint and is only used to set the physical boundary for offline component screening during the configuration phase. As for the time-domain supply and demand mismatch caused by photovoltaic power generation only existing during the day and heat demand being higher at night in the natural environment, it is specifically handled by the rolling time-domain control (RHC) during the operation phase (step S3).

[0073] For lightweight task scenarios, mandatory structural compatibility constraints are implemented, prohibiting the mounting of heavy batteries and rigid data acquisition devices on lightweight clothing:

[0074] in For lightweight scenarios, identify variables. It is a collection of heavy-duty hardware components.

[0075] The energy harvester quantity constraint limits the maximum number of harvesters that the system can accommodate:

[0076] in, The maximum number of energy harvesters allowed to be carried, set according to specific mission attributes or hardware physical limitations.

[0077] The power adequacy constraint requires that the total maximum rated power of the selected heating elements must meet the heat compensation requirements under extreme temperatures to ensure sufficient heating in extreme environments.

[0078] in, For part The maximum rated power of the heating element.

[0079] In some embodiments, the multi-objective optimization model uses the total initial cost Overall comfort score and total weight To optimize the objective:

[0080] The specific calculation formulas for each sub-objective function are as follows:

[0081] in, Represents the cost of fixed electronic components in the system. The series of parameters represents the unit cost of various variable hardware components, for example, This represents the cost of the i-th heating element; This represents the cost of the k-th battery; This represents the cost of the m-th energy harvester; The static balancing weights are set by the system based on scene preferences. During system initialization, these weights are read from a weight table (<0.7, 0.3> for skiing scenarios and <0.4, 0.6> for long-duration mountaineering scenarios). As the system runs, when the cumulative number of samples is ≥ N_min (configurable, default 1000) and the scene coverage is ≥ 90%, it automatically switches to a data-driven calibration mechanism: calling the Bayesian Calibration script and outputting... The posterior distribution is used, with the upper limit of the 95% confidence interval as the boundary.

[0082] In some embodiments, predictive energy management strategies abandon heuristic static rules and employ rolling time-domain control (RHC), which, during the offline policy generation phase, targets the remaining task time domain. Solve for the combined objective function that maximizes thermal comfort and minimizes battery aging penalty:

[0083] As a static balancing weight, it combines the functions of a unit conversion scalar and a preference adjustment factor, and is used to bridge the engineering trade-off between the dimensionless physiological comfort index and irreversible battery asset degradation in the objective function. The formula for calculating the segmented linear cumulative degradation cost of the battery based on the weighted energy throughput method is as follows:

[0084] in, The total capital cost of the battery modules determined for the configuration phase; The total energy throughput over the battery's entire lifespan is calculated as follows: ( The expected rated cycle life is indicated by the number 2, which represents bidirectional charge and discharge (denoted as 2). It is a piecewise linear cumulative function used to approximate the acceleration loss weights at the extreme SOC boundary.

[0085] The offline solution process is constrained by the following battery state of charge (SOC) evolution constraints. By simulating future time series that include photovoltaic transient fluctuations, this control strategy fundamentally solves the time-domain mismatch problem of energy during day-night or alternating cloudy / sunny periods:

[0086] in, The activation vector for the heating element in the future prediction time domain; and These are the power consumption of the heating array and the base power consumption, respectively. This represents the predicted transient power fluctuation of photovoltaic power generation.

[0087] At the same time, a safety red line constraint is set as a hard safety net, forcing the optimizer to actively store electrical energy during daylight hours, completely eliminating the risk of system failure due to battery depletion caused by overheating in the early stages of a task.

[0088] In some embodiments, pre-generating the control strategy mapping relationship based on system state specifically involves pre-generating an optimal control strategy lookup table. Based on the collected real-time state data of the human body power grid management system and the control strategy mapping relationship, obtaining the optimal real-time power allocation command specifically involves querying the optimal control strategy lookup table to obtain the optimal real-time power allocation command. The online real-time execution part is based on the principle of explicit model predictive control (MPC). To ensure high-frequency real-time performance on wearable embedded microcontrollers (MCUs) with limited computing power, in the offline phase, the system state space is discretized into a three-dimensional grid of remaining time, ambient temperature, and current SOC. The aforementioned rolling time domain control (RHC) is calculated offline for each grid node, and the activation vector of the optimal heating element at the current moment is stored in a pre-computed optimal control strategy lookup table (LUT). In the online operation phase of the wearable embedded device, the MCU only needs to collect real-time sensor data and project it onto neighboring grid nodes every set control step (e.g., every minute or every second), achieving a computational complexity of only [missing information]. The lookup table operation can quickly output the optimal heating element activation command. This mechanism successfully overcomes the bottleneck of embedded computing power, achieving optimal control equivalent to full-order MIP solving while ensuring extremely high control frequency.

[0089] In some embodiments, a human body electrical grid management system is also provided, comprising: The modeling module is configured to: establish a local thermal comfort model that reflects the differences in thermal sensitivity of different parts of the human body and a weight comfort model that includes the cumulative effect over time; The configuration module is configured to: establish a multi-objective optimization model based on specific task scenario information; incorporate the local thermal comfort model and weight comfort model into the optimization objective set of the multi-objective optimization model; introduce harsh scenario protection logic and battery low temperature derating processing logic into the multi-objective optimization model; and determine the hardware configuration scheme of the system by solving the multi-objective optimization model. The operation control module is configured to perform real-time energy management using a predictive energy management strategy. It is implemented using a decoupled architecture, including an offline strategy generation unit and an online real-time execution unit. The offline policy generation unit is configured to: pre-generate control policy mapping relationships based on system state by solving finite-time domain optimization problems offline; The online real-time execution unit is configured to obtain real-time power allocation instructions by collecting real-time status data of the system and querying the optimal control strategy lookup table. One or more processors are used to execute the operations of the modeling module, configuration module, and operation control module; Memory is used to store control strategy mapping relationships, system hardware configuration schemes, and instructions executed by the processor.

[0090] In some embodiments, a computer-readable storage medium is also provided, on which a computer program is stored, which, when executed by a processor, implements the steps of the method of this application.

[0091] To make the objectives, technical solutions, and advantages of this invention clearer, the technical solutions of this invention will be described in detail below with reference to specific embodiments. This invention generates corresponding hardware architectures and management strategies for three representative extreme environments and differences in requirements.

[0092] Example 1: Harbin commercial skiing scenario (emphasizing economy and high mobility).

[0093] Application Background Description: Assume the wearer plans a one-day commercial skiing trip in Harbin. The task is characterized by: short duration (approximately 4-6 hours), moderate to low ambient temperature (-15℃ to -20℃), high user sensitivity to rental costs, and frequent skiing maneuvers requiring exceptional mobility and lightweight design.

[0094] Step 1 (Configuration): During the configuration phase, the system prioritizes minimizing cost and weight. The final hardware configuration quantization results generated from solving the Pareto front are shown in the table below:

[0095] The formula for calculating the system's economic return (expected payback period) is as follows:

[0096] The above The total initial hardware cost of the system determined during the configuration phase. Net income from a single lease This represents the average number of effective rental days per year. Assuming... The estimate is 40 times per year (based on the fact that the Harbin ski season lasts approximately 90 days, from December 1st to March 1st of the following year, and each use of the equipment requires maintenance, the annual maximum number of uses for a single piece of equipment is 90 times / 2 = 45 times. Considering that equipment damage would require even longer repair times, the assumed value of 40 times per year is more reasonable). The value is 100 yuan. Therefore, in this scenario, the estimated payback period is 1450 / (40...). 100) = 0.36 (years).

[0097] Step 2 (Run): As shown Figure 4 As shown, during a skiing trip expected to last 4 hours, the system predicts the control state using the RHC model and operates with a "comfort-first" strategy. This is because no photovoltaic modules are mounted. As the ambient temperature drops from -15℃ to -18.5℃, the power of the distributed heating element gradually increases, while the battery... It decreases to a threshold over time ( (Safety red line), the system ensures that it can maintain a high level of comfort (0.93 to 0.96) even if the system does not shut down before the end of skiing.

[0098] Example 2: Extreme mountaineering scenario on Muztagh Ata (emphasizing extreme lightweighting and life support).

[0099] Application Background Description: Assume the wearer plans to climb Muztagh Ata, an altitude of over 7000 meters. The mission is characterized by: extremely harsh environment (temperatures below -30°C, accompanied by strong winds), long mission duration (summit attempts typically take over ten hours), and high-altitude hypoxia causing significant physiological strain for every additional 1000 grams of load (weight fatigue attenuation coefficient). (Extremely high).

[0100] Step 1 (Configuration): The system establishes a multi-objective optimization model and selects a scheme that balances "limit weight control" and "endurance constraints" in the Pareto front. The final hardware configuration results are shown in the table below:

[0101] Where, assuming The estimate is 3 times per year (based on the Muztagh Ata climbing season having three climbing windows from June 1st to September 1st each year, such as June 21st-July 9th, July 10th-July 28th, and July 29th-August 16th, with a minimum usage period of 18 days per window. Considering the maintenance required after equipment use and the potential repair cycle, the assumption of 3 times per year is more reasonable). The cost is 2000 yuan. Therefore, in this scenario, the estimated payback period is 14800 / (3). 2000) = 2.47 (years).

[0102] Step 2 (Run): As shown Figure 5 As shown, during a 14-hour climb, the alternation of day and night severely limited battery power in the later stages (10 hours). Relying solely on total power constraints or simple threshold controls (such as shutting off non-core areas when battery levels drop below 20%) would fail because it ignores temporal mismatch. Blindly activating the heating elements during the day when sunlight is abundant would lead to instantaneous battery depletion at night when there is no light and temperatures plummet, posing a risk of frostbite to the wearer's extremities. Thanks to a predictive RHC model and real-time LUT lookup, the system dynamically simulated and "anticipated" nighttime light loss and low temperatures early in the mission. Its core lies in proactively ensuring the entire prediction time domain... The dynamic evolution of the internal power level always exceeds the red line for life safety:

[0103] Through offline solution and online mapping of the aforementioned forward-looking time-domain constraints, the system proactively reduces power consumption in non-lethal areas by 15% during the day to smoothly accumulate power. In actual testing, the system of this invention can still maintain basic safe heating after 14 hours, while traditional systems without predictive control face power depletion and shutdown after 11 hours.

[0104] Example 3: Human-building collaboration scenario at Antarctic research station (focusing on long-cycle endurance and microgrid collaboration).

[0105] Application Background Description: Assume researchers are undertaking long-term deployments at and around an Antarctic research station. The mission is characterized by extreme cold (-40°C and below), heavy reliance on diesel fuel or microgrids for power, and extremely high logistical costs. This scenario demands not only individual thermal comfort but also significant reductions in energy consumption across the entire research station's building complex.

[0106] Step 1 (Modeling and Configuration): The system establishes a collaborative model between the human body's electrical grid and the building's HVAC system, mapping individual needs to the energy requirements of the entire site. The system equips the research team with full sets of thermal clothing and high-capacity backpack power supplies, serving as "distributed micro-terminals" for building heating. The final quantitative results of the hardware configuration are shown in the table below:

[0107] Collaborative Optimization Overall Objective and Energy Consumption Model: To achieve collaborative energy conservation, the system minimizes total energy consumption by balancing the supply and demand distribution between "kilowatt-level (kW) building heating" and "watt-level (W) individual heating." Its collaborative optimization overall objective function is defined as:

[0108] in, The total aggregate energy consumption of the system. For the evaluation period (e.g., 24 hours). For heating power of HVAC systems, The total number of people inside the building. For the first Individual user's human body electrical grid heating power, The core decision variable for optimization is the indoor global environmental setpoint temperature. The building HVAC power model depends on heat loss from the building shell and deducts heat dissipation from occupants.

[0109] in, The heat transfer coefficient of the building envelope. The outdoor extreme air temperature, The total free heat generated by the metabolism of indoor occupants can offset the heat load. This refers to the coefficient of performance (COP) of an air conditioning system. Physically, it represents the coefficient of performance when the indoor temperature is set to a certain level. When the temperature is reduced, the temperature difference between the inside and outside decreases, and heat loss is significantly reduced.

[0110] Thermal Neutrality Constraint: When a building actively lowers its indoor set temperature... At this time, the environment will deviate from thermal neutrality, and BGEMS must intervene to fill the heat gap. Its hard compensation constraint is:

[0111] The physical significance lies in: the target temperature of human skin. and the ambient temperature after the downgrade The heat loss that exists must be addressed by precise localized heating power. To make amends.

[0112] Step 2 (Run): As shown Figure 6 As shown, the research station building actively adjusts the global set temperature of the central high-power HVAC system. The temperature dropped from the baseline of 20.0℃ to 15.0℃, which makes the kilowatt-level... Significant reduction. As the game logic behind synergistic energy saving, BGEMS increases watt-level output to compensate for indoor cooling. This huge efficiency arbitrage reduces HVAC power from 25.33 kW to 21.79 kW (saving approximately 3.54 kW), while 30 team members only need to compensate about 20.0 W each (a total increase in micro-energy consumption of 0.60 kW).

[0113] Step 3 (Performance Verification): (e.g.) Figure 6 As shown, in the traditional mode without this system, a high-power HVAC system must be used to maintain a temperature of 20°C to ensure that researchers do not suffer from hypothermia while at rest, consuming as much as 607.84 kWh per day. However, through the coordinated operation of this formula, the research station's expected total daily energy consumption is reduced to 537.44 kWh, achieving a net energy saving of 11.58% (70.40 kWh). This significantly improves the overall endurance of the research station's microgrid, perfectly realizing a low-carbon and energy-saving response from individual needs to the microgrid level. This has significant resource security value in extreme environments that heavily rely on expensive diesel logistics.

[0114] In summary, the advantages of this invention over the prior art are as follows: (1) Rigorous physiological basis and extreme survival guarantee: The model not only integrates the calibration coefficient of JOS-3 with the time cumulative fatigue of the actual fitting, but also deliberately leaves enough conservative safety redundancy for the life support system in the "worst state (such as hypothermia and rest)" to avoid configuration failure caused by overly optimistic overestimation of heat generation.

[0115] (2) Possesses future foresight and completely solves the time-domain supply-demand mismatch: It abandons the crude approach of treating energy as a static pool, and relies on predictive models and strict SOC protection limits to proactively "prepare for rain" during the day to reserve power for severe weather at night, effectively avoiding excessive power loss in the early stages of a task. Furthermore, it incorporates battery degradation penalties based on full-lifetime throughput to protect hardware assets.

[0116] (3) Perfectly solves the bottleneck of embedded real-time control: It is the first to reduce the dimensionality of the highly complex time-domain optimization problem and decouple it into "PC offline solution + MCU online lookup table (LUT)". This solution cleverly resolves the time-consuming defect of mixed integer programming algorithm, and can still stably output the global optimal control even in micro wearable chips with limited computing power.

[0117] To clarify the meaning of the variables in the relevant formulas of this invention, the following parameter comparison table is provided:

[0118]

[0119] Those skilled in the art will understand that embodiments of the present invention can be provided as methods, human body electrical grid management systems, or computer program products. Therefore, the present invention can take the form of a completely hardware embodiment, a completely software embodiment, or an embodiment combining software and hardware aspects. Furthermore, the present invention can take the form of a computer program product embodied on one or more computer-usable storage media (including, but not limited to, disk storage, CD-ROM, optical storage, etc.) containing computer-usable program code.

[0120] This invention is described with reference to flowchart illustrations and / or block diagrams of methods, apparatus (human energy grid management system), and computer program products according to embodiments of the invention. It will be understood that each block of the flowchart illustrations and / or block diagrams, and combinations of blocks in the flowchart illustrations and / or block diagrams, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, special-purpose computer, embedded processor, or other programmable data processing device to produce a machine, such that the instructions, which execute via the processor of the computer or other programmable data processing device, generate instructions for implementing the flowchart illustrations and / or block diagrams. Figure 1 One or more processes and / or boxes Figure 1 A device that provides the functions specified in one or more boxes.

[0121] These computer program instructions may also be stored in a computer-readable storage medium that can direct a computer or other programmable data processing device to function in a particular manner, such that the instructions stored in the computer-readable storage medium produce an article of manufacture including instruction means, which are implemented in a process Figure 1 One or more processes and / or boxes Figure 1 The function specified in one or more boxes.

[0122] These computer program instructions may also be loaded onto a computer or other programmable data processing equipment to cause a series of operational steps to be performed on the computer or other programmable equipment to produce a computer-implemented process, thereby providing instructions that execute on the computer or other programmable equipment for implementing the process. Figure 1 One or more processes and / or boxes Figure 1 The steps of the function specified in one or more boxes.

[0123] The background section of this invention may include background information about the problems or environment in which the invention is being developed, and is not necessarily a description of prior art. Therefore, the content included in the background section does not constitute an admission of prior art by the applicant.

[0124] The above description provides a further detailed explanation of the present invention in conjunction with specific / preferred embodiments, and it should not be construed that the specific implementation of the present invention is limited to these descriptions. For those skilled in the art, various substitutions or modifications can be made to these described embodiments without departing from the concept of the present invention, and all such substitutions or modifications should be considered within the scope of protection of the present invention. In the description of this specification, the reference to terms such as "an embodiment," "some embodiments," "preferred embodiment," "example," "specific example," or "some examples," etc., indicates that the specific features, structures, materials, or characteristics described in connection with that embodiment or example are included in at least one embodiment or example of the present invention. In this specification, the illustrative expressions of the above terms do not necessarily refer to the same embodiment or example. Moreover, the specific features, structures, materials, or characteristics described can be combined in any suitable manner in one or more embodiments or examples. Without contradiction, those skilled in the art can combine and integrate the different embodiments or examples described in this specification and the features of different embodiments or examples. Although the embodiments of the present invention and their advantages have been described in detail, it should be understood that various changes, substitutions, and modifications can be made herein without departing from the scope of protection of the patent application.

Claims

1. A human body electrical grid management method, applied to a human body electrical grid management system, characterized in that, Includes the following steps: Modeling phase: Establish a local thermal comfort model that reflects the differences in thermal sensitivity of different parts of the human body and a weight comfort model that includes the cumulative effect over time; Configuration Phase: Based on specific task scenario information, a multi-objective optimization model is established; the local thermal comfort model and the weight comfort model are incorporated into the optimization objective set of the multi-objective optimization model; the multi-objective optimization model introduces a minimum protection logic for harsh scenarios and a battery derating logic for low temperatures; by solving the multi-objective optimization model, the hardware configuration scheme of the human body power grid management system is determined; Operation phase: Based on the hardware configuration scheme, a predictive energy management strategy is adopted for real-time energy management; the predictive energy management strategy includes an offline strategy generation phase and an online real-time execution phase; In the offline policy generation stage, a control policy mapping relationship based on the system state is pre-generated by solving a finite-time domain optimization problem offline. During the online real-time execution phase, the optimal real-time power allocation command is obtained based on the collected real-time status data of the human power grid management system and the mapping relationship of the control strategy.

2. The human body electrical grid management method according to claim 1, characterized in that, In the modeling phase, the local thermal comfort model is established based on human physiological data, which includes thermal sensitivity coefficients and comfort contribution weights of human body parts extracted from standard dynamic thermophysiological models.

3. The human body electrical grid management method according to claim 2, characterized in that, The standard dynamic thermophysiological model includes the JOS-3 model, the heat sensitivity coefficient is extracted from the heat receptor distribution coefficient (SKINR) in the JOS-3 model, and the comfort contribution weight is mapped from the perspiration / comfort distribution coefficient (SKINS) in the JOS-3 model.

4. The human body electrical grid management method according to claim 3, characterized in that, The local thermal comfort model The calculation formula is: in, It is a dimensionless comfort index; Human body parts derived from the thermal receptor distribution coefficient (SKINR) calibrated in human experiments from the JOS-3 model. Thermosensitive coefficient; Weights are assigned to comfort levels derived from the "Sweat / Comfort Distribution Coefficient (SKINS)" mapped from the JOS-model. For human body parts The rated power of the heating element, The binary decision variable for whether or not to use the heating element. The total compensation power required to maintain thermal neutrality, For real-time ambient temperature, Choose a variable for clothing type; This is a unified set that includes the parameters mentioned above.

5. The human body electrical grid management method according to claim 1, characterized in that, In the modeling phase, the weight comfort model is established based on human physiological data, which includes physiological data and subjective index data collected from multiple real-world scenarios; the weight comfort model The calculation formula is: The weight comfort model includes a time variable. The dynamic exponential cumulative effect; among which, The fatigue attenuation coefficient under load ( To ensure its physiological accuracy, It is not a fixed empirical value; it is calculated based on normalized physiological and subjective index characteristics of real subjects collected in multiple scenarios, using a least squares method with non-negative constraints. in, This represents the number of experimental samples. For the first The load (kg) of the test strip; For the first Duration of the test (h); For the first The observational normalized comfort labels for the test were obtained by fusing and truncating heart rate, skin temperature, and subjective fatigue scales; constraints This is to ensure that the weight comfort model conforms to the basic principle that "increased load / time should not increase comfort"; The total weight of the hardware representing the human body electrical grid management system is calculated using the following formula: in, To fix the weight of the electronic components in the human body electrical grid management system, For the selected clothing type The weight, and For batteries The weight and quantity of individual units. and For collector The individual weight and selection variables.

6. The human body electrical grid management method according to claim 1, characterized in that, The severe scenario protection logic in the configuration phase includes: calculating the total compensation power required to maintain thermal neutrality. At that time, the resting metabolic rate was used as a premise, and the effects of the body's own high metabolic heat production and evaporative heat dissipation were ignored; the total compensation power The calculation formula is: in, A constant temperature is set for the target skin to maintain thermal neutrality in the human body; For the selected clothing type The inherent thermal resistance; Thermal resistance of the external air layer; This represents the real-time ambient temperature.

7. The human body electrical grid management method according to claim 1, characterized in that, The battery low-temperature derating logic in the configuration phase includes: the actual usable capacity of the battery capacity used in the multi-objective optimization model after temperature derating based on the extreme ambient temperature of the target scenario.

8. The human body electrical grid management method according to claim 1, characterized in that, In the configuration phase, the decision variables of the multi-objective optimization model are uniformly defined as the hardware decision variable set. Its mathematical expression and variable domain constraint formula are as follows: in, For the first The quantity of each type of battery; The heating element and the data collector are defined as 0-1 Boolean variables respectively; The clothing type is limited to a preset clothing collection. .

9. The human body electrical grid management method according to claim 8, characterized in that, The multi-objective optimization model in the configuration phase must simultaneously satisfy the following constraints: system energy balance constraint, structural compatibility constraint, energy harvester quantity constraint, and power sufficiency constraint. The system energy balance constraint is: The total capacity of all configured batteries This refers to the actual usable capacity after temperature derating based on the extreme ambient temperatures of the target scenario. This actual usable capacity is then increased by the time step during the mission. Discrete total energy generated by the internal collector Greater than or equal to the discrete total energy consumption of the human power grid management system within the expected task time. The system energy balance constraint, as a static total constraint, is used to set the physical boundary for offline component screening during the configuration phase. The structural compatibility constraints are as follows: in For lightweight scenarios, identify variables. It is a collection of heavy hardware components; the structural compatibility constraints are used to adapt to lightweight task scenarios; The number of energy harvesters is constrained as follows: in, The maximum number of energy harvesters that can be installed in the human body power grid management system is limited by the maximum number of energy harvesters allowed to be installed, which is set according to specific task attributes or hardware physical limitations. The power adequacy constraint is: in, For human body parts The maximum rated power of the heating element is determined; under the power adequacy constraint, the total maximum rated power of the selected heating element meets the heat compensation requirements under extreme temperatures, so as to ensure sufficient heating in extreme environments.

10. The human body electrical grid management method according to claim 1, characterized in that, The optimization objective set of the multi-objective optimization model includes: minimizing the total initial cost of the system. Maximize the overall comfort score Minimize the total weight of the system The comprehensive comfort score The weighted sum of the local thermal comfort model and the weight comfort model; the multi-objective optimization model determines the hardware configuration scheme by solving the Pareto front of the optimization objective set; The optimization objective of the multi-objective optimization model is: ； in: in, The cost of fixed electronic components representing the human body electrical grid management system; This represents the cost of the i-th heating element; This represents the cost of the k-th battery; This represents the cost of the m-th energy harvester; The static balance weights for thermal comfort and weight comfort are set by the human body electrical grid management system based on scenario preferences and are read from the weight table during initialization.

11. The human body electrical grid management method according to claim 1, characterized in that, The predictive energy management strategy employs rolling time-domain control (RHC). During the offline strategy generation phase, it targets the remaining task time domain... Solve for the combined objective function that maximizes thermal comfort and minimizes battery aging penalty: in As a static balancing weight, it combines the functions of a unit conversion scalar and a preference adjustment factor, and is used to bridge the engineering trade-off between the dimensionless physiological comfort index and irreversible battery asset degradation in the comprehensive objective function; The formula for calculating the segmented linear cumulative degradation cost of the battery based on the weighted energy throughput method is as follows: in, The total capital cost of the battery modules determined for the configuration phase; The total energy throughput over the entire battery life cycle is given by the following formula: in For the expected rated cycle life, the number 2 represents bidirectional charge and discharge, denoted as 2; It is a piecewise linear cumulative function used to approximate the accelerated loss weights at the extreme state of charge (SOC) boundary; The offline solution process in the offline strategy generation stage is constrained by the following SOC evolution constraints, which realize the temporal mismatch of energy during day and night or alternating cloudy and sunny periods by simulating future time series containing photovoltaic transient fluctuations: in, The activation vector for the heating element in the future prediction time domain; and These are the power consumption of the heating array and the base power consumption, respectively. This represents the predicted transient power fluctuation of photovoltaic power generation; where, This is to prevent the risk of the human electrical grid management system failing due to overheating and battery depletion in the early stages of the mission.

12. The human body electrical grid management method according to claim 11, characterized in that, The pre-generated control strategy mapping relationship based on system state specifically involves pre-generating an optimal control strategy lookup table. Obtaining the optimal real-time power allocation command based on the collected real-time state data of the human body power grid management system and the control strategy mapping relationship specifically involves querying the optimal control strategy lookup table to obtain the optimal real-time power allocation command. In the online real-time execution phase, explicit model predictive control (MPC) is used to ensure high-frequency real-time performance on a wearable embedded microcontroller (MCU) with limited computing power. In the offline strategy generation phase, the state space of the human body power grid management system is discretized into a three-dimensional grid of remaining time, ambient temperature, and current SOC. For each grid node, a rolling time-domain control (RHC) optimization problem based on the comprehensive objective function and SOC evolution constraints is solved offline, and the activation vector of the optimal heating element at the current moment is stored in a pre-computed optimal control strategy lookup table (LUT). In the online real-time execution phase, the MCU collects real-time sensor data only at set control steps and projects it onto neighboring grid nodes, achieving a computational complexity of... The lookup operation outputs the current optimal heating element activation command.

13. A human body electrical grid management system, characterized in that, include: The modeling module is configured to: establish a local thermal comfort model that reflects the differences in thermal sensitivity of different parts of the human body and a weight comfort model that includes the cumulative effect over time; The configuration module is configured to: establish a multi-objective optimization model based on specific task scenario information; incorporate the local thermal comfort model and the weight comfort model into the optimization objective set of the multi-objective optimization model; and introduce a severe scenario protection logic and a battery low-temperature derating processing logic into the multi-objective optimization model. The hardware configuration scheme of the system is determined by solving the multi-objective optimization model; The operation control module is configured to perform real-time energy management using a predictive energy management strategy. It is implemented using a decoupled architecture, including an offline strategy generation unit and an online real-time execution unit. The offline strategy generation unit is configured to: pre-generate a control strategy mapping relationship based on the system state by solving a finite-time domain optimization problem offline; The online real-time execution unit is configured to obtain the optimal real-time power allocation instruction based on the collected real-time status data of the human power grid management system and the mapping relationship of the control strategy. One or more processors are used to perform the operations of the modeling module, configuration module, and operation control module; The memory is used to store the control strategy mapping relationship, the hardware configuration scheme of the system, and the instructions executed by the processor.

14. A computer-readable storage medium having a computer program stored thereon, characterized in that, When executed by a processor, the computer program implements the steps of the human body electrical grid management method as described in any one of claims 1-12.

Images