A multi-working-condition temperature control method and system for lithium batteries of shopping carts in a low-temperature environment

Abstract

The application relates to the technical field of battery thermal management, and discloses a multi-working-condition temperature control method and system for shopping cart lithium batteries in a low-temperature environment, which comprises the following steps: acquiring battery multi-measurement-point temperature, physical docking signals and three-axis acceleration data, generating a motion anti-shake flag bit through variance feature extraction, determining a homing charging, pushing use or parking standby state in combination with the docking signals, calculating a heat compensation power in the pushing use state by substituting an air inlet and outlet temperature difference and internal heat production into a heat balance equation; introducing a heat dissipation integral factor accumulated in a dormant period as a feedforward control to drive preheating in the homing charging state; calculating a deep power shortage bottom line power in combination with an environment temperature drop in the parking standby state and executing a ladder dormancy, and the system integrates multi-modal sensors and hardware modules, a power supply link is dispatched by a main control unit, and a heating control signal is output. The application carries out differential heat compensation for multiple working conditions, shortens a cold start time and guarantees battery thermal safety.

Description

Technical Field

[0001] This invention relates to the field of battery thermal management technology, specifically to a multi-condition temperature control method and system for lithium batteries in shopping carts under low-temperature environments. Background Technology

[0002] Low temperatures can cause a decrease in the rate of chemical reactions inside lithium batteries, an increase in internal resistance, and a reduction in the charging and discharging platform voltage, which severely limits the output performance and lifespan of the batteries. In order to maintain the working activity of the batteries at low temperatures, most existing intelligent shopping cart lithium battery temperature control systems use fixed temperature thresholds to perform basic heating start-stop actions. However, in actual application scenarios, intelligent shopping carts face different physical conditions such as returning to charging, manual pushing, and long-term parking. The external convection environment and heat dissipation characteristics of the batteries differ under different conditions.

[0003] Conventional temperature control solutions do not establish a dynamic adjustment mechanism deeply coupled with physical operating conditions, and do not incorporate spatial motion posture and windward air convection factors for comprehensive quantitative evaluation. When the shopping cart is being pushed and moved, the surface air cooling effect will take away a lot of heat. The fixed power heating strategy that relies on a single closed-loop logic results in insufficient heat compensation, causing a sudden drop in the internal temperature of the battery. When the shopping cart is in a long-term parked state, blindly maintaining the basic temperature scale and performing high-frequency heating will excessively consume the battery's own power and reduce the device's usable battery life.

[0004] Furthermore, existing battery thermal management logic has limitations in handling long-term low-power parking and cold start scenarios. During the low-temperature sleep period, the shopping cart lacks a long-term cumulative assessment mechanism for environmental heat loss. When the device is connected to an external power source and enters the charging state, the system can only rely on the basic proportional-integral-differential algorithm for slow temperature ramp-up adjustment, lacking feedforward compensation parameters for early heat loss. This results in excessively long cold start preheating time, severely affecting charging efficiency. At the same time, the existing discharge protection strategy uses an absolutely constant value as the bottom line for power depletion protection without dynamic correction based on changes in external ambient temperature. When the battery discharge voltage drops significantly in extremely cold environments, the fixed power protection threshold cannot trigger the deep sleep power-off protection action in time, ultimately leading to irreversible physical damage to the battery due to deep over-discharge. Summary of the Invention

[0005] To address the shortcomings of existing technologies, this invention provides a multi-condition temperature control method and system for lithium batteries in shopping carts under low-temperature environments. This solves the problems of abnormal power consumption caused by the lack of dynamic temperature control combining motion and air cooling effects, slow cold start due to lack of heat dissipation feedforward compensation, and battery over-discharge damage caused by a fixed discharge baseline under extremely cold conditions.

[0006] To achieve the above objectives, the present invention provides the following technical solution: In a first aspect, the present invention provides a multi-condition temperature control method for a lithium battery in a shopping cart under low-temperature conditions, the method comprising the following steps: Configure the temperature control threshold matrix and power protection reference parameters, and write the characteristic coefficients for dynamic thermal balance calculation; The system acquires data on battery core temperature, external ambient temperature, battery compartment air inlet and outlet temperatures, physical docking signal, external power supply valid signal, output current, remaining power, and spatial triaxial acceleration matrix. Based on this data, it extracts the low-temperature heat dissipation integral factor and calculates the minimum power level for deep power depletion protection. It also performs sliding time window filtering on the battery compartment air inlet and outlet temperatures to extract smooth air inlet and outlet temperature difference feature values, and performs variance feature extraction on the spatial triaxial acceleration matrix data to generate motion stabilization flags. Based on the physical docking signal and motion anti-shake flag, the system determines the charging status, the pushing and using status, or the parking and standby status. In the parking and standby status, the system controls the heating module to perform intermittent heat preservation. In the charging status, the system calculates the target preheating power to drive the heating module by combining the low temperature heat dissipation integral factor. In the pushing and using status, the system substitutes the smooth inlet and outlet air temperature difference characteristic value into the dynamic heat balance calculation formula to obtain the target control power and outputs a dynamic heat replenishment command. Monitor the temperature at all measuring points. If the temperature at any measuring point exceeds the maximum safe temperature, cut off the heating execution circuit and output a high temperature fault alarm signal. If the standby state continues for more than a preset time threshold, the hibernation wake-up mechanism will be activated. At the periodic wake-up time node, the accumulated data during the hibernation period will be calculated and supplemented into the low temperature heat dissipation integral factor. When the remaining power is lower than the deep power loss protection bottom line, the heating control channel will be locked and the deep hibernation power-off protection state will be entered.

[0007] This method introduces a spatial triaxial acceleration matrix for variance calculation to identify the physical motion state of the smart shopping cart, and combines the temperature difference data of the battery compartment air inlet and outlet to quantify the wind-cooling effect during movement. Differentiated working logic is established for different operating conditions. During implementation, dynamic thermal balance calculation is performed by combining battery discharge Joule heat and windward heat loss. When parked for a long time, the heat dissipation factor is extracted using a temperature integral algorithm to achieve dynamic thermal management under multiple operating conditions and ensure the safety of the battery discharge limit.

[0008] Furthermore, the steps for configuring the temperature control threshold matrix and the power protection benchmark parameters include: writing the preheating start temperature, the discharge insulation lower limit temperature, the heating stop temperature, and the maximum safe temperature, and combining them into a temperature control threshold matrix; writing the basic reserve power and the environmental compensation coefficient to establish the power protection benchmark parameters, and constructing a mathematical model for the deep power loss protection baseline power based on the basic reserve power and the environmental compensation coefficient; introducing the external ambient temperature as an independent variable to perform slope compensation on the basic reserve power, and dynamically raising the deep power loss protection baseline power when the external ambient temperature drops. The solution logic for the deep power loss protection baseline power is expressed as a linear superposition of the basic reserve power and the environmental compensation correction term.

[0009] By dynamically adjusting the discharge baseline based on changes in external ambient temperature, the deep discharge protection baseline is raised when the ambient temperature drops, thus preventing excessive voltage drop in the battery discharge voltage caused by extreme cold environments, which could lead to power supply abnormalities.

[0010] Furthermore, the steps for writing the dynamic thermal balance calculation characteristic coefficients include: the dynamic thermal balance calculation characteristic coefficients include the internal Joule thermal conversion coefficient, the air-cooled convection compensation coefficient, and the heat dissipation attenuation coefficient; configuring the internal Joule thermal conversion coefficient to establish the heat offset mapping relationship generated by the battery internal resistance heating under the promotion and use state; configuring the air-cooled convection compensation coefficient to quantify the surface air convection heat loss during the movement of the smart shopping cart; configuring the heat dissipation attenuation coefficient to reconstruct the feedforward heating power compensation benchmark when the parking standby state transitions to the charging state.

[0011] By parameterizing the effects of internal battery heating and environmental heat dissipation as independent characteristic coefficients, the Joule heat compensation and air cooling heat dissipation are quantified, providing computational weights for the multivariable collaborative thermal balance equation.

[0012] Furthermore, the steps for performing variance feature extraction on the spatial triaxial acceleration matrix data to generate motion stabilization flags include: constructing a variance calculation window according to a continuously preset number of samples; performing variance feature extraction on the spatial triaxial acceleration matrix data in the discrete time dimension; evaluating the intensity of pose oscillation of the smart shopping cart in three-dimensional physical space to obtain the variance feature extraction result; comparing the variance feature extraction result with pre-configured motion entry variance threshold and motion exit variance threshold to generate motion stabilization flags used to isolate mechanical vibration interference signals; and calculating the stabilization gating coefficient based on the variance feature extraction result, with the value of the stabilization gating coefficient smoothly transitioning between zero and one.

[0013] A sliding window approach combined with variance calculation is introduced to process the underlying acceleration data, isolate transient interference caused by physical road bumps, extract the continuous propulsion motion state, and output the anti-shake gating coefficient.

[0014] Furthermore, the steps for determining the charging state, pushing and using state, or parking and standby state based on the physical docking signal and the motion stabilization flag include: when the physical docking signal is valid and the external power supply valid signal is valid, the smart shopping cart is determined to be in the charging state; when the physical docking signal is invalid and the motion stabilization flag is logic 1, the smart shopping cart is determined to be in the pushing and using state; when the physical docking signal is invalid and the motion stabilization flag is logic 0, the smart shopping cart is determined to be in the parking and standby state; when the physical docking signal is valid but the external power supply valid signal is invalid, the smart shopping cart is determined to be in the parking and standby state, and an external power supply abnormality flag is generated.

[0015] By combining hardware docking signals, power signals, and extracted motion stabilization flags, a combined judgment logic is constructed to distinguish typical usage scenarios and abnormal access conditions of smart shopping carts, providing a basis for judgment on temperature control strategy switching.

[0016] Furthermore, the step of calculating the target preheating power to drive the heating module in the homing charging state by combining the low-temperature heat dissipation integral factor includes: introducing the low-temperature heat dissipation integral factor as a feedforward compensation component into the closed-loop feedback regulation loop, and calculating the target preheating power by combining the low-temperature heat dissipation integral factor with the proportional-integral-derivative control equation; the specific calculation logic of the proportional-integral-derivative control equation is: calculating the basic regulation amount by combining the temperature deviation value, proportional coefficient, integral coefficient and derivative coefficient at the current moment, and multiplying the low-temperature heat dissipation integral factor by the heat dissipation compensation weight coefficient and then adding it to the basic regulation amount; The current allowable preheating power is determined based on the rated maximum power of the heating module and the rated output power of the external power supply. The target preheating power is then constrained by upper and lower limits to obtain the final target preheating power. A pulse width modulation signal is output based on the final target preheating power to drive the heating module to work continuously until the battery core temperature reaches the heating stop temperature in the temperature control threshold matrix.

[0017] The accumulated heat dissipation data during the dormancy period is used as a feedforward control variable to participate in the closed-loop regulation, providing power compensation output in the initial stage of charging with external power supply, and shortening the preheating time in the cold start stage.

[0018] Furthermore, the steps of substituting the characteristic value of the smooth inlet and outlet air temperature difference into the dynamic heat balance calculation formula to obtain the target control power and output the dynamic heat replenishment command in the implementation and use state include: obtaining the calibrated basic heat demand power to maintain the basic temperature scale, and calculating the deduction value of the self-heating generated during the battery discharge process by combining the internal Joule heat conversion coefficient and the output current in the dynamic heat balance calculation characteristic coefficient. The wind-cooled convection compensation coefficient in the dynamic heat balance calculation characteristic coefficient, the anti-shake gating coefficient derived from the variance feature extraction operation, and the smooth inlet and outlet air temperature difference characteristic value are combined to calculate the windward heat loss value during the movement of the smart shopping cart; the target control power is generated by subtracting the deduction value from the calibrated basic heat demand power and adding the windward heat loss value; the target control power is limited according to the current allowable heating power to obtain the final control power, and a dynamic heat replenishment command is output based on the final control power.

[0019] A dynamic thermal compensation model is introduced, which takes the heat generated by the battery itself during charging and discharging as a heat deduction item and the wind-cooling convection caused by external movement as a heat loss increase item, and adjusts the battery self-powered heating power under the promoted use state.

[0020] Furthermore, the steps of monitoring the global measurement point temperature and cutting off the heating execution circuit and outputting a high temperature fault alarm signal when the temperature of any measurement point exceeds the maximum safe temperature include: configuring a high-priority hardware timer to independently execute the global temperature over-limit detection task, performing differential operation on the temperature data sequence based on the battery core temperature in combination with a fixed sampling time interval, and extracting the real-time temperature rise rate. When the real-time temperature rise rate is greater than or equal to the preset temperature rise rate limit threshold within a continuous preset confirmation period, or when the temperature at any measuring point is greater than or equal to the maximum safe temperature, the high-priority temperature protection interrupt logic is triggered; a power-off control signal is issued to physically cut off the heating execution circuit, and a high-temperature fault alarm signal is simultaneously output to the outside through the communication bus, and the battery core temperature and timestamp at the moment of exceeding the limit are solidified and written into the non-volatile storage area.

[0021] By employing dual boundary conditions of absolute temperature value and real-time temperature rise rate, and setting up an independent hardware timed detection channel to perform over-limit detection, the execution priority of battery thermal runaway protection actions and the power-off response speed are improved.

[0022] Furthermore, if the standby state continues for more than a preset time threshold, the hibernation / wake-up mechanism is activated. The steps for calculating the cumulative integral data during the hibernation period and supplementing it to the low-temperature heat dissipation integral factor at the periodic wake-up time node include: The stepped backoff wake-up cycle is determined based on the deviation between the remaining power and the deep power loss protection baseline power. Before entering the low-power sleep mode, the current low-temperature heat dissipation integral factor is saved as the historical low-temperature heat dissipation integral factor and the sleep start timestamp. When the cycle wake-up time node corresponding to the stepped backoff wake-up cycle is reached, the external ambient temperature integral value between adjacent cycle wake-up time nodes is calculated using the trapezoidal integral algorithm to obtain the integral accumulation data during the sleep period. The integral accumulation data is multiplied by the heat dissipation attenuation coefficient in the dynamic heat balance calculation characteristic coefficient to obtain the sleep period heat dissipation compensation amount. The sleep period heat dissipation compensation amount is added to the historical low-temperature heat dissipation integral factor to obtain the updated low-temperature heat dissipation integral factor.

[0023] The timing wake-up cycle is dynamically adjusted based on the remaining battery power to control static power consumption. The environmental heat dissipation during the sleep cycle is accumulated and calculated through a segmented integral algorithm to obtain the thermal impact parameters caused by the low temperature of the environment after long-term parking.

[0024] Secondly, the present invention also provides a multi-condition temperature control system for a shopping cart lithium battery under low-temperature conditions, the system comprising: The lithium battery module is located inside the battery compartment of the smart shopping cart chassis; A thermal insulation shell is installed inside the battery compartment and wraps around the outer periphery of the lithium battery module; The heating module is located inside the heat-insulating housing and is used to perform heat conduction in close contact with the surface of the lithium battery module housing. The power switching module is configured at the input hub of the power supply link. It is used to receive the grid power introduced by the external charging interface and the battery power fed back by the lithium battery module and perform power supply link switching, thereby outputting the target driving power to the heating module. The temperature acquisition module is configured at the system's data acquisition node to collect the battery core temperature, external ambient temperature, and battery compartment air inlet and outlet temperatures. The repositioning detection module, configured at the data acquisition node, is used to acquire physical docking signals; The motion sensing module, configured at the data acquisition node, is used to acquire spatial triaxial acceleration matrix data in real time; The BMS main control unit is connected to the temperature acquisition module, the homing detection module, the motion sensing module, the power switching module, and the heating module to acquire physical docking signals, spatial triaxial acceleration matrix data, and various temperature data. The BMS main control unit uses the spatial triaxial acceleration matrix data to generate a motion stabilization flag, and then combines the physical docking signal and the motion stabilization flag to perform a working status determination operation, classifying the smart shopping cart into a homing charging state, a pushing and using state, or a parked standby state. The BMS main control unit is also used to generate a target control power based on various temperature data, and send a pulse width modulation control signal to the heating module according to the target control power to perform dynamic constant temperature compensation operation.

[0025] This system integrates sensor nodes and hardware control modules to establish a hardware link from data acquisition to heat control. It uses acceleration data and inlet / outlet temperature difference data to evaluate motion posture and thermal convection. The BMS main control unit performs power supply link scheduling and heating power calculation to complete the temperature control of lithium battery modules in different application scenarios.

[0026] This invention provides a multi-condition temperature control method and system for lithium batteries in shopping carts operating at low temperatures. It offers the following advantages: 1. This invention generates motion anti-shake flags by performing variance calculations on spatial triaxial acceleration matrix data and extracts the characteristic values ​​of the smooth airflow temperature difference in the battery compartment. In the pushing operation state, the wind-cooled convective heat loss and the Joule heat generation inside the battery are substituted into the dynamic thermal balance calculation model. By parametrically evaluating the physical motion posture and the amount of heat dissipated in the wind, the state misjudgment caused by mechanical bumps on the road surface is eliminated, and the self-powered heating power is accurately controlled under the pushing operation state, reducing ineffective power loss.

[0027] 2. This invention introduces a segmented sleep-wake mechanism in the parked standby state, uses a trapezoidal integral algorithm to accumulate the integral value of the external ambient temperature during the sleep cycle, extracts the low-temperature heat dissipation integral factor, and uses the low-temperature heat dissipation integral factor as a feedforward control quantity to participate in the closed-loop calculation when the vehicle switches to the charging state. Combined with the magnitude of the drop in external ambient temperature, the deep discharge protection baseline capacity is dynamically increased, which shortens the cold start preheating time and prevents the battery from being over-discharged and damaged in extremely cold environments.

[0028] 3. This invention executes a global temperature over-limit detection task through an independent high-priority hardware timer, and extracts the real-time temperature rise rate by combining a fixed sampling time interval. It establishes a dual judgment boundary of absolute temperature value and temperature rise rate. When the temperature or temperature rise rate of any measuring point triggers the limit threshold, it directly triggers a hardware interrupt and physically cuts off the heating execution circuit. The hardware-level detection channel, independent of the main control software, improves the execution priority and power-off response speed of thermal runaway protection, ensuring the physical safety of the battery under extreme conditions. Attached Figure Description

[0029] Figure 1 This is a schematic diagram of the system timing interaction of the present invention; Figure 2 This is a schematic diagram of the method flow of the present invention; Figure 3 This is a schematic diagram of the three-dimensional fitting relationship for the calibration of the air-cooled convection compensation coefficient of the present invention; Figure 4 This is a schematic diagram illustrating the mapping relationship between the anti-shake gate coefficient and the motion anti-shake flag bit of the present invention; Figure 5 This is a schematic diagram of the multi-condition working state determination and temperature over-limit protection process of the present invention; Figure 6 This is a schematic diagram illustrating the variation characteristics of the stepped backoff wake-up cycle and the periodic wake-up power consumption of the present invention. Figure 7 This is a schematic diagram showing the effect comparison of various operating conditions constant temperature management application examples of the present invention in extremely cold environments. Detailed Implementation

[0030] The technical solutions in the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0031] See attached document Figure 1 This invention provides a multi-condition temperature control system for a shopping cart lithium battery in low-temperature environments. The system includes a lithium battery module, a thermal insulation shell, a heating module, a power switching module, a temperature acquisition module, a homing detection module, a motion sensing module, and a BMS main control unit.

[0032] The lithium battery module and the thermal insulation shell are disposed inside the battery compartment of the smart shopping cart chassis; the lithium battery module outputs working power and transmits the working power to the electronic devices and drive mechanism of the smart shopping cart; the thermal insulation shell is wrapped around the outer periphery of the lithium battery module; the thermal insulation shell uses the heat-insulating material filled inside to isolate the internal space from the external environment, thereby reducing the heat exchange intensity between the lithium battery module and the external environment and suppressing the loss of heat from the lithium battery module to the external environment.

[0033] The heating module and the power switching module are configured in the heating execution circuit of the system; the heating module is arranged inside the heat insulation shell and performs heat conduction operation in close contact with the surface of the lithium battery module shell; the power switching module is configured at the input hub of the power supply link; the power switching module receives the grid power introduced by the external charging interface and the battery power fed back by the lithium battery module; then the power switching module performs the power supply link switching operation according to the grid power access status, and then outputs the target driving power to the heating module.

[0034] The temperature acquisition module, the positioning detection module, and the motion sensing module are configured at the system's data acquisition nodes. The temperature acquisition module is located inside and outside the smart shopping cart. It collects core battery temperature data from the surface of the lithium battery module and ambient temperature data. Simultaneously, it collects flow field temperature data at the air inlet and outlet of the battery compartment. The positioning detection module is located at the bottom docking area of ​​the smart shopping cart. It acquires the physical docking signal between the smart shopping cart and the external charging terminal. The motion sensing module is installed inside the chassis. It collects the spatial three-axis acceleration matrix data of the smart shopping cart in real time.

[0035] The BMS main control unit is connected to the temperature acquisition module, the homing detection module, the motion sensing module, the power switching module, and the heating module to acquire physical docking signals, spatial triaxial acceleration matrix data, and various temperature data. The BMS main control unit performs sliding time window filtering on the flow field temperature data to extract smooth inlet and outlet air temperature difference feature values ​​that eliminate flow field disturbances. At the same time, it uses spatial triaxial acceleration matrix data to perform variance extraction and threshold comparison operations to generate motion anti-shake flags. Then, it combines the physical docking signal and motion anti-shake flags to perform working status determination operations, thereby classifying the smart shopping cart into homing charging state, pushing and using state, or parking and standby state.

[0036] The BMS main control unit sends a link selection command to the power switching module based on the working status determination result. At the same time, the BMS main control unit inputs the smooth inlet and outlet air temperature difference characteristic value, motion anti-shake flag, battery core temperature data, and remaining power data into the dynamic thermal balance calculation formula to perform mathematical calculations, and then generates the target control power. Subsequently, the BMS main control unit sends a pulse width modulation control signal to the heating module according to the target control power, and then performs dynamic constant temperature compensation operation. Finally, when the BMS main control unit determines that a long period of no interaction has been triggered, it executes dynamic sleep wake-up logic. At the same time, the BMS main control unit performs a power-off operation on the non-core power supply circuits of the system, and finally enters the low-power timed wake-up sampling and monitoring process.

[0037] See attached document Figure 2 This invention provides a method for constant temperature management of lithium batteries in smart shopping carts under low-temperature environments, the method comprising the following steps: S1, the BMS main control unit configures the temperature control threshold matrix and power protection reference parameters in the system memory; at the same time, the BMS main control unit writes the characteristic coefficients required for dynamic thermal balance calculation, thereby constructing the initial constant temperature management control reference and constant temperature control parameter set of the system.

[0038] S2, the temperature acquisition module simultaneously acquires the battery core temperature, the external ambient temperature, and the battery compartment air inlet and outlet temperatures; the positioning detection module acquires the physical docking signal, while the BMS main control unit reads the output current and remaining power; then the BMS main control unit performs a sliding time window filtering operation on the battery compartment air inlet and outlet temperatures to extract smooth air inlet and outlet temperature difference feature values; subsequently, the motion sensing module acquires spatial three-axis acceleration matrix data, and the BMS main control unit performs variance feature extraction operation on the spatial three-axis acceleration matrix data to generate motion stabilization flags.

[0039] S3, the BMS main control unit performs working status determination based on the physical docking signal and motion anti-shake flag, classifying the working status of the smart shopping cart into charging state, push-and-use state, or parking standby state; when it is determined to be in parking standby state and the low temperature condition is met, the BMS main control unit calculates the deep power depletion protection baseline power based on the ambient temperature, and performs integral calculation on the low temperature exposure time to extract the low temperature heat dissipation integral factor, and then controls the heating module to perform intermittent heat preservation when the power is sufficient; When the system is determined to be in the charging state and meets the low temperature conditions, the BMS main control unit controls the power switching module to connect the external power supply. Then, it calculates the target preheating power by combining the low temperature heat dissipation integral factor and the proportional-integral-differential control equation, and drives the heating module to work continuously until the upper temperature limit is reached. When the system is determined to be in the push-to-use state and meets the low temperature conditions, the BMS main control unit substitutes the output current, the motion anti-shake flag, and the smooth inlet and outlet air temperature difference characteristic value into the dynamic heat balance calculation formula to perform the solution calculation, thereby obtaining the target control power. Finally, the BMS main control unit outputs a dynamic heat replenishment command to the heating module based on the target control power.

[0040] S4, the BMS main control unit monitors the global measurement point temperature data in real time. When the BMS main control unit determines that the temperature data of any measurement point exceeds the maximum safe temperature, it triggers the hardware forced interrupt logic. Then, the BMS main control unit sends a power-off command to the power switching module and the heating module, thereby physically cutting off the heating execution circuit, and simultaneously outputs a high temperature fault alarm signal to the outside.

[0041] S5, when the BMS main control unit determines that the duration of the standby state exceeds a preset time threshold, it initiates the hibernation wake-up mechanism. The BMS main control unit calculates the stepped back-off wake-up cycle based on the deviation of the remaining power from the bottom line. Then, the BMS main control unit collects ambient temperature data at the cycle wake-up time node, and then uses the trapezoidal integral algorithm to calculate the integral accumulation data during the hibernation period, and merges the integral accumulation data into the low temperature heat dissipation integral factor. Finally, when the BMS main control unit determines that the remaining power is lower than the deep power loss protection bottom line power, it locks the heating control channel, thereby controlling the temperature control system to enter the deep hibernation power-off protection state.

[0042] See attached document Figure 3 In specific implementations, step S1 provided by the present invention may include the following steps: S101, the BMS main control unit is powered on and reset, and performs a hardware status self-test on the system memory. At the same time, the BMS main control unit divides the non-volatile storage area of ​​the system memory into an independent threshold configuration area and a dynamic parameter cache area. The construction process of the constant temperature management control benchmark relies on the physical mapping coordination between the static thermodynamic boundary and the dynamic evolution parameters. Then, the BMS main control unit writes the preheating start temperature, the discharge heat preservation lower limit temperature, the heating stop temperature and the maximum safe temperature into the threshold configuration area, and configures the combination of the preheating start temperature, the discharge heat preservation lower limit temperature, the heating stop temperature and the maximum safe temperature into a temperature control threshold matrix. In addition, the BMS main control unit establishes the size constraint relationship of each temperature threshold in the temperature control threshold matrix in the internal control logic. In this embodiment, the preheating start temperature is set between -5℃ and 0℃, the discharge insulation lower limit temperature is set between 0℃ and 5℃, the heating stop temperature is set between 10℃ and 15℃, and the maximum safe temperature can be set between 55℃ and 65℃. However, the specific setting of the maximum safe temperature shall not exceed the minimum value among the maximum allowable operating temperature specified by the cell manufacturer, the maximum allowable temperature of the heating module, the maximum allowable temperature of the insulation shell material, and the maximum allowable temperature of the internal connectors of the battery module.

[0043] The BMS main control unit uses a configured temperature control threshold matrix to define the safe operating thermodynamic boundary of the lithium battery module in low-temperature environments, thereby avoiding control logic deadlock during cross-condition switching and thermal runaway caused by continuous heating exceeding limits.

[0044] S102, the BMS main control unit writes the basic minimum charge level and environmental compensation coefficient into the threshold configuration area. The basic minimum charge level is set between 10% and 20%, and the environmental compensation coefficient is set between 0.5% / ℃ and 1.0% / ℃. Then, the BMS main control unit establishes the basic minimum charge level and environmental compensation coefficient as the power protection benchmark parameters. Based on the basic minimum charge level and environmental compensation coefficient in the power protection benchmark parameters, the BMS main control unit constructs a mathematical model of the deep power loss protection baseline power level. In extremely cold environments, combined with the physical characteristics of non-linear reduction in battery usable capacity, the basic minimum charge level is compensated for by introducing ambient temperature as an independent variable. Thus, the forced dormancy protection baseline of the lithium battery module is dynamically raised when the ambient temperature drops. The solution logic of the adaptive deep power loss protection baseline power level is represented by the linear superposition of the basic minimum charge level and the environmental compensation correction term. The specific formula for calculating the adaptive deep power loss protection baseline capacity is as follows: ; In the formula, To provide adaptive deep power depletion protection minimum power level; Basic minimum power supply; This is the environmental compensation coefficient; The system's preset nominal room temperature reference value is set between 20°C and 25°C. The current ambient temperature.

[0045] S103, the BMS main control unit writes the characteristic coefficients required for dynamic thermal balance calculation into the dynamic parameter buffer area. The characteristic coefficients required for dynamic thermal balance calculation include the internal Joule thermal conversion coefficient, the air-cooled convection compensation coefficient, and the heat dissipation attenuation coefficient. For the initial calibration of the internal Joule thermal conversion coefficient, the lithium battery module is placed under multiple preset ambient temperatures between -20℃ and 0℃ and remaining charge conditions between 20% and 80%, and at least two different constant discharge currents such as 0.5C and 1C are applied to the lithium battery module. The changes in terminal voltage, battery core temperature, and discharge duration under each constant discharge current condition are recorded. The internal Joule thermal conversion coefficient is obtained by fitting the relationship between the DC internal resistance of the lithium battery module, the square of the current, and the equivalent heat generation power. The internal Joule thermal conversion coefficient is set between 0.05 and 0.15.

[0046] For the calibration of the air-cooled convection compensation coefficient, the smart shopping cart was placed under multiple fixed pushing speeds between 0.5m / s and 2.0m / s or simulated wind speeds between 1m / s and 3m / s. Under the condition that the initial battery temperature, external ambient temperature, and basic heating power between 10W and 50W were basically the same, the characteristic value of the smooth inlet and outlet air temperature difference, the pushing speed, and the additional heating power required to maintain the core battery temperature were recorded. Based on the correspondence between the characteristic value of the smooth inlet and outlet air temperature difference and the additional heating power, linear fitting or piecewise fitting was performed to obtain the air-cooled convection compensation coefficient, and the air-cooled convection compensation coefficient was set between 1.2 and 2.5.

[0047] For the calibration of the heat dissipation attenuation coefficient, the lithium battery module is heated to a preset initial temperature between 10°C and 20°C, and then placed in multiple preset low-temperature environments between -20°C and 0°C to stop active heating. The battery core temperature, external ambient temperature, and natural cooling time are recorded. The parameters are fitted according to the decrease curve of the battery core temperature relative to the external ambient temperature to obtain the heat dissipation attenuation coefficient, which is set between 0.01 and 0.05.

[0048] The BMS main control unit synchronously records the values, dimensions, applicable temperature ranges, and applicable remaining power ranges of each characteristic coefficient in the system memory; when the real-time operating condition is between adjacent calibration intervals, linear interpolation is used to determine the currently used characteristic coefficients.

[0049] The BMS main control unit configures an internal Joule heat conversion coefficient to establish a heat offset mapping relationship for the battery's internal resistance heating during operation. Simultaneously, the BMS main control unit configures a wind-cooling convection compensation coefficient to quantify the surface air convection heat loss during the smart shopping cart's movement. Furthermore, the BMS main control unit configures a heat dissipation attenuation coefficient to reconstruct the feedforward heating power compensation benchmark when transitioning from standby to charging. Finally, the BMS main control unit completes the initial construction of the system's constant temperature management control benchmark and constant temperature control parameter set by writing the characteristic coefficients required for dynamic thermal balance calculation, the temperature control threshold matrix, and the power protection benchmark parameters.

[0050] See attached document Figure 4 In specific implementations, step S2 provided by the present invention may include the following steps: S201, the temperature acquisition module uses temperature sensor nodes arranged in different physical locations to simultaneously acquire the battery core temperature, the external ambient temperature, and the temperature of the battery compartment air inlet and outlet. Among them, the first temperature sensor for acquiring the battery core temperature is set on the surface of a representative cell in the central area inside the lithium battery module or between adjacent cells to approximately characterize the temperature of the core area of ​​the battery module. The second temperature sensor is set on the surface of the battery module near the heating module, and the third temperature sensor is set on the edge area of ​​the battery module away from the heating module to monitor the local highest temperature and local lowest temperature, respectively.

[0051] The BMS main control unit uses the temperature collected by the first temperature sensor as the core battery temperature for constant temperature control, and uses the temperatures collected by each temperature sensor together as global measurement point temperature data.

[0052] The thermal insulation shell is located inside the battery compartment. The battery compartment air inlet and outlet are located at both ends of the battery compartment air flow channel outside the thermal insulation shell. The air in the battery compartment air flow channel does not directly enter the interior of the thermal insulation shell. The inlet temperature and outlet temperature are used to characterize the convective heat transfer changes caused by the air flow around the thermal insulation shell.

[0053] Simultaneously, the positioning detection module acquires the physical docking signal through the level transition state of the chassis docking interface; the power switching module detects the voltage and polarity of the output terminal of the external charging terminal, and generates an external power supply valid signal when the external power supply voltage is detected to be within the preset effective voltage range of 12V to 48V and the polarity is correct; the BMS main control unit reads the physical docking signal and the external power supply valid signal respectively, and reads the output current and remaining power of the lithium battery module through the internal current sampling circuit and the fuel gauge chip; then the BMS main control unit stores the received multi-dimensional sensing signals into the pre-divided dynamic parameter buffer area, and generates the basic measurement point data frame of the current acquisition cycle.

[0054] S202, the BMS main control unit reads the temperature time series corresponding to the battery compartment air inlet and outlet temperatures in the dynamic parameter buffer area; in an outdoor environment, the smart shopping cart will experience transient temperature sampling fluctuations due to irregular cold air convection. Then, the BMS main control unit performs a sliding time window filtering operation on the battery compartment air inlet and outlet temperatures based on the temperature sampling fluctuation characteristics; subsequently, the BMS main control unit calculates the difference between the battery compartment air outlet temperature and the battery compartment air inlet temperature at each sampling moment within the sliding time window, and performs window averaging on the difference to obtain a smoothed air inlet and outlet temperature difference feature value; the smoothed air inlet and outlet temperature difference feature value is used to characterize the temperature difference characteristics of the air flowing around the battery compartment before and after passing through the battery compartment; The specific formula for calculating the characteristic value of the smooth inlet and outlet air temperature difference is as follows: ; In the formula, To smooth out the characteristic values ​​of the temperature difference between the inlet and outlet air; This is the total number of sampling points included within the sliding time window, and the total number of sampling points is set between 10 and 50. This is the time step between two adjacent sampling points within the sliding window, and the time step is set between 0.1 seconds and 1.0 seconds. This refers to the sequence number of consecutive sampling points within the sliding time window; For the first time within the sliding time window Temperature of the battery compartment air outlet at each sampling point; For the first time within the sliding time window Temperature of the air inlet of the battery compartment at each sampling point.

[0055] S203, the motion sensing module continuously acquires spatial three-axis acceleration matrix data of the smart shopping cart at a fixed sampling frequency of 50Hz to 100Hz; the BMS main control unit constructs a variance calculation window according to 50 consecutive acceleration samples, so that each variance calculation window corresponds to a motion observation duration of 0.5 seconds to 1 second.

[0056] Next, the BMS main control unit synchronously receives spatial triaxial acceleration matrix data through the communication bus, and performs variance feature extraction calculation on the spatial triaxial acceleration matrix data in the discrete time dimension; relying on the mathematical principle of variance algorithm to characterize the deviation of discrete data from the mean, the BMS main control unit uses the variance feature extraction calculation results to quantitatively evaluate the intensity of pose oscillation of the smart shopping cart in three-dimensional physical space. The specific calculation formula for variance feature extraction is as follows: ; In the formula: The result of variance feature extraction; This represents the total number of acceleration samples accumulated within the variance calculation period. The variable is the sample index of acceleration within the variance calculation period; For the first The transient acceleration values ​​of each sample in the X-axis direction in three-dimensional space; For the first The transient acceleration value of each sample in the Y-axis direction in three-dimensional space; For the first The transient acceleration value of a sample in the Z-axis direction in three-dimensional space; The variance is calculated as the average value of the transient acceleration in the X-axis direction within the calculation period. The variance is calculated as the average value of the transient acceleration in the Y-axis direction within the period. This is the average value of the transient acceleration in the Z-axis direction within the variance calculation period.

[0057] S204, the BMS main control unit pre-configures the motion discrimination variance threshold in the system memory, and the BMS main control unit performs a numerical comparison operation between the variance feature extraction calculation result and the motion discrimination variance threshold; based on the baseline mechanical vibration variance statistical value during normal walking speed, the value range of the configured motion discrimination variance threshold is set to 0.01 to 0.1 standard gravitational acceleration square units.

[0058] The BMS main control unit configures motion entry variance threshold and motion exit variance threshold in the system memory, wherein the motion entry variance threshold is greater than the motion exit variance threshold. When the variance feature extraction result is greater than or equal to the motion entry variance threshold for no less than a first preset window number of consecutive operations, the motion stabilization flag is set to logic 1. When the variance feature extraction result is less than the motion exit variance threshold for no less than a second preset window number of consecutive operations, the motion stabilization flag is set to logic 0.

[0059] After the motion stabilization flag is set to logic 1, the BMS main control unit will maintain the motion stabilization flag at logic 1 even if the variance feature extraction result is briefly lower than the motion exit variance threshold during the preset motion hold time, so as to avoid the smart shopping cart frequently switching to the parking standby state due to brief stops during the push process; when the duration of the variance feature extraction result being lower than the motion exit variance threshold exceeds the preset motion hold time, the BMS main control unit will reset the motion stabilization flag to logic 0.

[0060] The first preset number of windows and the second preset number of windows are set to 2 to 5 respectively, and the preset motion holding time is set to 2 to 10 seconds.

[0061] Finally, the BMS main control unit sets the generated motion anti-shake flag as a rigid constraint input condition for switching working states, thereby isolating the mechanical vibration interference signal generated by occasional external force contact in the static state from the hardware interference of the working condition transition logic.

[0062] The BMS main control unit also calculates the anti-shake gating coefficient based on the variance feature extraction result. When the variance feature extraction result is less than or equal to the motion exit variance threshold, the anti-shake gating coefficient is set to zero. When the variance feature extraction result is greater than the motion exit variance threshold but less than the motion entry variance threshold, the anti-shake gating coefficient is linearly normalized according to the variance feature extraction result, so that the value of the anti-shake gating coefficient smoothly transitions between zero and one. When the variance feature extraction result is greater than or equal to the motion entry variance threshold, the anti-shake gating coefficient is set to one.

[0063] The motion stabilization flag is used to determine the working state, and the stabilization gating coefficient is used for dynamic thermal balance calculation. The two are used as discrete state variables and continuous compensation variables, respectively.

[0064] See attached document Figure 5 In specific implementations, step S3 provided by the present invention may include the following steps: S301, the BMS main control unit reads the received physical docking signal and the generated motion stabilization flag. Then, the BMS main control unit performs a working status determination based on the physical docking signal and the motion stabilization flag. The specific determination logic is as follows: when the physical docking signal is valid and the external power supply valid signal is valid, the smart shopping cart is determined to be in the charging state; when the physical docking signal is invalid and the motion stabilization flag is logic 1, the smart shopping cart is determined to be in the push-and-use state; when the physical docking signal is invalid and the motion stabilization flag is logic 0, the smart shopping cart is determined to be in the parking standby state; when the physical docking signal is valid but the external power supply valid signal is invalid, the smart shopping cart is determined to be in the parking standby state, and an external power supply abnormality flag is generated.

[0065] Subsequently, the BMS main control unit extracts the battery core temperature and sets the low-temperature control enable flag when the battery core temperature is less than or equal to the preheating start temperature. After the low-temperature control enable flag is set, it remains effective until the battery core temperature reaches the heating stop temperature, and it is not immediately reset when the battery core temperature is higher than the preheating start temperature, thus forming a temperature hysteresis control range between the preheating start temperature and the heating stop temperature. When the battery core temperature reaches the heating stop temperature, the smart shopping cart's working state changes, or the system triggers any of the following protection conditions: temperature exceeding the limit, insufficient power, or sensor failure, the BMS main control unit resets the low-temperature control enable flag and stops heating control. Finally, the BMS main control unit combines the working state judgment result with the low-temperature control enable flag to classify the smart shopping cart's working state into a low-temperature-constrained charging state, a pushing and using state, or a parking and standby state.

[0066] S302, when it is determined that the device is in a parking standby state and meets the low temperature conditions, the BMS main control unit extracts the external ambient temperature and calculates the deep power loss protection baseline power based on the preset model parameters according to the ambient temperature. At the same time, the BMS main control unit starts the internal timer to count the low temperature exposure time.

[0067] The heat dissipation phenomenon characterizes the process of continuous heat transfer from inside the lithium battery module to the external low-temperature environment. The BMS main control unit acquires the external ambient temperature at a preset sampling period and calculates the positive temperature difference between the nominal room temperature reference value and the external ambient temperature. When the external ambient temperature is lower than the nominal room temperature reference value, the BMS main control unit performs discrete cumulative updates on the low-temperature heat dissipation integral factor based on the positive temperature difference, the sampling time interval, and the heat dissipation attenuation coefficient. When the external ambient temperature is greater than or equal to the nominal room temperature reference value, the increase in the low-temperature heat dissipation integral factor stops.

[0068] The low-temperature heat dissipation integral factor is initialized to zero when the system is first powered on, and a preset maximum limit is set between 10,000 and 50,000. When the calculation result exceeds the preset maximum limit, the preset maximum limit is used as the effective low-temperature heat dissipation integral factor. When the lithium battery module completes its return-to-position preheating and reaches the heating stop temperature, the BMS main control unit decays the low-temperature heat dissipation integral factor according to a preset ratio between 50% and 80%, or clears the low-temperature heat dissipation integral factor to zero.

[0069] In normal operation and low-power sleep mode, the low-temperature heat dissipation integral factor is updated using the same state variable and the same temperature difference accumulation rule. In normal operation, it is updated using a fixed sampling period, which is set between 1 second and 5 seconds. In low-power sleep mode, it is compensated using the trapezoidal integral result between adjacent wake-up time nodes. The specific formula for calculating the low-temperature heat dissipation integral factor is as follows: ; In the formula, This is the integral factor for low-temperature heat dissipation; This is the upper limit of the integration time, representing the duration of continuous low-temperature exposure. This is the heat dissipation attenuation coefficient; The system's preset nominal room temperature reference value is set between 20°C and 25°C. It is a continuous integral time variable; For continuous integration time variables The transient external ambient temperature at that moment; For time variables The time element for continuous integration.

[0070] Next, the BMS main control unit compares the remaining power with the adaptive deep power loss protection baseline power. When the remaining power is greater than the adaptive deep power loss protection baseline power and the smart shopping cart has not yet entered the low-power sleep mode, the BMS main control unit controls the heating module to cycle between the heat preservation start temperature and the heating stop temperature to perform intermittent heat preservation, and the heat preservation start temperature is set between 2℃ and 8℃.

[0071] When the remaining power is less than or equal to the adaptive deep power loss protection baseline power, the BMS main control unit locks the heating control channel and triggers the deep sleep power-off protection logic; when the duration of the standby state reaches the sleep time threshold, and the sleep time threshold is set between 30 minutes and 2 hours, the BMS main control unit executes low-power sleep and periodic wake-up control.

[0072] S303: When it is determined that the system is in the charging state and the low temperature condition is met, the BMS main control unit sends a hardware control signal to control the power switching module to connect the DC power output by the external charging terminal after isolation AC-DC conversion, after confirming that both the physical docking signal and the external power supply valid signal are valid. When the external power supply valid signal is invalid, the external power supply voltage exceeds the preset range between 12V and 48V, or the polarity is abnormal, the BMS main control unit prohibits the power switching module from connecting the external power supply and outputs an external power supply fault signal.

[0073] The BMS main control unit reads the allowable charging temperature from the system memory, and the allowable charging temperature is set between 0°C and 5°C. When the core temperature of the battery is lower than the allowable charging temperature, the BMS main control unit keeps the battery charging channel closed and controls the external power supply to prioritize powering the heating module, so that the lithium battery module only performs preheating and does not perform charging.

[0074] When the battery core temperature reaches or exceeds the allowable charging temperature, the BMS main control unit releases the charging lockout and allows the charging module to charge the lithium battery module. When charging and heating are performed simultaneously, the BMS main control unit determines the current allowable preheating power of the heating module by subtracting the real-time charging power and the power of other electrical loads from the rated output power of the external power supply.

[0075] The permissible charging temperature is set according to the minimum permissible charging temperature specified by the battery cell manufacturer, and the permissible charging temperature is not lower than the minimum safe charging temperature of the battery cell.

[0076] Next, the BMS main control unit reads the generated low-temperature heat dissipation integral factor and introduces the low-temperature heat dissipation integral factor as a feedforward compensation component into the closed-loop feedback regulation loop; the BMS main control unit constructs the control equation by superimposing the feedforward compensation component, and then the BMS main control unit calculates the target preheating power by combining the low-temperature heat dissipation integral factor with the proportional-integral-derivative control equation. The specific calculation formulas for the proportional-integral-differential control equations are as follows: ; In the formula, Preheating power to the target; This is the scaling factor, and the scaling factor is set between 0.5 and 1.5; This represents the temperature deviation value at the current moment. This is the integral coefficient, and the integral coefficient is set between 0.01 and 0.1; For continuous time variables in the closed-loop feedback control circuit; This refers to the time differential element in closed-loop feedback control calculations. These are the differential coefficients, and the differential coefficients are set between 0.1 and 0.5; This represents the differential change of the temperature deviation value over an extremely small time interval. This is the weighting coefficient for heat dissipation compensation, and the weighting coefficient for heat dissipation compensation is set between 0.8 and 1.2. This is the integral factor for low-temperature heat dissipation.

[0077] The BMS main control unit determines the current allowable preheating power based on the rated maximum power of the heating module, the rated output power of the external power supply, the real-time power of other electrical loads in the smart shopping cart, and the allowable output power of the charging equipment. When the calculated target preheating power is less than zero, the target preheating power is set to zero. When the target preheating power is greater than the current allowable preheating power, the current allowable preheating power is used as the final target preheating power.

[0078] When the final target preheating power reaches the upper or lower limit, the BMS main control unit suspends the accumulation of integral terms in the proportional-integral-derivative control equation towards saturation, or performs reverse calculation on the integral terms to avoid integral saturation causing the heating module to maintain maximum power output for a long time.

[0079] The BMS main control unit outputs a pulse width modulation signal based on the final target preheating power after upper and lower amplitude limits, and drives the heating module to work continuously at the duty cycle corresponding to the final target preheating power until the battery core temperature rises to the upper temperature limit, which is specifically configured as the heating stop temperature. The BMS main control unit then cuts off the pulse width modulation signal and stops driving the heating module.

[0080] S304, when it is determined that the vehicle is in a push-to-use state and meets the low temperature conditions, the BMS main control unit synchronously extracts the real-time read output current, anti-shake gate coefficient, and smooth inlet and outlet air temperature difference characteristic value; then, the BMS main control unit calculates the deduction value of the self-heating generated during the battery discharge process in combination with the internal Joule heat conversion coefficient, and the BMS main control unit calculates the windward heat loss value during the movement of the smart shopping cart in combination with the air-cooled convection compensation coefficient; subsequently, the BMS main control unit substitutes the output current, anti-shake gate coefficient, and smooth inlet and outlet air temperature difference characteristic value into the dynamic heat balance calculation formula for solution, and then obtains the target control power under the condition of comprehensively considering the balance relationship between heat loss and self-heating; The specific calculation formula for dynamic heat balance is as follows: ; In the formula, Control power to the target; To maintain the calibration baseline heat demand power of the baseline temperature scale, the calibration baseline heat demand power is set between 5W and 20W; This refers to the internal Joule heat conversion coefficient; This represents the output current of the lithium battery module under current operating conditions. This is the air-cooled convection compensation coefficient; For anti-shake gate control coefficient; To smooth the characteristic value of the temperature difference between the inlet and outlet air.

[0081] The BMS main control unit calculates the current allowable heating power based on the rated maximum power of the heating module, the current allowable discharge power of the lithium battery module, the current output current, the remaining power, and the load requirements of the shopping cart drive mechanism. Here, the air-cooled convection compensation coefficient is calibrated based on the nonlinear mapping relationship between the actual wind speed and the smooth inlet and outlet air temperature difference characteristics, ensuring that the heat loss caused by increased wind speed can be equivalently converted into a positive power compensation requirement. When the target control power calculated by dynamic thermal balance is less than zero, the final control power is set to zero; when the target control power exceeds the current allowable heating power, the current allowable heating power is used as the final control power.

[0082] When the remaining charge of the lithium battery module is less than or equal to the adaptive deep discharge protection baseline charge, the output current reaches the upper limit of the battery's allowable discharge current, or the BMS main control unit detects battery undervoltage, the BMS main control unit will ultimately set the control power to zero and lock the heating control channel.

[0083] The BMS main control unit outputs a dynamic heating command to the heating module based on the target control power obtained from the solution, and controls the heating module to adjust the driving current of the heating component in real time according to the dynamic heating command.

[0084] In specific implementations, step S4 provided by this invention may include the following steps: In S401, the BMS main control unit configures a high-priority hardware timer in the underlying operating system. Then, the BMS main control unit uses this high-priority hardware timer to execute a global temperature limit detection task independently of the state machine logic. Through this global temperature limit detection task, the BMS main control unit monitors the global measurement point temperature data in the dynamic parameter buffer in real time. Since the initial stage of lithium battery thermal runaway is accompanied by an abnormal high-temperature rise, the BMS main control unit then extracts the battery core temperature sequence from the global measurement point temperature data. Furthermore, the BMS main control unit performs differential operations on the battery core temperature sequence based on a fixed sampling time interval. The BMS main control unit quantifies the severity of temperature fluctuations per unit time and extracts the real-time temperature rise rate through differential operations. The specific formula for calculating the real-time temperature rise rate is as follows: ; In the formula, For real-time temperature rise rate; This is the current discrete sampling period number variable for the global limit violation detection task; For the first The current core temperature of the battery is collected in a discrete sampling period; For the first The battery core temperature of the previous cycle is collected in each discrete sampling cycle. This is the fixed sampling time interval for the global limit violation detection task, and the fixed sampling time interval is set between 100 milliseconds and 500 milliseconds.

[0085] S402, the BMS main control unit reads the maximum safe temperature and the preset temperature rise rate limit threshold from the threshold configuration area in the system memory. The BMS main control unit then compares the temperature data of each measuring point in the global measuring point temperature data with the maximum safe temperature. Simultaneously, the BMS main control unit compares the real-time temperature rise rate with the temperature rise rate limit threshold. The temperature rise rate limit threshold is calibrated based on the safety test data of the battery cell used, the response time of the temperature sensor, the maximum temperature rise rate under normal heating conditions, and the rated power of the heating module, and can be set from 0.5℃ / s to 1.0℃ / s. The temperature rise rate limit threshold should be higher than the maximum allowable temperature rise rate under normal preheating conditions and lower than the temperature rise rate corresponding to when the battery cell enters an abnormally rapid heating state.

[0086] When the temperature of all global measuring points is less than the maximum safe temperature, and the real-time temperature rise rate is less than the temperature rise rate limit threshold within a continuous preset confirmation period, the BMS main control unit maintains the normal operation of the system; when the real-time temperature rise rate is greater than or equal to the temperature rise rate limit threshold within a continuous preset confirmation period, or when the temperature of any measuring point is greater than or equal to the maximum safe temperature, the BMS main control unit triggers the high-priority temperature protection interrupt logic and generates a software blocking instruction.

[0087] The preset confirmation period is set to 2 to 5 sampling periods to suppress false triggering caused by single sampling noise; when the temperature at any measuring point reaches the maximum safe temperature, no delayed confirmation is performed, and protection is triggered immediately.

[0088] Finally, the BMS master control unit generates a hardware lockout instruction in the underlying system while triggering the hardware forced interrupt logic.

[0089] At least one analog temperature sensor for acquiring the core temperature of the battery or the surface temperature of the heating module is simultaneously connected to an independent hardware comparator; the comparison threshold of the independent hardware comparator is set according to the maximum safe temperature, and the output of the independent hardware comparator is directly connected to the hardware enable terminal of the heating branch power switch.

[0090] When the output signal of the temperature sensor exceeds the comparison threshold of the independent hardware comparator, the independent hardware comparator does not need to wait for the BMS main program to execute, and directly outputs a hardware shutdown signal to put the heating branch power switch into the off state. The hardware latch circuit keeps it in the off state until the fault check and manual reset are completed or the preset safety reset conditions are met.

[0091] After generating the hardware lockout command, the S403 main control unit directly sends a power-off command to the power switching module and the heating module. The power switching module outputs a forced low-level signal according to the power-off command, controlling the heating branch relay or the heating branch power switch to disconnect.

[0092] Simultaneously, the heating module blocks the pulse width modulation signal output channel of the drive controller according to the power-off command; the system then physically cuts off the heating execution circuit and forces the heating component to stop working through the above disconnection and blocking operations; then the BMS main control unit broadcasts a thermal runaway fault message to the communication bus and simultaneously outputs a high temperature fault alarm signal. The existing alarm devices of the smart shopping cart issue audible and visual prompts based on the high temperature fault alarm signal, or the BMS main control unit sends the high temperature fault alarm signal to the smart shopping cart upper controller through the communication bus, and the upper controller controls the existing alarm devices to execute audible and visual prompts.

[0093] Finally, the BMS main control unit packages the battery core temperature and timestamp at the moment of exceeding the limit and writes them into the non-volatile storage area of ​​the system memory. Then, in the event of a system power failure, it relies on the hardware log writing mechanism to complete the solidification operation of system fault tracing data.

[0094] See attached document Figure 6 In specific implementations, step S5 provided by the present invention may include the following steps: S501, the BMS main control unit continuously monitors the duration of the smart shopping cart in standby mode. Then, when the BMS main control unit determines that the duration of standby mode exceeds a preset time threshold between 30 minutes and 2 hours, it activates the sleep-wake mechanism. The BMS main control unit reads the current remaining power and the deep power loss protection baseline power, and calculates the difference between the remaining power and the deep power loss protection baseline power to obtain the deviation of the remaining power from the baseline.

[0095] Subsequently, the BMS main control unit determines the stepped backoff wake-up cycle based on the deviation between the remaining power and the adaptive deep power loss protection bottom line power. The smaller the deviation, the longer the stepped backoff wake-up cycle, so as to reduce the periodic wake-up power consumption when the remaining power is close to the protection bottom line.

[0096] For example, when the deviation is greater than 20 percentage points, the step-back wake-up cycle is set as the first wake-up cycle; when the deviation is greater than 10 percentage points and less than or equal to 20 percentage points, it is set as the second wake-up cycle; when the deviation is greater than 5 percentage points and less than or equal to 10 percentage points, it is set as the third wake-up cycle; and when the deviation is greater than 0 and less than or equal to 5 percentage points, it is set as the fourth wake-up cycle. The first wake-up cycle is shorter than the second wake-up cycle, the second wake-up cycle is shorter than the third wake-up cycle, and the third wake-up cycle is shorter than the fourth wake-up cycle.

[0097] The first to fourth wake-up cycles can be set to 10 minutes, 20 minutes, 30 minutes and 1 hour respectively. The specific values ​​are adjusted according to the lithium battery module capacity, the static power consumption during hibernation and the rate of change of ambient temperature. If the external ambient temperature is lower than the preset extreme cold threshold before hibernation, the system will forcibly limit the longest wake-up cycle to within 1 hour to prevent the battery from undergoing irreversible over-discharge during the long hibernation period.

[0098] Before entering low-power sleep mode, the BMS main control unit stops the pulse width modulation signal output of the heating module, disconnects the heating execution circuit, saves the current low-temperature heat dissipation integral factor, external ambient temperature, battery core temperature, remaining power, and sleep start timestamp, and uses the data at the start of sleep as the zero sampling point of the sleep period discrete monitoring dataset; then the BMS main control unit sends peripheral sleep commands to non-critical power-consuming peripherals, cuts off the corresponding hardware power supply circuits, and shuts down the system main frequency clock, so that the system enters a low-power sleep mode that operates according to the stepped backoff wake-up cycle.

[0099] In low-power sleep mode, the real-time clock module, the homing detection input circuit, the external power supply valid signal detection circuit, and the low-power motion interruption unit of the motion sensing module remain in the constant power supply domain; the temperature acquisition module, the complete triaxial acceleration data acquisition circuit, the communication module, and the heating drive circuit are in a power-off or low-power state during non-sampling or non-heating periods.

[0100] In low-power sleep mode, the motion sensing module does not continuously output complete spatial triaxial acceleration matrix data. It only outputs a push wake-up signal to the BMS main control unit through an independent hardware interrupt pin when it detects an acceleration change exceeding a preset wake-up threshold between 0.1g and 0.3g.

[0101] During the low-power sleep mode, the real-time clock module of S502 continues to perform independent timing. Then, when the BMS main control unit reaches the periodic wake-up time node corresponding to the stepped backoff wake-up cycle, it receives the hardware wake-up interrupt signal. The BMS main control unit responds to the hardware wake-up interrupt signal and briefly restores the system main frequency clock.

[0102] Subsequently, the BMS main control unit controls the temperature acquisition module to quickly acquire the external ambient temperature and transient battery core temperature at the periodic wake-up time node, and simultaneously reads the remaining power of the lithium battery module; the BMS main control unit recalculates the adaptive deep power loss protection baseline power based on the current external ambient temperature, and compares the current remaining power with the adaptive deep power loss protection baseline power.

[0103] When the remaining power is less than or equal to the adaptive deep power loss protection baseline power, the BMS main control unit no longer waits for the push trigger signal or the charging trigger signal, directly locks the heating control channel, cuts off unnecessary power supply circuits, and controls the temperature control system to enter the deep sleep power failure protection state.

[0104] When the remaining battery power is greater than the adaptive deep power loss protection threshold, the BMS main control unit packages the external ambient temperature, transient battery core temperature, remaining battery power, and the timestamp corresponding to the wake-up time into discrete sparse sampling data points, and appends the discrete sparse sampling data points to the non-volatile memory.

[0105] When the transient battery core temperature is lower than the heat preservation start temperature, and the current remaining charge is higher than the sum of the adaptive deep discharge protection baseline charge and the preset charge margin, and the preset charge margin is set between 2% and 5%, the BMS main control unit allows the heating module to perform a time-limited pulse heat preservation once. The time-limited pulse heat preservation ends when the transient battery core temperature reaches the heating stop temperature, the heating duration reaches the preset maximum time between 10 and 20 minutes, or the remaining charge drops to the sum of the adaptive deep discharge protection baseline charge and the preset charge margin. After completing sampling or time-limited pulse heat preservation, the BMS main control unit shuts down the system main frequency clock again and returns to low-power sleep mode.

[0106] S503 When the low-power motion interruption unit of the motion sensing module outputs a push wake-up signal, or the homing detection input circuit and the external power supply valid signal detection circuit output a charging wake-up signal, the BMS main control unit exits the low-power sleep mode and restores the system main frequency clock.

[0107] Next, the BMS main control unit reads the discrete monitoring dataset of the hibernation period that is stored in the non-volatile memory; then the BMS main control unit performs discrete interval integral compensation operation on the discrete monitoring dataset. The BMS main control unit uses the external ambient temperature and hibernation start timestamp saved before entering hibernation as the zero sampling point, and uses the trapezoidal integral algorithm to construct trapezoidal integral elements between the zero sampling point and the first cycle wake-up sampling point and between each adjacent cycle wake-up sampling point.

[0108] When the discrete monitoring dataset during the dormant period contains only the zeroth sampling point and no subsequent periodic wake-up sampling points, trapezoidal integration is not performed; when the timestamps of adjacent sampling points are abnormal, the time interval is zero, or the temperature data exceeds the effective range of the sensor, the BMS main control unit discards the corresponding integration interval and generates a sampling abnormality flag.

[0109] Non-volatile memory uses circular storage, partitioning, or wear leveling to record discrete monitoring data during the dormant period in order to reduce memory lifespan loss caused by frequent writes.

[0110] The BMS main control unit calculates the temperature integral value over a discrete time interval using the trapezoidal integral algorithm, and the BMS main control unit calculates the cumulative integral data during the dormancy period. The specific formula for calculating cumulative integral data is as follows: ; In the formula, This refers to the cumulative points data during the dormant period; This represents the total number of samples taken at the periodic wake-up time points during the sleep period. This is the collection sequence number variable for the periodic wake-up time nodes during the hibernation period; The system's preset nominal room temperature reference value is set between 20°C and 25°C. For the first Ambient temperature data collected at each wake-up time point; For the first The ambient temperature data of the previous cycle collected at each wake-up time node; For the first The wake-up timestamp corresponding to each wake-up time node in the cycle; For the first Each wake-up time point corresponds to the wake-up timestamp of the previous cycle.

[0111] S504, during the hibernation period, the system's underlying heat dissipation continues to occur naturally. The BMS main control unit then extracts the historical low-temperature heat dissipation integral factor stored before entering hibernation. Subsequently, the BMS main control unit multiplies the accumulated integral data during hibernation by the heat dissipation attenuation coefficient to obtain the hibernation heat dissipation compensation amount. This compensation amount is then added to the historical low-temperature heat dissipation integral factor stored before entering hibernation, resulting in an updated low-temperature heat dissipation integral factor. When the updated low-temperature heat dissipation integral factor exceeds a preset maximum limit (set between 10,000 and 50,000), the BMS main control unit limits the updated low-temperature heat dissipation integral factor to the preset maximum limit.

[0112] Next, the BMS main control unit compares the real-time remaining power with the deep power loss protection threshold power again. Finally, when the BMS main control unit determines that the remaining power is lower than the deep power loss protection threshold power, it generates a low-level hardware cutoff command. Based on the low-level hardware cutoff command, the BMS main control unit locks the heating control channel and controls the temperature control system to enter the deep sleep power-off protection state. In the power-off state, the BMS main control unit maintains the static physical lock of the battery's low-level remaining energy until the return detection module obtains a valid physical docking signal and a valid external power supply signal again. Only then can the system exit the deep sleep power-off protection state and perform a hardware restart.

[0113] See attached document Figure 7 To aid in understanding the technical solution of this invention, the following is an application embodiment of multi-condition constant temperature management of lithium batteries in intelligent shopping carts under low-temperature environments.

[0114] The application example receives an extreme cold environment simulation test dataset for verifying the stability of dynamic thermal balance calculation and the accuracy of temperature control during multi-condition switching. The extreme cold environment simulation test dataset contains a series of dynamic operating conditions of a shopping cart, consisting of different ambient temperature step drops in the range of -20℃ to 0℃, multi-band irregular wind speed convection characteristics, and simulated high-frequency mechanical vibration interference. At the same time, the extreme cold environment simulation test dataset includes the initial battery core temperature data in a dormant state under the corresponding test conditions, the temperature drop data of traditional constant temperature control without wind-cooled convection compensation, and the real value of the physical reference of relative heat dissipation of the lithium battery module, which is synchronously recorded by an external traceable high-precision thermal imager and a power meter.

[0115] In addition, the extreme cold environment simulation test dataset also includes dynamic spatial triaxial acceleration features introduced through a preset route, and the system uses the real start-stop state switching nodes associated with the dynamic spatial triaxial acceleration features as the real value reference boundary for testing the system's motion anti-shake and working condition transition response capabilities.

[0116] For the extreme cold environment simulation test dataset, the BMS main control unit first completes the initial configuration of the temperature control threshold matrix and power protection reference parameters in the system memory, and uses the non-volatile storage area to divide the independent threshold configuration area and dynamic parameter cache area to write the internal Joule heat conversion coefficient, air-cooled convection compensation coefficient and other characteristic coefficients. Then, it performs the system hardware status self-test operation to establish the initial constant temperature management control reference.

[0117] During the dynamic disturbance acquisition phase of the current test frame, the temperature acquisition module, the homing detection module, and the motion sensing module work synchronously. The fixed sampling frequency of the motion sensing module is configured to cover the typical oscillation cycle of the shopping cart, and the time step of the sliding time window is kept smaller than the low-frequency micro-disturbance characteristic time of the ambient cold wind gusts. By performing continuous time series sampling of the temperature of the air inlet and outlet of the battery compartment and the spatial triaxial acceleration, which are spatially overlapping, the system captures comprehensive sensing information including thermal convection loss error and mechanical vibration interference. Finally, it outputs a high signal-to-noise ratio basic measurement point data frame that effectively freezes transient sampling fluctuations and has multi-source dimensional signal time alignment.

[0118] The data parsing and state determination stage acquires basic measurement point data frames and performs time-domain and discrete-domain feature extraction operations. The BMS main control unit uses a filtering algorithm with a sliding time window to extract the smooth inlet and outlet air temperature difference feature values ​​of the corresponding battery compartment air outlet and air inlet in parallel. Subsequently, the BMS main control unit calculates the transient anti-shake gating coefficient reflecting the intensity of global three-dimensional spatial pose oscillation based on the deviation characteristics of the variance feature extraction calculation results of continuous acceleration samples and the variance threshold of motion discrimination. At the same time, the BMS main control unit performs hysteresis confirmation on the extracted variance data in combination with the preset window number, and generates discrete motion anti-shake flags in combination with physical docking signals and external power supply valid signals. Finally, the BMS main control unit establishes a working condition mask based on the rigid constraint input conditions of working state switching, and accurately divides the smart shopping cart into the charging state, the pushing and using state, or the parking and standby state.

[0119] The control calculation module receives the smooth inlet and outlet air temperature difference characteristic value, motion anti-shake flag, and transient remaining power data. Based on the ambient temperature, the control calculation module calls the preset model parameters to complete the mathematical construction of the adaptive deep power loss protection baseline power. It also performs discrete accumulation on the low temperature exposure time based on the upper limit extreme value of the integration time. Then, the control calculation module uses the positive temperature difference between the nominal normal temperature reference value and the external ambient temperature to construct the transient heat dissipation surface, and combines the heat dissipation attenuation coefficient to compensate for the global heat loss, thereby obtaining the effective low temperature heat dissipation integral factor constrained by the limit of 10,000 to 50,000.

[0120] The control calculation module performs dynamic heat balance equation calculations for the operating conditions under the push-out state, thereby generating the target control power. After completing the equivalent compensation for heat loss based on the nonlinear mapping relationship between the air-cooled convection compensation coefficient and the smooth inlet and outlet air temperature difference characteristic value, the control calculation module calls the self-heating deduction formula with the weight of remaining power and output current to perform the internal heat source weight reduction operation according to the pre-established internal Joule heat conversion coefficient. This internal heat source weight reduction and external cooling compensation operation retain the actual heat generation difference caused by the shopping cart push load, while generating the final control power that filters out transient cold air disturbances and current spike noise.

[0121] The control solution module substitutes the low-temperature heat dissipation integral factor, real-time temperature deviation value, and proportional-integral-differential control equations into the preheating solution loop for the homing charging state. In the closed-loop feedback control operation, the control solution module performs differential operation and upper and lower limit rounding operation, and combines the regional consistency judgment of battery core temperature and charging allowable temperature to output the duty cycle of the absolute pulse width modulation signal matching the rated maximum power of the heating module. At the same time, the control solution module reverse-engineers the real-time temperature rise rate based on global measurement point temperature data and the system's high-priority hardware timer, and then outputs the corrected hardware-level over-limit protection trigger command. Finally, the control solution module jointly executes the equation to solve the hibernation step back-off wake-up cycle, trapezoidal integral accumulation data, and updated low-temperature heat dissipation integral factor, thereby forming a closed-loop solution for the long-term parking hibernation reconstruction of the extreme cold environment simulation test dataset, and finally outputs the deep hibernation power failure protection and intermittent wake-up control timing matrix in a unified underlying physical coordinate system.

[0122] Through the closed-loop collaborative operation of multiple core modules on the extreme cold environment simulation test dataset, the application example compares and verifies the independent solution process of turning on and off dynamic thermal balance compensation and stepped sleep trapezoidal integral smoothing. The verification results show that: compared with the traditional fixed threshold temperature control algorithm without feedforward compensation, the system relies on setting adaptive power loss bottom line constraints and variance feature extraction operations to block the logic dead zone of frequent working condition jumps caused by mechanical vibration noise amplification. Furthermore, the system utilizes the nonlinear mapping between actual wind speed and smooth inlet and outlet air temperature difference to correct the heat dissipation benchmark, eliminating the blind spots of outdoor heat loss and overheating waste caused by fixed heating power approximation. In the application example, the final output dynamic battery core temperature curve is compared with the true value of the external traceable physical benchmark by performing time axis registration and residual profile feature comparison. The temperature step consistency at the boundary of the working condition and the root mean square error of global power consumption prove that the system successfully avoids the nonlinear overshoot caused by high-frequency environmental perturbation signals, while accurately restoring the true three-dimensional thermodynamic boundary topology and the absolutely safe constant temperature calibration state of the shopping cart lithium battery.

[0123] While specific embodiments of the present invention have been described above, those skilled in the art should understand that these specific embodiments are merely illustrative. Those skilled in the art can omit, substitute, and modify the details of the above methods and systems in various ways without departing from the principles and essence of the present invention. For example, combining the above method steps to perform substantially the same function and achieve substantially the same result using substantially the same method falls within the scope of the present invention.

Claims

1. A multi-condition temperature control method for a shopping cart lithium battery under low-temperature conditions, characterized in that, The method includes the following steps: Configure the temperature control threshold matrix and power protection reference parameters, and write the characteristic coefficients for dynamic thermal balance calculation; Acquire battery core temperature, external ambient temperature, battery compartment air inlet and outlet temperatures, physical docking signal, output current, remaining power, and spatial triaxial acceleration matrix data; perform sliding time window filtering on the battery compartment air inlet and outlet temperatures to extract smooth air inlet and outlet temperature difference feature values, and perform variance feature extraction on the spatial triaxial acceleration matrix data to generate motion stabilization flags; The charging status, the push-to-use status, or the parking standby status are determined based on the physical docking signal and the motion stabilization flag. In the standby state, the heating module is controlled to perform intermittent heat preservation; in the charging state, the target preheating power is calculated by combining the low temperature heat dissipation integral factor to drive the heating module; in the push-to-use state, the smooth inlet and outlet air temperature difference characteristic value is substituted into the dynamic heat balance calculation formula to obtain the target control power and output the dynamic heat replenishment command. Monitor the temperature at all measuring points. If the temperature at any measuring point exceeds the maximum safe temperature, cut off the heating execution circuit and output a high temperature fault alarm signal. If the standby state continues for more than a preset time threshold, the hibernation wake-up mechanism will be activated. At the periodic wake-up time node, the accumulated data during the hibernation period will be calculated and supplemented into the low temperature heat dissipation integral factor. When the remaining power is lower than the deep power loss protection bottom line power, the heating control channel will be locked and the deep hibernation power-off protection state will be entered.

2. The multi-condition temperature control method for shopping cart lithium batteries under low-temperature environments according to claim 1, characterized in that, The steps for configuring the temperature control threshold matrix and the power protection reference parameters include: Write the preheating start temperature, the discharge heat preservation lower limit temperature, the heating stop temperature, and the maximum safe temperature, and combine them to configure the temperature control threshold matrix. The baseline power reserve and environmental compensation coefficient are established as the power protection benchmark parameters, and a mathematical model of the deep power loss protection baseline power reserve is constructed based on the baseline power reserve and environmental compensation coefficient. The external ambient temperature is introduced as an independent variable to compensate the slope of the basic reserve power. When the external ambient temperature drops, the deep power loss protection baseline power is dynamically increased. The solution logic of the deep power loss protection baseline power is expressed as a linear superposition of the basic reserve power and the environmental compensation correction term.

3. The multi-condition temperature control method for shopping cart lithium batteries under low-temperature environments according to claim 1, characterized in that, The step of writing the characteristic coefficients for dynamic thermal equilibrium calculation includes: The characteristic coefficients for dynamic heat balance calculation include the internal Joule heat conversion coefficient, the air-cooled convection compensation coefficient, and the heat dissipation attenuation coefficient. Configure the internal Joule thermal conversion coefficient and establish the heat offset mapping relationship generated by the internal resistance heating of the battery under the promoted use state; Configure the air-cooled convection compensation coefficient to quantify the surface air convection heat loss during the movement of the smart shopping cart; Configure a heat dissipation attenuation coefficient to reconstruct the feedforward heating power compensation benchmark when the parking standby state transitions to the return charging state.

4. The multi-condition temperature control method for shopping cart lithium batteries under low-temperature environments according to claim 1, characterized in that, The steps for performing variance feature extraction on the spatial triaxial acceleration matrix data to generate motion stabilization flags include: A variance calculation window is constructed according to a continuously preset number of samples. Variance feature extraction is performed on the spatial triaxial acceleration matrix data in the discrete time dimension to quantitatively evaluate the intensity of pose oscillation of the smart shopping cart in three-dimensional physical space and obtain the variance feature extraction result. The variance feature extraction result is compared with the pre-configured motion entry variance threshold and motion exit variance threshold to generate the motion anti-shake flag used to isolate mechanical vibration interference signals; The anti-shake gating coefficient is calculated based on the variance feature extraction results, and the value of the anti-shake gating coefficient smoothly transitions between zero and one.

5. The multi-condition temperature control method for shopping cart lithium batteries under low-temperature environments according to claim 1, characterized in that, The steps for determining the charging status, the push-to-use status, or the parking standby status based on the physical docking signal and the motion stabilization flag include: When the physical docking signal is valid and the external power supply signal is valid, the smart shopping cart is determined to be in the charging state. When the physical docking signal is invalid and the motion stabilization flag is logic 1, it is determined that the smart shopping cart is in the push-and-use state; When the physical docking signal is invalid and the motion stabilization flag is logic 0, it is determined that the smart shopping cart is in the parking standby state. When the physical connection signal is valid but the external power supply signal is invalid, the smart shopping cart is determined to be in the parking standby state, and an external power supply abnormality flag is generated.

6. The multi-condition temperature control method for shopping cart lithium batteries under low-temperature environments according to claim 1, characterized in that, The step of calculating the target preheating power to drive the heating module in the homing charging state, combined with the low-temperature heat dissipation integral factor, includes: The low-temperature heat dissipation integral factor is introduced as a feedforward compensation component into the closed-loop feedback control loop, and the target preheating power is calculated by combining the low-temperature heat dissipation integral factor with the proportional-integral-derivative control equation. The specific calculation logic of the proportional-integral-derivative control equation is as follows: the basic adjustment amount is calculated by combining the temperature deviation value, proportional coefficient, integral coefficient and derivative coefficient at the current moment, and the low temperature heat dissipation integral factor is multiplied by the heat dissipation compensation weight coefficient and then superimposed on the basic adjustment amount. The current allowable preheating power is determined based on the rated maximum power of the heating module and the rated output power of the external power supply. The target preheating power is then subject to upper and lower limit constraints to obtain the final target preheating power. Based on the final target preheating power output pulse width modulation signal, the heating module is driven to work continuously until the battery core temperature reaches the heating stop temperature in the temperature control threshold matrix.

7. The multi-condition temperature control method for a shopping cart lithium battery under low-temperature conditions according to claim 1, characterized in that, The steps of substituting the smooth inlet and outlet air temperature difference characteristic value into the dynamic heat balance calculation formula to obtain the target control power and outputting the dynamic heat replenishment command in the promoted use state include: Obtain the calibrated basic thermal demand power to maintain the basic temperature scale, and combine the internal Joule thermal conversion coefficient in the dynamic thermal balance calculation characteristic coefficient and the output current to calculate the deduction value of the self-heating generated during the battery discharge process. The wind-cooled convection compensation coefficient in the dynamic heat balance calculation characteristic coefficient, the anti-shake gate coefficient derived from the variance feature extraction operation, and the smooth inlet and outlet air temperature difference characteristic value are combined to calculate the windward heat loss value during the movement of the smart shopping cart. The target control power is generated by subtracting the deduction value from the calibrated basic heat demand power and adding the windward heat loss value. The target control power is limited based on the current allowable heating power to obtain the final control power, and the dynamic heat replenishment command is output based on the final control power.

8. The multi-condition temperature control method for a shopping cart lithium battery under low-temperature conditions according to claim 1, characterized in that, The step of monitoring the global temperature measurement points and cutting off the heating execution circuit and outputting a high-temperature fault alarm signal when the temperature at any measurement point exceeds the maximum safe temperature includes: Configure a high-priority hardware timer to independently execute the global temperature over-limit detection task, and perform differential operation on the temperature data sequence constructed based on the core temperature of the battery in combination with a fixed sampling time interval to extract the real-time temperature rise rate; When the real-time temperature rise rate is greater than or equal to the preset temperature rise rate limit threshold within a continuous preset confirmation period, or when the temperature at any measuring point is greater than or equal to the maximum safe temperature, the high-priority temperature protection interruption logic is triggered. A power-off control signal is issued to physically disconnect the heating execution circuit, and the high-temperature fault alarm signal is simultaneously output to the outside via the communication bus. The battery core temperature and timestamp at the time of the over-limit event are then permanently written into the non-volatile storage area.

9. The multi-condition temperature control method for a shopping cart lithium battery under low-temperature conditions according to claim 1, characterized in that, The step of initiating a sleep / wake-up mechanism when the parking standby state continues for more than a preset time threshold, and calculating the cumulative integral data during the sleep period to supplement the low-temperature heat dissipation integral factor at the periodic wake-up time node includes: The stepped backoff wake-up cycle is determined based on the deviation between the remaining power and the deep power depletion protection baseline power. Before entering low-power sleep mode, the current low-temperature heat dissipation integral factor is saved as the historical low-temperature heat dissipation integral factor and sleep start timestamp. When the periodic wake-up time node corresponding to the stepped back wake-up cycle is reached, the external ambient temperature integral value between adjacent periodic wake-up time nodes is calculated using the trapezoidal integral algorithm to obtain the integral cumulative data during the hibernation period. Multiply the accumulated integral data by the heat dissipation attenuation coefficient in the dynamic heat balance calculation characteristic coefficient to obtain the dormant heat dissipation compensation amount. Add the dormant heat dissipation compensation amount to the historical low temperature heat dissipation integral factor to obtain the updated low temperature heat dissipation integral factor.

10. A multi-condition temperature control system for a shopping cart lithium battery under low-temperature conditions, characterized in that, The multi-condition temperature control method for a shopping cart lithium battery under low-temperature conditions according to any one of claims 1 to 9, the system comprising: The lithium battery module is located inside the battery compartment of the smart shopping cart chassis; A thermal insulation shell is disposed inside the battery compartment and surrounds the outer periphery of the lithium battery module; A heating module is arranged inside the thermal insulation shell and is used to perform heat conduction operation in close contact with the surface of the lithium battery module shell. The power switching module is configured at the input hub of the power supply link. It is used to receive the grid power introduced by the external charging interface and the battery power fed back by the lithium battery module and perform power supply link switching, thereby outputting the target driving power to the heating module. The temperature acquisition module is configured at the system's data acquisition node to collect the battery core temperature, external ambient temperature, and battery compartment air inlet and outlet temperatures. The repositioning detection module, configured at the data acquisition node, is used to acquire physical docking signals; A motion sensing module, configured at the data acquisition node, is used to acquire spatial triaxial acceleration matrix data in real time. The BMS main control unit is connected to the temperature acquisition module, the homing detection module, the motion sensing module, the power switching module, and the heating module to acquire the physical docking signal, the spatial triaxial acceleration matrix data, and various temperature data. The BMS main control unit is used to generate a motion stabilization flag using the spatial triaxial acceleration matrix data, and then combines the physical docking signal and the motion stabilization flag to perform a working status determination operation, classifying the smart shopping cart into a homing charging state, a pushing and using state, or a parking and standby state. The BMS main control unit is also used to generate a target control power based on the various temperature data, and send a pulse width modulation control signal to the heating module according to the target control power to perform dynamic constant temperature compensation operation.

Images