Battery intelligent control method under cold chain environment

Abstract

This invention aims to provide a smart battery control method for cold chain environments, comprising the following steps: A) Sensor data acquisition, real-time acquisition of ambient temperature T, ambient humidity H, condensate level W, battery voltage U, and battery temperature Tb; B) Data preprocessing; C) Setting preset thresholds, including temperature threshold, humidity threshold, SOC threshold, and insulation resistance threshold; D) Establishing a fuzzy rule base, inputting ambient temperature T, ambient humidity H, condensate level, alarm level, and BMS status into corresponding functions of the fuzzy rule base, outputting various control parameters for control by a controller, while simultaneously providing feedback on the status. An adaptive neural fuzzy control algorithm is used to correct and optimize the above functions in real time, achieving nonlinear dynamic regulation. The method of this invention can effectively solve the technical problems of battery range degradation, high short-circuit risk, and shortened lifespan in cold chain environments characterized by low temperature, high humidity, and easy condensation.

Description

Technical Field

[0001] This invention relates to the field of power battery protection structure technology, specifically to a battery intelligent control method under cold chain conditions. Background Technology

[0002] With the rapid development of the cold chain logistics industry, electric operating vehicles have been widely used in cold chain warehouses, refrigerated transport vehicles, and low-temperature logistics centers due to their advantages such as environmental friendliness, low noise, and low operating costs. However, the unique characteristics of the cold chain environment pose severe challenges to power batteries: Low-temperature challenges: When the ambient temperature is below 0℃, the viscosity of lithium battery electrolyte increases significantly, and the lithium-ion migration rate decreases, leading to increased internal resistance and capacity decay (capacity decay can reach 30% to 50% at -20℃), and in severe cases, even failure to start. At the same time, low-temperature charging is prone to lithium dendrite growth, causing permanent capacity loss and safety hazards.

[0003] High humidity and condensation challenges: The relative humidity in cold chain environments is generally higher than 70%, and the temperature fluctuates frequently (such as when the cold storage door is open). When the surface temperature of the battery pack is lower than the ambient dew point temperature, water vapor condenses on the inner wall of the casing and the surface of the cells, resulting in a decrease in insulation resistance (which can drop sharply from hundreds of MΩ to the MΩ level), corrosion of the terminals, and a sharp increase in the risk of short circuits.

[0004] Thermal runaway risk: During cold chain operations, vehicles frequently enter and exit high and low temperature areas, and the battery undergoes severe temperature cycles, which accelerates aging; if local overheating causes thermal runaway, the consequences will be extremely serious in the closed cold chain space.

[0005] Vibration and shock: Frequent starting and stopping of cold chain vehicles and loading and unloading of goods generate continuous vibration (15Hz~20Hz medium and high frequency) and impact (up to 30g), which can easily lead to loose cell connections and structural fatigue.

[0006] Existing technologies have significant shortcomings: traditional insulation structures use a single material, resulting in limited thermal insulation performance and low-temperature brittleness; temperature control systems are mostly simple on / off controls, unable to achieve coordinated temperature and humidity regulation; there is a lack of effective anti-condensation prediction and proactive intervention methods; thermal management, dehumidification, and safety protection subsystems operate independently without synergy; and there is no self-learning optimization capability based on cloud-based big data. Therefore, a battery intelligent control method under cold chain conditions that combines efficient protection and intelligent regulation is needed to adapt to the usage requirements of cold chain environments. Summary of the Invention

[0007] This invention aims to provide a smart battery control method for cold chain environments. This method effectively addresses the technical problems of reduced battery range, high short-circuit risk, and shortened lifespan caused by low temperature, high humidity, and condensation in cold chain environments. The technical solution of the present invention is as follows: The aforementioned intelligent battery control method under cold chain conditions includes the following steps: A. Sensor data acquisition: Real-time acquisition of ambient temperature T, ambient humidity H, condensate water level W, battery voltage U, and battery temperature Tb through temperature sensor, humidity sensor, water level sensor, and battery management system. B. Data preprocessing: Filtering and denoising the collected data, removing outliers, and obtaining effective monitoring data. C. Set preset threshold Temperature threshold: T heat,on =5℃; T heat,off =15℃; T fan,on =28℃; T safe =45℃; Humidity threshold: R Hon =65%; R Hoff =55%; R Hdanger =85%; SOC threshold: SOC min =5%; SOC heat,max =20%, maximum heating power limit at low SOC; Insulation resistance threshold: R iso,min =1.0 MΩ; D. Establish a fuzzy rule base, in which temperature control function, humidity control function, water level control calculation function, and BMS system enable function are defined respectively. Ambient temperature T, ambient humidity H, water level and alarm level, and BMS status are input into the corresponding functions, respectively, and the output variables are heating power, dehumidifier operating level, ventilation solenoid valve status, drain pump status, and alarm level. These are controlled by a controller, which also provides status feedback. An adaptive neural fuzzy control algorithm is used to correct and optimize the above functions in real time, achieving nonlinear dynamic regulation. Specific variable outputs are as follows: P h ∈[0,1], heater power factor 0-100%; P d ∈[0,1], dehumidifier power factor; F v ∈{0,1}, Ventilation solenoid valve status: 0 closed, 1 open; F p ∈{0,1}, Drainage pump status; A∈{0,1,2}, Alarm level: 0 None, 1 Warning, 2 Critical.

[0008] In step D, the core decision-making logic of the adaptive neural fuzzy control algorithm is as follows: a. Temperature control: The preset ambient low temperature threshold is 0℃, the target temperature range for discharge is 5-10℃, and the target temperature range for charging is 10-15℃. When T < 0℃, the PTC heating film is activated, and the heating power is dynamically adjusted according to the difference between T and the target temperature of 5℃ until the temperature reaches the corresponding target range. When the battery core temperature Tb > 25℃ is detected, the heating power is reduced or the heating film is turned off. At the same time, the positive pressure relief valve is used to accelerate the air circulation inside and outside the casing to enhance the heat dissipation effect. b. Humidity Control: The preset high humidity threshold is 60%RH, and the target humidity range is 30%-50%RH. When H>60%RH is detected, the semiconductor dehumidifier is activated, and the operating level is adjusted according to the degree of humidity exceeding the standard until the humidity drops to the target range, avoiding excessive dehumidification that could cause battery components to dry out and crack.

[0009] c. Water level control: The condensate water level threshold is preset, and a safe upper limit is set according to the capacity of the collection box. When W≥ this threshold is detected, the controller triggers an audible and visual drainage reminder, controls the opening of the drainage pump and the solenoid valve of the water pipe, and closes the solenoid valve and drainage pump after the water level drops to the safe lower limit.

[0010] The temperature control function Specifically as follows: It is divided into high-temperature ventilation mode and low-temperature heating mode: High-temperature ventilation mode: (1); Low temperature heating mode: (2); Gradient adjustment rules: (3); Among them, T b T represents the current temperature of the battery. heat,on The threshold temperature for heating to start; T heat,off The heating shut-off threshold temperature; T target Target temperature; F cool This refers to the fan speed setting in high-temperature ventilation mode; P h This represents the percentage of heating power in low-temperature heating mode; ΔP h The heating power gradient adjustment amount; T is the cold chain ambient temperature; SOC is the remaining battery charge; R i This represents the battery's internal resistance.

[0011] The humidity control function Specifically as follows: Dehumidifier power control: (4); Where RH represents the current ambient relative humidity; P d R represents the percentage of the dehumidifier's output power.Hdanger Hazardous humidity threshold; RH on Set the humidity threshold for the dehumidifier; RH off The dehumidifier is set to the humidity threshold for shutdown. Enhanced anti-condensation control: (5); Among them, the safety threshold Condensation control gain ;P h,add To prevent condensation, the auxiliary heating power is increased; T dew This is the dew point temperature.

[0012] Natural ventilation and dehumidification conditions: (6); Among them, the minimum humidity difference threshold is RH out The relative humidity of the external environment; T out The external ambient temperature.

[0013] The aforementioned anti-condensation enhancement control is activated based on anti-condensation judgment; the anti-condensation judgment is calculated using the dew point temperature calculation formula, and the object surface temperature is below T. dew Condensation will occur at this time, and the specific formula is as follows: (7); in: And parameters: T e RH represents ambient temperature and relative humidity.

[0014] The calculation formula in the water level and alarm level calculation module is as follows: Water level control algorithm: (8); Alarm level: (9); Where W is the water level status flag, h w F represents the water level. p For drainage pump control flags; A water This is the flood alarm level.

[0015] The BMS system enable function is defined as follows: System enable function : (10); in, F represents logical AND; bms For BMS fault flags; T b,max This is the highest battery temperature; T safe SOC refers to the battery's safe temperature threshold; SOC refers to the remaining battery capacity.min R is the minimum permissible SOC threshold for the battery. iso R is the battery insulation resistance. iso,min This is the minimum permissible insulation resistance threshold.

[0016] Considering safety interlocking, the final output of all actuators is: ; ; (11); in, This is an indicator function; it returns 1 if the condition is met, and 0 otherwise. P represents the final output value of the heating power. h This is the original heating power control value; This represents the final output power of the dehumidifier. The final output value of the natural ventilation control flag; F v This is the original natural ventilation control marker position.

[0017] When the BMS reports a critical fault: if or Then we have: .

[0018] The adaptive neural fuzzy control algorithm process is as follows: a. Input / output definition: Input layer (5-dimensional): ; Output layer (5-dimensional): ; Among them, RH avg P represents the average relative humidity of the environment. h P is the heating power. d For dehumidification power, F v Ventilation status, F p In drainage state, A l Alarm level; b. Perform fuzzy rules and reasoning: The j-th fuzzy rule (j=1,2,...,10): ; This includes fuzzy subsets (such as "low temperature" and "medium humidity"). For fuzzy rule consequent parameters; c. Self-learning parameter correction: Error function: (12) Where M is the total number of training samples; k is the sample iteration index; Y represents the target true value for the k-th sample. k This is the model prediction value for the k-th sample. Parameter update (gradient descent): (13) Where η is the learning rate. This is the partial derivative of the error with respect to the parameter.

[0019] Iterate every 24 hours to correct the fuzzy rule parameters and minimize the error E.

[0020] In step B, the data preprocessing uses the Kalman filter algorithm to remove outliers caused by sensor errors and environmental interference, and then normalization is used to unify the scale of data from different dimensions to obtain accurate and effective monitoring data. The multi-objective optimization function for Kalman filtering denoising is as follows: ;

[0021] in, , , These are weighting coefficients, typically This indicates that temperature control has the highest priority.

[0022] The aforementioned intelligent battery control method for cold chain environments utilizes a battery protection device with the following structure for cold chain environments: The battery protection device for cold chain environments includes a shock-absorbing buffer layer, an outer shell, an inner shell, a heating film assembly, a microchannel liquid cooling plate, a controller, and a remote monitoring module. Both the outer shell and the inner shell are sealed shells; A shock-absorbing buffer layer is provided on the bottom surface of the outer shell; The inner bottom surface of the outer shell is provided with multiple heat-insulating supports, and the outer bottom surface of the inner shell is fixedly installed on these heat-insulating supports. A gap is left between the outer wall of the inner shell and the inner wall of the outer shell, which is filled with nano-aerogel to form a nano-aerogel layer. A heating film assembly is provided on the inner bottom surface of the inner shell, a microchannel liquid cooling plate is provided on the heating film assembly, and a battery cell module is installed on the top surface of the microchannel liquid cooling plate. At this time, the large surface of the battery cell module is in contact with the microchannel liquid cooling plate. The top of the outer casing is equipped with a positive pressure relief valve; the battery cell module is equipped with a temperature sensor and a humidity sensor. The controller is connected to the temperature sensor, humidity sensor, and BMS of the battery cell module. The controller uploads monitoring data and control status information to the cloud platform in real time via the Internet of Things, issues alarms according to the alarm level, and accepts remote parameter commands for control.

[0023] The outer shell is made of aluminum alloy and has a fire-retardant coating on its outer surface; the inner shell is made of stainless steel and its inner wall is coated with high-temperature resistant insulating paint. Preferably, the fire-retardant coating is a 0.2-0.4mm ceramicized silicone rubber; Preferably, the nano-aerogel is a SiO2 aerogel felt with a thickness of 4mm to 6mm, a thermal conductivity of ≤0.018 W / (m·K) at 25℃, and no cracks when folded 180° at -30℃; Preferably, the outlet of the positive pressure relief valve is covered with a hydrophobic and breathable membrane with a pore size of 0.08-0.12μm, and the opening pressure of the positive pressure relief valve is 4kPa±0.5kPa.

[0024] The shock-absorbing buffer layer includes a honeycomb aluminum layer and rubber supports. The honeycomb aluminum layer is disposed on the bottom surface of the outer shell, and multiple sets of rubber supports are evenly spaced on the bottom surface of the honeycomb aluminum layer, which are then installed on the vehicle chassis.

[0025] The heating film assembly includes a PTC heating film and a graphene heat-spreading film. The PTC heating film is disposed on the bottom surface of the outer shell, and the graphene heat-spreading film is disposed on the top surface of the PTC heating film. The graphene heat-spreading film is in contact with the microchannel liquid cooling plate. Preferably, the thickness of the graphene heat spreader film is 0.4-0.6 mm.

[0026] The heat insulation support is made of ceramic material; preferably, the battery module includes multiple battery cells, and a heat insulation pad with a thickness of 1-2mm is provided between each battery cell. The heat insulation pad has a double-layer structure, consisting of a ceramicized silicone rubber layer and a SiO2 aerogel felt layer.

[0027] The sensor data acquisition system of this invention is based on "multi-dimensional perception + real-time transmission + precise synchronization" to build a full-scene monitoring network covering battery environment, battery body, and structural status. The system architecture is divided into three layers: 1. Sensing Layer: Deploy 6 types of core sensors to collect ambient temperature, ambient humidity, condensate water level, battery cell voltage, battery core temperature, and casing vibration acceleration, achieving full coverage of five parameters: temperature, humidity, water level, electricity, and vibration. 2. Transmission layer: The system adopts a dual-bus architecture of CAN bus (500kbps) + RS485 (115200bps). The CAN bus is responsible for transmitting voltage and battery temperature data related to the battery management system (BMS), while the RS485 is responsible for transmitting ambient temperature, humidity, water level, and vibration data to ensure real-time data transmission (delay ≤100ms). 3. Processing layer: Equipped with an STM32H743 microcontroller (480MHz), it performs preliminary caching and synchronization processing on the acquired raw data, providing a foundation for subsequent data preprocessing.

[0028] The beneficial effects of this invention are as follows: The method of this invention is based on multi-sensor data fusion to achieve real-time monitoring and dynamic control of temperature, humidity and battery status, ensuring safe, efficient and long-lasting operation of the battery in a cold chain environment of -30℃ to 60℃.

[0029] The method of this invention preferably employs a battery protection device for cold chain environments, which utilizes a double-layer sealed shell structure consisting of an outer shell and an inner shell. The outer shell is preferably made of high-strength aluminum alloy with a fire-retardant coating, combining lightweight, wear resistance, corrosion resistance, and fire resistance to withstand impacts and environmental erosion during cold chain operations. The inner shell is installed inside the outer shell, and a nano-aerogel insulation layer is filled between the outer and inner shells, with a preferred thickness of 4mm to 6mm and a thermal conductivity ≤0.018W / (m·K) at room temperature. Even at a low temperature of -30℃, it maintains its flexibility and does not crack, effectively blocking the transfer of thermal bridges, reducing heat exchange between the inside and outside of the battery pack, and inhibiting the accumulation of condensate on the inner wall of the shell.

[0030] The battery protection device for cold chain environments in this invention incorporates a microchannel liquid cooling plate, a PTC heating film, and a graphene heat dissipation film. This allows for preheating of the battery at low temperatures and heat dissipation at high temperatures. During installation, the microchannel liquid cooling plate is tightly fitted to the large surface of each battery cell, with an ethylene glycol-water mixture flowing within it, adapting to the wide temperature range requirements of cold chain applications. The PTC heating film is integrated with the liquid cooling plate, with the graphene heat dissipation film sandwiched between them, ensuring that the temperature difference on the cell surface is controlled within ≤3℃, guaranteeing temperature uniformity. This component enables preheating and startup of the battery at extreme low temperatures of -30℃, and also achieves efficient heat dissipation when the battery generates heat during operation or when the ambient temperature rises, maintaining the battery within its optimal operating temperature range.

[0031] The battery protection device in the cold chain environment of this invention is equipped with an anti-condensation and pressure balance structure, namely a positive pressure relief valve mounted on the shell, whose opening pressure is set to 3kPa~5kPa, and has the dual functions of one-way venting and preventing condensate backflow. Preferably, a hydrophobic and breathable membrane is integrated at the outlet of the pressure relief valve, with a permeability of ≥2000mL / min, which can effectively block droplets and salt spray with a particle size ≥0.1μm from entering the shell. This design achieves three functions: ① internal and external pressure balance to prevent negative pressure from drawing in humid air; ② one-way venting to block external water vapor backflow; ③ overpressure relief to suppress the accumulation of thermal runaway pressure.

[0032] In the cold chain environment of the battery protection device of the present invention, the inner wall of the inner shell is sprayed with high temperature resistant insulating paint to form an insulating layer. The insulating layer is resistant to electrolyte corrosion for ≥1000h and has a breakdown voltage of ≥10kV. It can effectively isolate the battery cell from the shell and prevent short circuit faults caused by condensate or electrolyte leakage.

[0033] The thermal insulation pads installed between the cells of the battery module in the cold chain environment battery protection device of the present invention have a thermal conductivity of ≤0.03W / (m·K) and a flame retardant rating of UL94V-0. This not only reduces heat transfer between cells and inhibits the spread of thermal runaway, but also further improves the insulation protection effect.

[0034] The battery protection device in the cold chain environment of the present invention is also equipped with a shock-absorbing buffer layer. It is connected to the vehicle chassis by a honeycomb structure and rubber support, and bolts. This structure can effectively absorb impact acceleration of ≥30g, while isolating medium and high frequency vibrations of 15Hz to 20Hz. It is suitable for the vibration and impact generated during frequent start-stop, turning and loading and unloading of goods in cold chain operations, avoids loosening of battery cells and breakage of connecting pieces, and ensures the integrity of battery structure and connection stability.

[0035] In the method of this invention, managers can remotely view the battery's working status via computer or mobile APP, and remotely set target thresholds such as temperature and humidity. When faults such as persistently excessive humidity, excessively high water level, abnormal battery temperature, or decreased insulation performance occur, the cloud platform immediately sends alarm information, facilitating timely response and maintenance by staff and reducing operation and maintenance costs.

[0036] This invention effectively solves the problems of rapid battery capacity decay, charging difficulties, condensation short circuits, and thermal runaway propagation in cold chain operations. It is suitable for electric operation in cold chain warehousing, cold chain transportation, and other scenarios, and has good application prospects. Attached Figure Description

[0037] Figure 1 This is a cross-sectional structural schematic diagram of the intelligent battery control method in a cold chain environment according to an embodiment of the present invention; The names and numbers of the parts in the diagram are as follows: 1-Shock-resistant buffer layer, 2-Outer shell, 3-Inner shell, 4-Microchannel liquid cooling plate, 5-Battery cell module, 6-Insulation support, 7-Nano aerogel layer, 8-Positive pressure relief valve, 9-Honeycomb aluminum layer, 10-Rubber support, 11-PTC heating film, 12-Graphene heat spreader film. Detailed Implementation

[0038] The present invention will now be described in detail with reference to the accompanying drawings and embodiments.

[0039] Example 1 like Figure 1 As shown, the intelligent battery control method in the cold chain environment applies a battery protection device in the cold chain environment. The battery protection device in the cold chain environment includes a shock-absorbing buffer layer 1, an outer shell 2, an inner shell 3, a heating film group, a microchannel liquid cooling plate 4, a cell module 5, a controller, and a remote monitoring module. An anti-vibration buffer layer 1 is provided on the bottom surface of the outer shell 2; a positive pressure relief valve 8 is provided on the top of the outer shell 2.

[0040] The inner bottom surface of the outer shell 2 is provided with multiple heat-insulating supports 6, which are made of ceramic material; the outer bottom surface of the inner shell 3 is fixedly installed on these heat-insulating supports 6, and a gap is left between the outer wall of the inner shell 3 and the inner wall of the outer shell 2, which is filled with nano-aerogel to form a nano-aerogel layer 7; the thickness of the nano-aerogel layer 7 is 5mm.

[0041] A heating film assembly is provided on the inner bottom surface of the inner shell 3, and a microchannel liquid cooling plate 4 is provided on the heating film assembly. The large surface of the battery cell module 5 is mounted on the microchannel liquid cooling plate 4. A temperature sensor and a humidity sensor are provided on the battery cell module 5.

[0042] The outer shell 2 is made of aluminum alloy, and its outer surface is coated with a fire-retardant coating, which is a 0.3mm thick ceramicized silicone rubber. The inner shell 3 is made of stainless steel, and its inner wall is coated with a high-temperature resistant insulating varnish.

[0043] The nano-aerogel is a SiO2 aerogel felt with a thickness of 5 mm, a thermal conductivity of ≤0.018 W / (m·K) at 25℃, and no cracks when folded 180° at -30℃.

[0044] The outlet of the positive pressure relief valve 8 is covered with a hydrophobic and breathable membrane with a pore size of 0.1 μm, and the opening pressure of the positive pressure relief valve 8 is 4 kPa ± 0.5 kPa.

[0045] The shock-absorbing buffer layer includes a honeycomb aluminum layer 9 and rubber pillars 10. The honeycomb aluminum layer 9 is disposed on the bottom surface of the outer shell 2. Multiple sets of rubber pillars 10 are evenly arranged on the bottom surface of the honeycomb aluminum layer 9 and are installed on the vehicle chassis through the rubber pillars 10.

[0046] The heating film assembly includes a PTC heating film 11 and a graphene heat-spreading film 12. The PTC heating film 11 is disposed on the bottom inner surface of the outer shell 2, and the graphene heat-spreading film 12 is disposed on the top surface of the PTC heating film 11. The graphene heat-spreading film 12 is in contact with the microchannel liquid cooling plate 4. The thickness of the graphene heat-spreading film 12 is 0.5 mm.

[0047] The controller is connected to the temperature sensor, humidity sensor, and BMS of the battery cell module 5. The controller uploads monitoring data and control status information to the cloud platform in real time through the Internet of Things, issues alarms according to the alarm level, and accepts remote parameter commands for control.

[0048] The intelligent control method includes the following steps: A. Sensor data acquisition: Real-time acquisition of ambient temperature T, ambient humidity H, condensate water level W, battery voltage U, and battery temperature Tb through temperature sensor, humidity sensor, water level sensor, and battery management system (BMS). B. Data preprocessing: Filtering and denoising the collected data, removing outliers, and obtaining effective monitoring data. Data preprocessing uses the Kalman filter algorithm to remove outliers caused by sensor errors and environmental interference, and then normalization is used to unify the scale of data from different dimensions to obtain accurate and effective monitoring data. The multi-objective optimization function for Kalman filtering denoising is as follows: ;

[0049] in, , , These are weighting coefficients, typically This indicates that temperature control has the highest priority.

[0050] C. Set preset threshold: Temperature threshold: T heat,on =5℃; T heat,off =15℃; T fan,on =28℃; T safe =45℃; Humidity threshold: R Hon =65%; R Hoff =55%; R Hdanger =85%; SOC threshold: SOC min =5%; SOC heat,max =20%, maximum heating power limit at low SOC; Insulation resistance threshold: R iso,min =1.0 MΩ; D. Establish a fuzzy rule base, in which temperature control function, humidity control function, water level control calculation function, and BMS system enable function are defined respectively. Ambient temperature T, ambient humidity H, water level and alarm level, and BMS status are input into the corresponding functions, respectively, and the output variables are heating power, dehumidifier operating level, ventilation solenoid valve status, drain pump status, and alarm level. These are controlled by a controller, which also provides status feedback. An adaptive neural fuzzy control algorithm is used to correct and optimize the above functions in real time, achieving nonlinear dynamic regulation. Specific variable outputs are as follows: P h ∈[0,1], heater power factor 0-100%; P d ∈[0,1], dehumidifier power factor; F v ∈{0,1}, Ventilation solenoid valve status: 0 closed, 1 open; F p ∈{0,1}, Drainage pump status; A∈{0,1,2}, Alarm level: 0 None, 1 Warning, 2 Critical.

[0051] The temperature control function Specifically as follows: It is divided into high-temperature ventilation mode and low-temperature heating mode: High-temperature ventilation mode: (1); Low temperature heating mode: (2); Gradient adjustment rules: (3); Among them, T b T represents the current temperature of the battery. heat,on The threshold temperature for heating to start; T heat,off The heating shut-off threshold temperature; T target Target temperature; F cool This refers to the fan speed setting in high-temperature ventilation mode; P h This represents the percentage of heating power in low-temperature heating mode; ΔP h The heating power gradient adjustment amount; T is the cold chain ambient temperature; SOC is the remaining battery charge; R i This refers to the battery's internal resistance. The core decision-making logic of the adaptive neural fuzzy control algorithm is as follows: a. Temperature control: The preset ambient low temperature threshold is 0℃, the target temperature range for discharge is 5-10℃, and the target temperature range for charging is 10-15℃. When T < 0℃, the PTC heating film is activated, and the heating power is dynamically adjusted according to the difference between T and the target temperature of 5℃ until the temperature reaches the corresponding target range. When the battery core temperature Tb > 25℃ is detected, the heating power is reduced or the heating film is turned off. At the same time, the positive pressure relief valve is used to accelerate the air circulation inside and outside the casing to enhance the heat dissipation effect. b. Humidity Control: The preset high humidity threshold is 60%RH, and the target humidity range is 30%-50%RH. When H>60%RH is detected, the semiconductor dehumidifier is activated, and the operating level is adjusted according to the degree of humidity exceeding the standard until the humidity drops to the target range, avoiding excessive dehumidification that could cause battery components to dry out and crack.

[0052] c. Water level control: The condensate water level threshold is preset, and a safe upper limit is set according to the capacity of the collection box. When W≥ this threshold is detected, the controller triggers an audible and visual drainage reminder, controls the opening of the drainage pump and the solenoid valve of the water pipe, and closes the solenoid valve and drainage pump after the water level drops to the safe lower limit.

[0053] The humidity control function Specifically as follows: Dehumidifier power control: (4); Where RH represents the current ambient relative humidity; P d R represents the percentage of the dehumidifier's output power. Hdanger Hazardous humidity threshold; RH on Set the humidity threshold for the dehumidifier; RH off The humidity threshold for turning off the dehumidifier.

[0054] Enhanced anti-condensation control: (5); Among them, the safety threshold Condensation control gain ;P h,add To prevent condensation, the auxiliary heating power is increased; T dew This is the dew point temperature.

[0055] Natural ventilation and dehumidification conditions: (6); Among them, the minimum humidity difference threshold is RH out The relative humidity of the external environment; T out The external ambient temperature.

[0056] The aforementioned anti-condensation enhancement control is activated based on anti-condensation judgment; the anti-condensation judgment is calculated using the dew point temperature calculation formula, and the object surface temperature is below T. dew Condensation will occur at this time, and the specific formula is as follows: (7); in: And parameters: T e RH represents ambient temperature and relative humidity.

[0057] The calculation formula in the water level and alarm level calculation module is as follows: Water level control algorithm: (8); Alarm level: (9); Where W is the water level status flag, h w F represents the water level. p For drainage pump control flags; A water This is the flood alarm level.

[0058] The BMS system enable function is defined as follows: System enable function : (10); in, F represents logical AND; bms For BMS fault flags; T b,max This is the highest battery temperature; T safe SOC refers to the battery's safe temperature threshold; SOC refers to the remaining battery capacity. min R is the minimum permissible SOC threshold for the battery. iso R is the battery insulation resistance. iso,min This is the minimum permissible insulation resistance threshold.

[0059] Considering safety interlocking, the final output of all actuators is: ;

[0060] ;

[0061] (11); in, This is an indicator function; it returns 1 if the condition is met, and 0 otherwise. P represents the final output value of the heating power. h This is the original heating power control value; This represents the final output power of the dehumidifier. The final output value of the natural ventilation control flag; F v This is the original natural ventilation control marker position.

[0062] When the BMS reports a critical fault: if or Then we have: .

[0063] The adaptive neural fuzzy control algorithm process is as follows: a. Input / output definition: Input layer (5-dimensional): ;

[0064] Output layer (5-dimensional): ;

[0065] Among them, RH avg P represents the average relative humidity of the environment. h P is the heating power. d For dehumidification power, F v Ventilation status, F p In drainage state, A l Alarm level; b. Perform fuzzy rules and reasoning The j-th fuzzy rule (j=1,2,...,10): ;

[0066] in For fuzzy subsets (such as "low temperature" and "medium humidity"), For fuzzy rule consequent parameters; c. Self-learning parameter correction Error function: (12) Where M is the total number of training samples; k is the sample iteration index; Y represents the target true value for the k-th sample. k This is the model prediction value for the k-th sample. Parameter update (gradient descent): (13); Where η is the learning rate. This is the partial derivative of the error with respect to the parameter.

[0067] Iterate every 24 hours to correct the fuzzy rule parameters and minimize the error E.

[0068] The battery intelligent control method for cold chain environments also includes a remote monitoring module. The controller uploads monitoring data and control status information to the cloud platform in real time via the Internet of Things, issues alarms according to the alarm level, and receives remote parameter commands for control.

Claims

1. A method for intelligent battery control in a cold chain environment, characterized in that, Includes the following steps: A. Sensor data acquisition: Real-time acquisition of ambient temperature T, ambient humidity H, condensate water level W, battery voltage U, and battery temperature Tb through temperature sensor, humidity sensor, water level sensor, and battery management system (BMS). B. Data preprocessing: Filtering and denoising the collected data, removing outliers, and obtaining effective monitoring data. C. Set preset threshold Temperature threshold: T heat,on =5℃; T heat,off =15℃; T fan,on =28℃; T safe =45℃; Humidity threshold: R Hon =65%; R Hoff =55%; R Hdanger =85%; SOC threshold: SOC min =5%; SOC heat,max =20%, maximum heating power limit at low SOC; Insulation resistance threshold: R iso,min =1.0 MΩ; D. Establish a fuzzy rule base, in which temperature control function, humidity control function, water level control calculation function, and BMS system enable function are defined respectively. Ambient temperature T, ambient humidity H, water level and alarm level, and BMS status are input into the corresponding functions, respectively, and the output variables are heating power, dehumidifier operating level, ventilation solenoid valve status, drain pump status, and alarm level. These are controlled by a controller, which also provides status feedback. An adaptive neural fuzzy control algorithm is used to correct and optimize the above functions in real time, achieving nonlinear dynamic regulation. Specific variable outputs are as follows: P h ∈[0,1], heater power factor 0-100%; P d ∈[0,1], dehumidifier power factor; F v ∈{0,1}, Ventilation solenoid valve status: 0 closed, 1 open; F p ∈{0,1}, Drainage pump status; A∈{0,1,2}, Alarm level: 0 None, 1 Warning, 2 Critical.

2. The intelligent battery control method under cold chain conditions as described in claim 1, characterized in that: In step D, the core decision-making logic of the adaptive neural fuzzy control algorithm is as follows: a. Temperature control: The preset ambient low temperature threshold is 0℃, the target temperature range for discharge is 5-10℃, and the target temperature range for charging is 10-15℃. When T < 0℃, the PTC heating film is activated, and the heating power is dynamically adjusted according to the difference between T and the target temperature of 5℃ until the temperature reaches the corresponding target range. When the battery core temperature Tb > 25℃ is detected, the heating power is reduced or the heating film is turned off. At the same time, the positive pressure relief valve is used to accelerate the air circulation inside and outside the casing to enhance the heat dissipation effect. b. Humidity Control: The preset high humidity threshold is 60%RH, and the target humidity range is 30%-50%RH. When H>60%RH is detected, the semiconductor dehumidifier is activated, and the operating level is adjusted according to the degree of humidity exceeding the standard until the humidity drops to the target range, to avoid excessive dehumidification that could cause battery components to dry out and crack. c. Water level control: The condensate water level threshold is preset, and a safe upper limit is set according to the capacity of the collection box. When W≥ this threshold is detected, the controller triggers an audible and visual drainage reminder, controls the opening of the drainage pump and the solenoid valve of the water pipe, and closes the solenoid valve and drainage pump after the water level drops to the safe lower limit.

3. The intelligent battery control method under cold chain conditions as described in claim 1, characterized in that: The temperature control function Specifically as follows: It is divided into high-temperature ventilation mode and low-temperature heating mode: High-temperature ventilation mode: (1); Low temperature heating mode: (2); Gradient adjustment rules: (3); Among them, T b T represents the current temperature of the battery. heat,on The threshold temperature for heating to start; T heat,off The heating shut-off threshold temperature; T target Target temperature; F cool This refers to the fan speed setting in high-temperature ventilation mode; P h This represents the percentage of heating power in low-temperature heating mode; ΔP h The heating power gradient adjustment amount; T is the cold chain ambient temperature; SOC is the remaining battery charge; R i This refers to the battery's internal resistance. The humidity control function Specifically as follows: Dehumidifier power control: (4); Where RH represents the current ambient relative humidity; P d R represents the percentage of the dehumidifier's output power. Hdanger Hazardous humidity threshold; RH on Set the humidity threshold for the dehumidifier; RH off The dehumidifier is set to the humidity threshold for shutdown. Enhanced anti-condensation control: (5); Among them, the safety threshold Condensation control gain ;P h,add To prevent condensation, the auxiliary heating power is increased; T dew This refers to the dew point temperature. Natural ventilation and dehumidification conditions: (6); Among them, the minimum humidity difference threshold is RH out The relative humidity of the external environment; T out External ambient temperature; The aforementioned anti-condensation enhancement control is activated based on anti-condensation judgment; the anti-condensation judgment is calculated using the dew point temperature calculation formula, and the object surface temperature is below T. dew Condensation will occur at this time, and the specific formula is as follows: （7）； in: And parameters: T e Here, RH represents ambient temperature and relative humidity. The calculation formula in the water level and alarm level calculation module is as follows: Water level control algorithm: (8); Alarm level: (9); Where W is the water level status flag, h w F represents the water level. p For drainage pump control flags; A water The water accumulation alarm level; The BMS system enable function is defined as follows: System enable function : （10）； in, F represents logical AND; bms For BMS fault flags; T b,max This is the highest battery temperature; T safe SOC refers to the battery's safe temperature threshold; SOC refers to the remaining battery capacity. min R is the minimum permissible SOC threshold for the battery. iso R is the battery insulation resistance. iso,min The minimum permissible insulation resistance threshold; Considering safety interlocking, the final output of all actuators is: ； ； （11）； in, This is an indicator function; it returns 1 if the condition is met, and 0 otherwise. P represents the final output value of the heating power. h This is the original heating power control value; This represents the final output power of the dehumidifier. The final output value of the natural ventilation control flag; F v This is the original natural ventilation control sign position; When the BMS reports a critical fault: if or Then we have: 。 4. The intelligent battery control method under cold chain conditions as described in claim 1, characterized in that: The adaptive neural fuzzy control algorithm process is as follows: a. Input / output definition: Input layer (5-dimensional): ; Output layer (5-dimensional): ; Among them, RH avg P represents the average relative humidity of the environment. h P is the heating power. d For dehumidification power, F v Ventilation status, F p In drainage state, A l Alarm level; b. Perform fuzzy rules and reasoning The j-th fuzzy rule (j=1,2,...,10): ； in For fuzzy subsets (such as "low temperature" and "medium humidity"), For fuzzy rule consequent parameters; c. Self-learning parameter correction Error function: （12）； Where M is the total number of training samples; k is the sample iteration index; Y represents the target true value for the k-th sample. k This is the model prediction value for the k-th sample. Parameter update (gradient descent): （13）； Where η is the learning rate. This is the partial derivative of the error with respect to the parameter; Iterate every 24 hours to correct the fuzzy rule parameters and minimize the error E.

5. The intelligent battery control method under cold chain conditions as described in claim 1, characterized in that: In step B, the data preprocessing uses the Kalman filter algorithm to remove outliers caused by sensor errors and environmental interference, and then normalization is used to unify the scale of data from different dimensions to obtain accurate and effective monitoring data. The multi-objective optimization function for Kalman filtering denoising is as follows: ； in, , , These are weighting coefficients, typically This indicates that temperature control has the highest priority.

6. The intelligent battery control method under cold chain conditions as described in claim 1, characterized in that, The following battery protection device with a cold chain environment structure was applied: The battery protection device in the cold chain environment includes a shock-absorbing buffer layer (1), an outer shell (2), an inner shell (3), a heating film assembly, a microchannel liquid cooling plate (4), a controller, and a remote monitoring module; Both the outer shell (2) and the inner shell (3) are sealed shells; An anti-vibration buffer layer (1) is provided on the bottom surface of the outer shell (2); The inner bottom surface of the outer shell (2) is provided with multiple heat-insulating supports (6), and the outer bottom surface of the inner shell (3) is fixedly installed on these heat-insulating supports (6). There is a gap between the outer wall of the inner shell (3) and the inner wall of the outer shell (2), which is filled with nano aerogel to form a nano aerogel layer (7). A heating film assembly is provided on the inner bottom surface of the inner shell (3), and a microchannel liquid cooling plate (4) is provided on the heating film assembly. A battery cell module (5) is installed on the top surface of the microchannel liquid cooling plate (4). At this time, the large surface of the battery cell module (5) is in contact with the microchannel liquid cooling plate (4). The top of the outer casing (2) is provided with a positive pressure relief valve (8); the battery cell module (5) is provided with a temperature sensor and a humidity sensor; The controller is connected to the temperature sensor, humidity sensor, and BMS of the battery cell module (5); The controller uploads monitoring data and control status information to the cloud platform in real time via the Internet of Things, issues alarms according to the alarm level, and accepts remote parameter commands for control.

7. The intelligent battery control method under cold chain conditions as described in claim 1, characterized in that: The outer shell (2) is made of aluminum alloy and its outer surface is provided with a fireproof coating; the inner shell (3) is made of stainless steel and its inner wall is coated with high temperature resistant insulating paint. Preferably, the fire-retardant coating is a 0.2-0.4mm ceramicized silicone rubber; Preferably, the nano-aerogel is a SiO2 aerogel felt with a thickness of 4mm to 6mm, a thermal conductivity of ≤0.018 W / (m·K) at 25℃, and no cracks when folded 180° at -30℃; Preferably, the outlet of the positive pressure relief valve (8) is covered with a hydrophobic and breathable membrane with a pore size of 0.08-0.12μm, and the opening pressure of the positive pressure relief valve (8) is 4kPa±0.5kPa.

8. The intelligent battery control method under cold chain conditions as described in claim 1, characterized in that: The shock-absorbing buffer layer includes a honeycomb aluminum layer (9) and rubber pillars (10). The honeycomb aluminum layer (9) is disposed on the bottom surface of the outer shell (2). Multiple sets of rubber pillars (10) are evenly arranged on the bottom surface of the honeycomb aluminum layer (9) and are installed on the vehicle chassis through the rubber pillars (10).

9. The intelligent battery control method under cold chain conditions as described in claim 1, characterized in that: The heating film assembly includes a PTC heating film (11) and a graphene heat-spreading film (12). The PTC heating film (11) is disposed on the bottom surface of the inner shell (2), and the graphene heat-spreading film (12) is disposed on the top surface of the PTC heating film (11). The graphene heat-spreading film (12) is in contact with the microchannel liquid cooling plate (4). Preferably, the thickness of the graphene heat spreader (12) is 0.4-0.6 mm.

10. The intelligent battery control method under cold chain conditions as described in claim 1, characterized in that: The heat insulation support (6) is made of ceramic material; preferably, the battery module (5) includes multiple battery cells, and a heat insulation pad with a thickness of 1-2 mm is provided between each battery cell. The heat insulation pad has a double-layer structure, consisting of a ceramicized silicone rubber layer and a SiO2 aerogel felt layer.

Images