Flexible battery-based energy storage charging garment and mobile power supply system

Abstract

The present application belongs to the field of wearable electronic equipment and flexible energy storage technology, and particularly relates to a flexible battery-based energy storage charging garment and mobile power supply system, which comprises an energy storage charging garment body and an external mobile power supply management unit, the garment body is integrated with a non-continuous island-shaped flexible battery array, a distributed thermal management fabric layer, an embedded state sensing network and a flexible interconnection conductive path; the external unit comprises an intelligent power adapter, a photovoltaic backboard and a cloud scheduling server, and the two units are cooperated through a standardized interface and bidirectional communication, so as to realize dynamic thermal management, multi-source power supply switching, three-level safety redundancy and user behavior adaptive adjustment. Through the above scheme, the safety, energy efficiency and intelligent level of the mobile power supply system are significantly improved while the flexibility and wearing comfort of the garment are ensured.

Description

Technical Field

[0001] This invention belongs to the field of wearable electronic devices and flexible energy storage technology, specifically an energy storage and charging garment and mobile power supply system based on flexible batteries. Background Technology

[0002] Wearable electronic devices are rapidly developing and deeply penetrating various scenarios such as health monitoring and outdoor work, creating an urgent need for portable, highly integrated, comfortable, and reliable mobile energy supply systems. Traditional rigid batteries, due to their fixed size, large weight, and lack of adaptability, cannot be effectively integrated with the human body contours and flexible clothing substrates, severely limiting the battery life and user experience of wearable devices. Flexible batteries, with their excellent extensibility, lightweight design, and fabric compatibility, have become the core direction of the next generation of wearable energy systems. Embedding them into clothing to create energy storage and charging garments integrates energy with the wearable carrier, providing a new paradigm for distributed self-sustaining mobile power supply systems.

[0003] The current mainstream solution in the industry is to integrate flexible lithium-ion batteries or thin-film batteries into the lining / layer of clothing through patching, stitching, or lamination, and connect them to external power interfaces using conductive fabrics or flexible circuits. This design initially solved the problem of flexible adaptation of energy modules, and early designs focused on improving battery energy density and bending durability, which could meet the basic power supply needs of early low-power wearable devices. However, with the popularization of high-power terminals such as AR glasses and real-time physiological monitoring arrays, the power supply capacity of a single flexible battery module is insufficient, and system-level energy management and multi-source coordination have become new technical focuses.

[0004] In existing technologies, while pursuing high energy density and high output power, it is difficult to simultaneously ensure the flexibility, breathability, and long-term wearing comfort of clothing. This contradiction stems from the inherent conflict between the energy module and clothing in terms of physical properties and functional requirements. Increasing energy storage per unit area requires thickening the electrodes or stacking batteries, which leads to an increase in local stiffness and damages the fit of the clothing. Using large-area thin batteries to maintain softness is limited by current collection efficiency and internal resistance, which cannot support instantaneous high power output. Moreover, when the system needs to meet the fast charging requirements of mainstream devices, significant Joule heat will be generated inside. The heat dissipation in the closed environment of clothing is limited, which can easily accelerate battery aging and cause overheating risks. Existing systems lack dynamic perception and feedback on battery status, environmental parameters, and user activity intensity, making it impossible to adaptively adjust charging and discharging strategies. Under extreme conditions, safety and stability drop sharply. Furthermore, clothing is often treated as a passive energy container, without deep coupling with external energy replenishment units and user behavior data, resulting in low energy utilization efficiency and difficulty in building a closed-loop intelligent energy supply ecosystem. Simply improving the performance of individual battery cells cannot break through the architectural bottleneck of high energy, high flexibility, and high safety.

[0005] Therefore, the present invention provides an energy storage and charging garment and a mobile power supply system based on a flexible battery. Summary of the Invention

[0006] In order to overcome the shortcomings of the prior art, at least one technical problem raised in the background art is solved.

[0007] The technical solution adopted by this invention to solve its technical problem is as follows: An energy storage and charging garment and mobile power supply system based on flexible batteries, the garment body comprising a flexible battery array, a distributed thermal management fabric layer, an embedded state sensing network, flexible interconnected conductive paths, and a human-machine interface module. The flexible battery array consists of multiple independently packaged flexible solid-state battery units embedded in non-critical deformation areas of the garment body in a discontinuous island-like layout. The battery units are connected in parallel via low-impedance flexible busbars, and micro-current sensors and temperature sensing nodes are integrated at the positive and negative terminals of each battery unit. The discontinuous island-like layout means that the battery units maintain minimum spacing in both the warp and weft directions, ensuring that the garment has no rigid embedding in highly curved areas such as the shoulders, elbows, and knees, thereby maintaining overall flexibility and natural drape.

[0008] Preferably, the distributed thermal management fabric layer covers the inner and outer sides of the flexible battery array and is woven from phase change microcapsule fibers and thermally conductive graphene composite yarns. Its structural design ensures that when the ambient temperature is below the phase change threshold, the microcapsules remain solid, and the fabric exhibits high thermal insulation to reduce heat loss from the body. When the local temperature rises above the phase change threshold due to battery discharge, the microcapsules melt and absorb heat, while the graphene network forms a continuous thermally conductive path, rapidly diffusing heat along the fabric plane and preventing localized hotspot accumulation. This thermal management fabric layer is fixed to the garment substrate using an ultrasonic spot welding process, ensuring structural integrity and thermal response performance after multiple washes and stretching.

[0009] Preferably, the embedded state-aware network consists of a multimodal sensor array carried on a flexible printed circuit board, including but not limited to an inertial measurement unit, a skin contact humidity sensor, an ambient temperature and humidity probe, and a heart rate monitoring electrode. All sensor nodes are connected to an edge computing module located in a concealed cavity in the waist of the garment via flexible shielded cables. The edge computing module has a built-in low-power microcontroller that runs a lightweight state recognition algorithm. It can determine the user's current activity intensity level based on inertial data and, combined with the trends in skin humidity and heart rate changes, infer the user's physiological load state, thereby generating dynamic constraint commands for regulating the upper limit of charging and discharging power.

[0010] Preferably, the flexible interconnect conductive path adopts a double-layer structure design. The inner layer is an elastic polymer film coated with silver nanowires for carrying the main power transmission; the outer layer is a copper foil braided strip covered with polyethylene terephthalate for providing electromagnetic shielding and mechanical reinforcement. Electrical connection between the two layers is achieved through laser micropores, and an overcurrent fuse structure is set at key nodes. When an abnormal current is detected to exceed the safety threshold, this structure automatically disconnects the circuit to prevent the spread of thermal runaway. All conductive paths are laid along the garment seam, avoiding major stress areas to minimize conductivity degradation caused by repeated bending.

[0011] Preferably, the human-computer interaction interface module includes a magnetic fast charging output port, a wireless near-field communication tag, and a haptic feedback vibrator. The magnetic fast charging output port is integrated into the cuff or hem of the garment and contains a protocol identification chip compliant with the USBPowerDelivery specification, which can negotiate the optimal voltage and current combination with the access device. The wireless near-field communication tag stores the garment's unique identification code, a summary of the battery health status, and the last calibration timestamp for external mobile power management units to read. The haptic feedback vibrator is used to provide non-visual warnings to the user when the system enters a power-limiting mode or detects a potential overheating risk.

[0012] Preferably, the external mobile power supply management unit includes a smart power adapter, a detachable photovoltaic energy replenishment backplane, a cloud-based collaborative scheduling server, and a mobile terminal application. The smart power adapter connects to the human-machine interface module of the energy storage charging garment via a standard cable. Internally, it includes a bidirectional AC / DC converter, an isolated DC / DC conversion module, and a safety authentication coprocessor. During the charging phase, the adapter executes a constant current-constant voltage-trickle charge three-stage charging strategy. During the discharging phase, it adjusts the output power in real time according to the dynamic constraint instructions uploaded by the garment's edge computing module to ensure that it does not exceed the safe heat load limit under the current operating conditions.

[0013] Preferably, the detachable photovoltaic power replenishment back panel adopts a composite structure of flexible perovskite solar cells and breathable mesh substrate, and is connected to the back of the garment through a snap-on mechanical interface. Its output end is connected to the MPPT (maximum power point tracking) control loop of the smart power adapter through a dedicated photovoltaic input interface. The MPPT control loop adopts the perturbation observation method, but its perturbation step size is adaptively adjusted according to the rate of change of light intensity reported by the embedded state sensing network of the garment, so as to avoid power oscillation under transient light conditions such as rapid cloud movement.

[0014] Preferably, the cloud-based collaborative scheduling server is deployed on a private cloud platform and establishes a bidirectional connection with the mobile terminal application and smart power adapter through an encrypted communication channel. The server maintains a database of users' historical electricity consumption patterns, regional power grid load status information, and weather forecast data, and can predict users' potential electricity demand in the next four hours. Based on this, it generates pre-charging scheduling instructions to activate photovoltaic power generation in advance or guide users to connect to the grid for charging during off-peak hours. At the same time, the server also receives anonymized battery aging data from multiple users, continuously optimizes the global charging and discharging strategy model through a federated learning mechanism, and pushes the updated strategy parameters to each local edge computing module.

[0015] Preferably, the mobile terminal application runs on the user's smartphone and provides a graphical interface to display the current remaining battery power, available output power, thermal management status, and suggested operations. The application also supports users to manually set priority modes, including battery life priority, fast charging priority, and comfort priority. Different modes correspond to different charge and discharge curve slopes and thermal management trigger thresholds. All user settings are transmitted to the garment edge computing module via a secure Bluetooth 5.0 pairing channel and cached locally to ensure that they can still be executed in offline mode.

[0016] Preferably, in terms of system-level coordination mechanism, the energy storage charging clothing and the external mobile power supply management unit exchange data through the communication protocol defined by the IEC61850 extended specification. This protocol adds a status code field specific to wearable devices to the standard GOOSE (General Object-Oriented Substation Event) message, including activity intensity level, local hotspot location index and battery cell health status indicator. After receiving a status code containing a high activity intensity level, the smart power adapter immediately starts a predictive power reduction process. Even if the current load has not yet reached the hardware limit, it limits the maximum output current in advance to reserve thermal buffer margin.

[0017] Preferably, the system incorporates a three-level safety redundancy mechanism: the first level is the hardware level, consisting of an overcurrent fuse structure in the flexible interconnect conductive path and a ceramic separator thermal shutdown layer inside the battery cell, which can cut off the fault circuit within milliseconds; the second level is the firmware level, where the edge computing module continuously monitors the voltage consistency of each battery cell, and once it detects that the voltage of a cell deviates from the average value by more than a preset safety threshold, it is immediately logically isolated from the parallel system and the load is redistributed to the remaining healthy cells; the third level is the cloud level, where when the server detects that a user device has reported abnormal high temperature events three times in a row, it will automatically push a forced firmware update and lock the high-power output function until remote diagnosis is completed.

[0018] Preferably, in terms of thermal management coordination, the edge computing module dynamically adjusts the effective working area of ​​the thermal management fabric layer based on multi-source data collected by the embedded state-aware network: when the user is detected to be stationary and the ambient temperature is low, only the local area covering the battery array is activated; when the user enters a state of high-intensity activity and the battery load is high, the entire area of ​​the thermal conduction network is connected through a micro-relay array, and a request is simultaneously sent to the smart power adapter to reduce the output power slope to match the heat dissipation capacity. This adjustment process is implemented through a state machine, and the state transition conditions include the rate of change of temperature gradient, the amount of accumulated humidity, and the duration of activity intensity, ensuring that the thermal management actions are strictly synchronized with the actual heat load.

[0019] Preferably, in terms of multi-source power supply coordination, the system adopts a priority-based power arbitration strategy: mains power input has the highest priority, photovoltaic input is second, and the flexible battery array, as the basic energy carrier, has the lowest priority. When mains power is available, the system prioritizes using mains power for external power supply while simultaneously performing trickle charging on the flexible battery. When only photovoltaic and battery power are available, if the light intensity is higher than a preset threshold, direct photovoltaic power supply is prioritized, and excess energy is used for battery charging. If the light intensity is insufficient, the system switches to battery power supply and dynamically adjusts the upper limit of output power based on the remaining power. This arbitration logic is executed by the power management controller within the intelligent power adapter, and its decision-making is based on the energy status summary periodically reported by the garment edge computing module.

[0020] Preferably, regarding user behavior adaptation, the edge computing module incorporates an activity-energy consumption mapping table. This table, obtained through offline training, records typical power consumption curves corresponding to different activity types (such as walking, running, and sitting). During runtime, the module matches the identified activity types with the mapping table in real time, predicts the average power demand over the next five minutes, and adjusts the battery discharge depth and thermal management intensity accordingly to avoid instantaneous system overload due to sudden high loads. This mapping table supports online incremental learning; when a new user activity pattern is detected, feature data is automatically collected and uploaded to the cloud-based collaborative scheduling server, and incorporated into the global model update after verification.

[0021] The beneficial effects of this invention are as follows: This invention discloses an energy storage and charging clothing and mobile power supply system based on flexible batteries. Through the structural coupling of a non-continuous island-shaped flexible battery array with a distributed thermal management fabric layer, it avoids the damage to the clothing's flexibility caused by high energy density configurations. It utilizes the closed-loop feedback of an embedded state-aware network and an edge computing module to achieve dynamic matching of charging and discharging strategies with the user's physiological state and environmental conditions. It implements a dual safety mechanism of hardware-level overcurrent protection and battery unit logic isolation in the flexible interconnected conductive path, preventing system-level failures caused by local faults. It ensures seamless switching between multiple power supply units under a unified arbitration strategy, maintaining the continuity of external power supply. Furthermore, through a cloud-based collaborative scheduling server and a federated learning framework, it transforms the operating experience of individual devices into global optimization knowledge, continuously improving the overall energy efficiency and safety of the system. Attached Figure Description

[0022] The invention will now be further described with reference to the accompanying drawings.

[0023] Figure 1 This is a perspective view of the energy storage and charging garment body in this invention; Figure 2 This is a schematic diagram of the flexible battery array in this invention; Figure 3 This is a structural block diagram of the external mobile power supply management unit in this invention.

[0024] In the diagram: 1. Energy storage and charging garment body; 2. Flexible battery array; 3. Distributed thermal management fabric layer; 4. Embedded state sensing network; 5. Flexible interconnected conductive path; 6. Human-machine interface module. Detailed Implementation

[0025] To make the technical means, creative features, objectives and effects of this invention easier to understand, the invention will be further described below in conjunction with specific embodiments.

[0026] like Figures 1-3 As shown in the figure, an energy storage and charging garment and mobile power supply system based on flexible batteries according to an embodiment of the present invention consists of an energy storage and charging garment body 1 and an external mobile power supply management unit. The two are physically coupled and interact with each other through a standardized electrical connection interface and a two-way communication link, and jointly build a closed-loop energy ecosystem with dynamic thermal management, multi-source collaborative energy supply, safety redundancy control and user behavior adaptive adjustment capabilities. The energy storage and charging garment body 1 includes a flexible battery array 2, a distributed thermal management fabric layer 3, an embedded state sensing network 4, a flexible interconnected conductive path 5 and a human-machine interaction interface module 6. The flexible battery array 2 consists of multiple independently packaged flexible solid-state battery cells embedded in the non-critical deformation area of ​​the garment body in a non-continuous island layout. The battery cells are connected in parallel through low-impedance flexible busbars, and micro current sensors and temperature sensing nodes are integrated at the positive and negative terminals of each battery cell. The non-continuous island layout means that the battery cells maintain a minimum spacing of 15 mm in both the warp and weft directions, ensuring that there is no rigid embedding in the garment in high bending areas such as the shoulders, elbows, and knees, thereby maintaining the overall flexibility and natural drape. In one specific embodiment, the flexible solid-state battery cell adopts a LiFePO4 / PEO-LiTFSI all-solid-state system, with a single cell thickness of 0.8 mm, an areal energy density of 45 mAh / cm², an open-circuit voltage of 3.2 V, and a biaxially stretched polyimide film with a thickness of 25 micrometers and a water vapor permeability of less than 10%. -6 Barrier performance of g / (m²·day).

[0027] The distributed thermal management fabric layer 3 covers the inner and outer sides of the flexible battery array 2 and is made of phase change microcapsule fibers and thermally conductive graphene composite yarns interwoven together. The phase change microcapsules use n-octadecane as the core material and polymethyl methacrylate as the shell material, with an average particle size of 3.5 micrometers, a phase change temperature of 38℃, and a melting enthalpy of 180 J / g. The thermally conductive graphene composite yarn is formed by coating polyester filaments with 8% by mass of reduced graphene oxide, and twisting it to form a composite yarn with a diameter of 0.12 mm. Its axial thermal conductivity is 12 W / (m·K). The fabric adopts a plain weave structure with warp and weft densities of 48 threads / cm and 42 threads / cm, respectively. This ensures that when the ambient temperature is below 38℃, the microcapsules are in a solid state, and the fabric exhibits high thermal insulation properties with an apparent thermal conductivity of less than 0.04 W / (m·K). When the local temperature rises above 38℃ due to battery discharge, the microcapsules melt and absorb heat. At the same time, the graphene network forms a continuous heat conduction path in the fabric plane, rapidly diffusing heat along the warp and weft directions, keeping the local temperature rise gradient within 1.5℃ / cm. The heat management fabric layer is fixed to the garment base by ultrasonic spot welding with a weld spacing of 8 mm, a welding head frequency of 20 kHz, and a welding pressure of 0.3 MPa. This ensures that the structure maintains its structural integrity and thermal response performance after 50 standard water washes and 10,000 repeated stretching cycles, with a phase change enthalpy retention rate of not less than 92%.

[0028] The embedded state-aware network 4 consists of a multimodal sensor array carried on a flexible printed circuit board, including an inertial measurement unit (IMU), a skin contact humidity sensor, an ambient temperature and humidity probe, and a heart rate monitoring electrode. The inertial measurement unit uses a six-axis MEMS chip, including a three-axis accelerometer and a three-axis gyroscope; the skin contact humidity sensor is based on the impedance method principle, operates at a frequency of 100kHz, has a sensitivity of 0.8Ω / %RH, and is fitted inside the clothing under the armpit and back area; the ambient temperature and humidity probe is integrated into the clothing collar, with a temperature measurement range of -10℃ to +60℃ and an accuracy of ±0.3℃, and a relative humidity measurement range of 10% to 95%RH and an accuracy of ±2%RH; the heart rate monitoring electrode is a silver / silver chloride dry electrode with an area of ​​1.2cm², extracts the PPG signal through a differential amplifier circuit, and has a sampling frequency of 100Hz; All sensor nodes are connected to an edge computing module located in a concealed cavity at the waist of the garment via flexible shielded cables. This edge computing module uses an ARM Cortex-M4F core microcontroller with a main frequency of 80MHz, 256KB Flash and 64KBS RAM, and runs a lightweight state recognition algorithm. It can determine the user's current activity intensity level based on inertial data: divided into four levels: sitting, walking, brisk walking and running. It also combines the skin moisture change rate (dH / dt) and heart rate variability (HRV) indicators to infer the user's physiological load state, and then generate dynamic constraint commands for regulating the upper limit of charging and discharging power. The algorithm uses a data segment with a sliding window length of 10 seconds, updates the output every 2 seconds, and has a delay of no more than 300 milliseconds.

[0029] The flexible interconnect conductive path 5 adopts a double-layer structure design. The inner layer is an elastic polymer film coated with silver nanowires, which is used to carry the main power transmission. The outer layer is a copper foil braided tape covered with polyethylene terephthalate (PET), which is used to provide electromagnetic shielding and mechanical reinforcement. The silver nanowire film is made by spin-coating silver nanowires with a diameter of 30nm and a length of 15μm onto a thermoplastic polyurethane (TPU) base film at an areal density of 12mg / m². After hot pressing at 120℃, the sheet resistance is 35mΩ / sq and the elongation at break is greater than 300%. The copper foil braided tape is made by cutting 12μm thick electrolytic copper foil into 0.3mm wide strips and weaving them into a PET mesh at a helix angle of 45° with a weaving density of 16 strands / cm. Its shielding effectiveness (SE) is not less than 65dB in the frequency band from 30MHz to 1GHz. Electrical connection between the two layers is achieved through laser micropores with a diameter of 80μm and a spacing of 2mm. The micropores are filled with conductive silver paste. An overcurrent fusing structure is set at the critical node. This structure is composed of a nickel-chromium alloy film with a width of 0.15mm and a thickness of 8μm. The fusing current threshold is 8A and the response time is less than 50ms. All conductive paths are laid along the seams of the garment, avoiding major stress areas, in order to minimize the attenuation of conductivity caused by repeated bending. After several dynamic bending tests, the change rate of resistance in the conductive path is less than 5%.

[0030] Human-computer interaction interface module 6 includes a magnetic fast charging output port, a wireless near-field communication tag, and a tactile feedback vibrator; The magnetic fast charging output port is integrated into the cuff or hem of the garment. Internally, it contains a protocol identification chip compliant with the USB Power Delivery 3.0 specification, supporting four preset voltage and current combinations: 5V / 3A, 9V / 3A, 15V / 3A, and 20V / 5A, with a maximum output power of 100W. The wireless near-field communication tag adopts the ISO / IEC 14443 Type B standard, with a storage capacity of 1KB. The first 128 bytes store the garment's unique identification code, the middle 256 bytes store a summary of the battery health status, including SOH, cycle count, and maximum allowable discharge current, and the last 64 bytes store the last calibration timestamp. The haptic feedback vibrator is a linear resonant actuator (LRA) with a resonant frequency of 180Hz, a driving voltage of 3.3V, and a peak vibration acceleration of 0.8g. It provides non-visual warnings to the user when the system enters power-limiting mode or detects potential overheating risks. Vibration modes include single pulse, double pulse, and continuous pulse.

[0031] The external mobile power management unit includes a smart power adapter, a detachable photovoltaic backplane, a cloud-based collaborative scheduling server, and a mobile terminal application. The smart power adapter connects to the human-machine interface module 6 of the energy storage charging garment via a standard cable. It contains a bidirectional AC / DC converter, an isolated DC / DC conversion module, and a safety authentication coprocessor. During the charging phase, the adapter employs a three-stage charging strategy: constant current, constant voltage, and trickle charge. The constant current phase has a current of 2C (C being the battery's rated capacity), the constant voltage phase has a voltage of 3.65V, and the trickle charge phase has a current of 0.05C. The battery is considered fully charged when the voltage stabilizes at 3.65V ± 0.01V and the current decreases to below 0.05C for 10 minutes. During the discharging phase, the adapter adjusts its output power in real time based on the dynamic constraint instructions uploaded by the clothing edge computing module to ensure that it does not exceed the safe thermal load limit under the current operating conditions. For example, when the edge computing module reports a high-intensity motion state and the local temperature exceeds 42°C, the adapter will limit the maximum output current to 1.5A, even if the load device requests higher power.

[0032] The detachable photovoltaic power-generating back panel adopts a composite structure of flexible perovskite solar cells and breathable mesh substrate, and is connected to the back of clothing through a snap-on mechanical interface; The perovskite solar cell uses a CH3NH3PbI3 light-absorbing layer with a thickness of 500 nm, deposited on an ITO / PET flexible substrate, with an effective area of ​​300 cm². Under standard test conditions (AM1.5G, 1000 W / m², 25℃), the photoelectric conversion efficiency is 18.7%. The breathable mesh substrate is a polyester fiber woven mesh with a porosity of 65% and an air permeability >500 L / (m²·s). Its output is connected to the MPPT (Maximum Power Point Tracking) control loop of the smart power adapter via a dedicated photovoltaic input interface; the MPPT control loop adopts the perturbation observation method, but its perturbation step size ΔV is adaptively adjusted according to the rate of change of light intensity dE / dt reported by the clothing embedded state sensing network 4, and the specific relationship is as follows: in, It represents the rate of change of light intensity per unit time (unit: W / (m²·s)). Voltage perturbation step size (unit: V); This mechanism avoids power oscillations under transient lighting conditions such as rapidly moving clouds. Real-world measurements show that in typical urban outdoor scenarios, the MPPT convergence time is reduced by 37%, and the standard deviation of steady-state power fluctuations is reduced to 1.2W.

[0033] The cloud-based collaborative scheduling server is deployed on a private cloud platform and establishes a two-way connection with the mobile terminal application and smart power adapter through a TLS1.3 encrypted communication channel. The server maintains a database of users' historical electricity consumption patterns, records electricity consumption, power, ambient temperature, activity type, regional power grid load status information and weather forecast data every 5 minutes, and can predict users' potential electricity demand in the next four hours and generate pre-charging scheduling instructions accordingly. For example, if it is predicted that users will engage in high-intensity outdoor activities between 18:00 and 19:00 and the current remaining battery power is less than 40%, then the mains power can be activated to charge to 80% SOC during the off-peak electricity price period of 14:00 to 16:00 (0.3 yuan / kWh). Meanwhile, the server also receives anonymized battery aging data from multiple users (including ΔV / ΔQ curves and internal resistance growth rates for each cycle) and continuously optimizes the global charge-discharge strategy model through a federated learning mechanism. Federated learning uses the FedAvg algorithm, which aggregates 100 client-local models in each round, with a learning rate η=0.01, a number of local training rounds E=5, a batch size B=32, and a loss function that is the sum of mean squared error (MSE) and a penalty term for violating security constraints. The updated strategy parameters (such as the maximum discharge C-rate under different SOH conditions) are pushed to each local edge computing module via OTA, with an update cycle of 7 days.

[0034] The mobile application runs on the user's smartphone and provides a graphical interface to display the current remaining battery power (accuracy ±2%), available output power (unit: W), thermal management status (partial activation or full area activation), and suggested operations (such as suggesting connecting to mains power and reducing the use of high-power devices); the application also supports users to manually set priority modes, including battery life priority, fast charging priority, and comfort priority; In the battery life priority mode, the discharge cutoff voltage is set to 3.0V and the thermal management trigger threshold is 45℃; in the fast charging priority mode, the charging current is increased to 2.5C and the thermal management trigger threshold is reduced to 40℃; in the comfort priority mode, the discharge power limit is set to 15W and the thermal management trigger threshold is 38℃. All user settings are transmitted to the garment edge computing module via a secure Bluetooth 5.0 pairing channel and cached in local Flash to ensure they can still be executed offline.

[0035] In terms of system-level coordination mechanisms, the energy storage charging clothing and the external mobile power supply management unit exchange data through the communication protocol defined by the IEC61850 extended specification; This protocol adds wearable device-specific status code fields to the standard GOOSE (GenericObjectOrientedSubstationEvent) message, including activity intensity level (0-3 integer code), local hotspot location index (represented by battery cell number, such as B03, B07), and battery cell health indicator (0=normal, 1=slight degradation, 2=severe degradation). Upon receiving a status code indicating a high activity level (≥2), the smart power adapter immediately initiates a predictive power reduction process, limiting the maximum output current even if the current load has not yet reached the hardware limit. For example, when the activity intensity level changes from 1 to 3, the adapter reduces the upper limit of the output current from 3A to 2A within 200ms, reserving thermal buffer margin.

[0036] The system has a built-in three-level safety redundancy mechanism: the first level is the hardware level, which consists of the overcurrent fuse structure in the flexible interconnect conductive path 5 and the ceramic separator thermal shutdown layer inside the battery cell. The ceramic diaphragm uses an Al2O3-coated polyolefin-based membrane with a coating thickness of 3 μm. When the temperature exceeds 130℃, the pores of the diaphragm close, and the ionic conductivity drops sharply to 10. -8The first level is below S / cm, which can cut off the fault circuit in milliseconds; the second level is firmware level, where the edge computing module continuously monitors the voltage consistency of each battery cell with a sampling frequency of 10Hz. Once it is found that the voltage of a cell deviates from the average value by more than the preset safety threshold (±50mV), it is immediately logically isolated from the parallel system through the MOSFET switch array and the load is redistributed to the remaining healthy cells. The load redistribution algorithm ensures that the current deviation of each cell does not exceed 10%; the third level is cloud level. When the server detects that a user device reports abnormal high temperature events three times in a row (single duration > 60s and temperature > 55℃), it will automatically push a forced firmware update, lock the high power output function (maximum output power limited to 5W) until the remote diagnosis is completed. The diagnosis process includes reading historical temperature logs, verifying sensor calibration status and checking the integrity of thermal management fabric.

[0037] In terms of thermal management collaboration, the edge computing module dynamically adjusts the effective working area of ​​the thermal management fabric layer based on multi-source data collected by the embedded state-aware network 4. This adjustment is achieved through a miniature relay array, with the relays being solid-state optocouplers, a drive voltage of 3.3V, and an on-resistance of <0.1Ω; When the user is detected to be stationary (activity intensity level 0) and the ambient temperature is below 25°C, only a local area covering the battery array (approximately 40% of the total area) is activated; when the user enters a high-intensity exercise state (activity intensity level ≥ 2) and the battery load power exceeds 20W, the entire area heat conduction network is activated, and a request is simultaneously sent to the smart power adapter to reduce the output power slope to match the heat dissipation capacity. The adjustment process is implemented through a finite state machine. The state transition conditions include the rate of change of temperature gradient (dT / dx > 2℃ / cm), the cumulative humidity (∫Hdt > 500%RH·s), and the duration of the activity intensity (t > 120s), ensuring that the thermal management action is strictly synchronized with the actual heat load. Actual tests show that, under running conditions (30W power), enabling full-area thermal management can reduce the maximum battery temperature from 58℃ to 46℃, reducing the rate of temperature rise by 62%.

[0038] In terms of multi-source power supply coordination, the system adopts a priority-based power arbitration strategy: the mains power input has the highest priority, the photovoltaic input is second, and the flexible battery array 2, as the basic energy carrier, has the lowest priority. The arbitration logic is executed by the power management controller within the smart power adapter, and its decision is based on the energy status summary periodically (every 5 seconds) reported by the garment edge computing module. The specific arbitration rules are as follows: When mains power is available (AC input voltage > 85V), the system prioritizes using mains power to supply external power, while simultaneously performing trickle charging (current 0.05C) on the flexible battery. When only photovoltaic and battery are available, if the light intensity E > 300W / m², photovoltaic direct supply should be used first, and excess energy should be used for battery charging (charging current = min(1C,(Ppv-Pload) / Vbat)). If E ≤ 300W / m², the system switches to battery power and dynamically adjusts the maximum output power based on the remaining SOC: Pmax = 30W when SOC > 60%; Pmax = 20W when 30% < SOC ≤ 60%; and Pmax = 10W when SOC ≤ 30%. This strategy ensures that the system can continuously power devices such as mobile phones and drones for at least 4 hours (based on an average load of 15W) in scenarios without mains power.

[0039] In terms of user behavior adaptation, the edge computing module has a built-in activity-energy consumption mapping table, which is obtained through offline training and records the typical power consumption curves corresponding to different activity types. The training data came from synchronously collected data of 50 volunteers performing standardized activities (sitting, walking at 4km / h, brisk walking at 6km / h, and running at 10km / h) in a laboratory environment, with a sampling frequency of 100Hz and a duration of 30 minutes for each activity. The mapping table is stored in the form of a lookup table, with the key being the activity type code and the value being the probability distribution of power demand in the next 5 minutes (mean μ and standard deviation σ). During operation, the module will match the activity type identified in real time with the mapping table, predict the average power demand in the next five minutes, and adjust the battery discharge depth and thermal management intensity accordingly. For example, when the system detects that a user is switching from sitting to running, it raises the maximum depth of discharge from 30% to 70% 10 seconds in advance and activates full-area thermal management. This mapping table supports online incremental learning: when a new user activity pattern, such as cycling or mountain climbing, is detected, feature data, including acceleration spectrum, heart rate rise slope, and humidity change trend, is automatically collected and uploaded to the cloud-based collaborative scheduling server; the server performs cluster analysis on the new data (using the DBSCAN algorithm, ε=0.5, MinPts=10), and if it is confirmed as a new category, a new mapping entry is generated, which is then incorporated into the global model update after A / B testing verification, with an update cycle of 14 days.

[0040] The foregoing has shown and described the basic principles, main features, and advantages of the present invention. Those skilled in the art should understand that the present invention is not limited to the above embodiments. The embodiments and descriptions in the specification are merely illustrative of the principles of the invention. Various changes and modifications can be made to the invention without departing from its spirit and scope, and all such changes and modifications fall within the scope of the present invention as claimed. The scope of protection of the present invention is defined by the appended claims and their equivalents.

Claims

1. A flexible battery-based energy storage and charging garment and mobile power supply system, characterized in that, The system includes an energy storage and charging garment body (1) and an external mobile power supply management unit; The energy storage and charging garment body (1) includes: The flexible battery array (2) consists of multiple flexible solid-state battery units embedded in the non-critical deformation area of ​​the garment body in a non-continuous island layout. The multiple flexible solid-state battery units are connected in parallel through low-impedance flexible busbars, and micro current sensors and temperature sensing nodes are integrated at the positive and negative terminals of each flexible solid-state battery unit. A distributed thermal management fabric layer (3) covers the inner and outer sides of the flexible battery array (2). It is made of phase change microcapsule fibers and thermally conductive graphene composite yarns. It is used to exhibit high thermal insulation when the ambient temperature is below the phase change threshold and melt to absorb heat and form a planar thermal conductive path to diffuse heat when the local temperature exceeds the phase change threshold. The embedded state-aware network (4) includes an inertial measurement unit, a skin contact humidity sensor, an ambient temperature and humidity probe, and a heart rate monitoring electrode. All sensor nodes are connected to the edge computing module located at the waist of the garment via flexible shielded cables. Flexible interconnect conductive path (5) is a double-layer structure. The double-layer structure is electrically connected through laser micro-holes, and an overcurrent fuse structure is set at key nodes. Human-computer interaction interface module (6); The external mobile power supply management unit includes a smart power adapter, a detachable photovoltaic energy replenishment backplane, a cloud-based collaborative scheduling server, and a mobile terminal application. The edge computing module runs a lightweight state recognition algorithm to determine the user's activity intensity level based on inertial data and infer the physiological load state by combining skin humidity and heart rate change trends, generating dynamic constraint instructions for adjusting the upper limit of charging and discharging power; the smart power adapter adjusts the output power in real time according to the dynamic constraint instructions; the cloud-based collaborative scheduling server predicts future electricity demand based on the user's historical electricity consumption patterns, regional power grid load, and weather forecast data, and generates pre-charging scheduling instructions.

2. The energy storage and charging clothing and mobile power supply system based on a flexible battery according to claim 1, characterized in that, The discontinuous island layout refers to the fact that each flexible solid-state battery cell maintains a minimum spacing of not less than 15 mm in both the longitudinal and latitudinal directions, ensuring that there is no rigid embedding in the high bending areas of the shoulders, elbows, and knees.

3. The energy storage and charging clothing and mobile power supply system based on a flexible battery according to claim 1, characterized in that, In the distributed thermal management fabric layer (3), the phase change microcapsules use n-octadecane as the core material and polymethyl methacrylate as the shell material.

4. The energy storage and charging clothing and mobile power supply system based on a flexible battery according to claim 1, characterized in that, It also includes a three-level safety redundancy mechanism, including hardware-level overcurrent fuses and thermal shutdown, firmware-level battery cell logic isolation, and cloud-level forced firmware updates for abnormal events.

5. The energy storage and charging clothing and mobile power supply system based on a flexible battery according to claim 1, characterized in that, The overcurrent fusing structure in the flexible interconnect conductive path (5) is made of a nickel-chromium alloy thin film; The battery cell is equipped with a ceramic separator thermal shutdown layer.

6. The energy storage and charging clothing and mobile power supply system based on a flexible battery according to claim 1, characterized in that, During the discharge phase, the intelligent power adapter performs a predictive power reduction process based on the received activity intensity level status code. The communication protocol adds a wearable device-specific status code field to the GOOSE message, including activity intensity level, local hotspot location index, and battery cell health indicator.

7. The energy storage and charging clothing and mobile power supply system based on a flexible battery according to claim 1, characterized in that, The detachable photovoltaic backsheet uses flexible perovskite solar cells, and its MPPT control loop adopts the perturbation observation method, with the perturbation step size ΔV adaptively adjusted according to the rate of change of light intensity |dE / dt|. When |dE / dt| < 50 W / (m²·s), ΔV = 0.1 V; When 50 ≤ |dE / dt| < 200 W / (m²·s), ΔV = 0.3 V; When |dE / dt|≥200W / (m²·s), ΔV=0.5V.

8. The energy storage and charging clothing and mobile power supply system based on a flexible battery according to claim 1, characterized in that, The multi-source power supply coordination adopts a priority arbitration strategy: The order of power input, photovoltaic, and flexible battery array (2) is sorted in descending order.

9. The energy storage and charging clothing and mobile power supply system based on a flexible battery according to claim 1, characterized in that, The coordinated thermal management is implemented using a finite state machine: When the user is stationary and the ambient temperature is <25℃, thermal management is activated only in a local area of ​​the battery array. When the load power is greater than 20W, the entire area heat conduction network is activated, and a request is made to reduce the output power slope simultaneously. The conditions for state transition include the rate of change of temperature gradient dT / dx > 2℃ / cm, the cumulative humidity ∫Hdt > 500%RH·s, and the duration of activity intensity t > 120s.

10. The energy storage and charging clothing and mobile power supply system based on a flexible battery according to claim 1, characterized in that, The edge computing module has a built-in mapping table that records the probability distribution of the average power demand for the next five minutes corresponding to different activity types. When an activity type switch is detected, adjust the upper limit of discharge depth and thermal management intensity in advance; The mapping table supports online incremental learning: when a new activity pattern is detected, acceleration spectrum, heart rate rise slope and humidity change trend feature data are collected, uploaded to the cloud collaborative scheduling server, and incorporated into the global model update after cluster analysis and A / B testing verification.

Images