Remote control method for emergency starting of a parking air conditioner battery

Abstract

The application relates to the technical field of battery remote control, and discloses a remote control method for a parking air conditioner battery for emergency starting, wherein the method comprises the following steps: a background management system carries out geographic fence matching and digital certificate verification on geographic coordinate information obtained by a built-in GPS module of an emergency starting battery to obtain a remote control permission authentication state; load mode identification is carried out to obtain a current power consumption mode classification result; battery state analysis is carried out to obtain an SOC (state of charge) value; when a BMS (battery management system) discharge switch carries out pre-charging timing control in an emergency starting scene, a target current limiting parameter is obtained; and the BMS charge-discharge switch carries out power level division and time period scheduling in a parking air conditioner scene to obtain a voltage output control parameter. The method can accurately identify the dual-mode load characteristics of an emergency starting large-current impact and a parking air conditioner small-current continuous discharge, solves the problem of inaccurate load identification of a traditional system, and improves the battery use efficiency.

Description

Technical Field

[0001] This invention relates to the field of battery remote control technology, and in particular to a remote control method for a parking air conditioner battery used for emergency starting. Background Technology

[0002] Emergency start parking air conditioner batteries need to simultaneously meet the dual requirements of instantaneous high current output during emergency starts and continuous low current power supply for parking air conditioners over extended periods. Their load characteristics exhibit significant dual-mode features. In emergency start scenarios, the battery needs to provide an instantaneous high current of 400-800A within 3-8 seconds to start the vehicle engine, while in parking air conditioner scenarios, it needs to stably provide a low current of 15-35A and operate continuously for several hours.

[0003] Existing battery management systems employ a single-load mode control strategy, which cannot effectively identify and distinguish between the instantaneous high-current surge of emergency startup and the continuous low-current discharge characteristics of parking air conditioning. This results in insufficient accuracy in battery state estimation, easily leading to SOC estimation deviations and inaccurate battery life predictions. Furthermore, existing systems lack differentiated charge and discharge control strategies for different application scenarios. Emergency startups in low-temperature environments can easily cause irreversible battery damage, while prolonged use of parking air conditioning lacks an intelligent power scheduling mechanism, failing to maximize battery efficiency while ensuring user comfort. Summary of the Invention

[0004] This invention provides a remote control method for a parking air conditioner battery used for emergency start-up. This invention can accurately identify the dual-mode load characteristics of high current surge during emergency start-up and low current continuous discharge of the parking air conditioner, solving the problem of inaccurate load identification in traditional systems and improving battery utilization efficiency.

[0005] In a first aspect, the present invention provides a remote control method for an emergency start parking air conditioner battery, the remote control method for the emergency start parking air conditioner battery comprising: The back-end management system uses the geographic coordinate information obtained by the built-in GPS module of the emergency start-up battery to perform geofence matching and digital certificate verification to obtain the remote control permission authentication status. The battery edge computing module identifies the load mode of the emergency start-up high current impact of the emergency start-up battery and the low current continuous discharge of the parking air conditioner, and obtains the current power consumption mode classification result. The battery edge computing module performs battery state analysis based on the current power consumption pattern classification results to obtain the SOC state of charge value. The battery edge computing module performs pre-charge timing control on the BMS discharge switch of the emergency start battery in the emergency start scenario based on the SOC state of charge value to obtain the target current limit parameters. The battery edge computing module performs power level classification and time period scheduling of the BMS charging and discharging switch of the emergency start battery in the parking air conditioning scenario based on the SOC state of charge value, and obtains voltage output control parameters.

[0006] In conjunction with the first aspect, in a first implementation of the first aspect of the present invention, the background management system performs geofence matching and digital certificate verification on the geographic coordinate information obtained by the built-in GPS module of the emergency start-up battery to obtain the remote control permission authentication status, including: The background management system performs coordinate analysis and accuracy verification on the raw satellite positioning signal transmitted by the built-in GPS module of the emergency start-up battery to obtain standardized geographic coordinate data; The backend management system performs Euclidean distance calculation and boundary judgment based on the standardized geographic coordinate data and the preset elliptical geofence boundary parameters to obtain the geographic location matching verification result; The back-end management system parses and compares the hash value of the RSA-2048-bit digital certificate in the remote control command to obtain the certificate verification status code; The backend management system performs logical AND operations and timestamp validity checks based on the geographic location matching verification result and the certificate verification status code to obtain the remote control permission authentication status.

[0007] In conjunction with the first aspect, in a second implementation of the first aspect of the present invention, the battery edge computing module performs load mode identification on the emergency start-up high current impact of the emergency start-up battery and the low current continuous discharge of the parking air conditioner, and obtains the current power consumption mode classification result, including: The battery edge computing module performs digital filtering on the raw charging and discharging signals collected by the current sensor to obtain denoised current waveform data. The battery edge computing module inputs the denoised current waveform data into the fast Fourier transform algorithm for frequency domain decomposition and feature frequency extraction to obtain the frequency domain feature parameters of the current waveform. The battery edge computing module performs current slope calculation and peak detection based on the frequency domain feature parameters of the current waveform to obtain the current slope and peak current. When the current slope exceeds the slope threshold and the peak current reaches the first current range, the current power consumption mode classification result is output as emergency start large current impact. When the current slope is less than the slope threshold and the peak current is stable in the second current range, the current power consumption mode classification result is output as parking air conditioner small current continuous discharge.

[0008] In conjunction with the first aspect, in a third implementation of the first aspect of the present invention, the battery edge computing module performs battery state analysis based on the current power consumption pattern classification result to obtain the SOC (State of Charge) value, including: When the current power consumption mode classification result is an emergency start high current impact, the weight factor of the battery edge computing module is set to the first weight factor parameter; when the current power consumption mode classification result is a parking air conditioner low current continuous discharge, the weight factor is set to the second weight factor parameter. Dynamic weight factor parameters are constructed based on the first weight factor parameter and the second weight factor parameter. The battery edge computing module inputs the dynamic weight factor parameters and the state variables of the emergency start battery into the extended Kalman filter for state prediction and covariance matrix update to obtain the prior distribution parameters of the battery state. The battery edge computing module updates the posterior probability based on the prior distribution parameters of the battery state to obtain the posterior distribution data of the battery state. The battery edge computing module resamples and weights the battery state posterior distribution data to generate the SOC state of charge value.

[0009] In conjunction with the first aspect, in a fourth implementation of the first aspect of the present invention, the battery edge computing module performs pre-charging timing control on the BMS discharge switch of the emergency start battery in an emergency start scenario based on the SOC state of charge value to obtain target current limiting parameters, including: The battery edge computing module collects internal battery temperature data through a temperature sensor. When the SOC state of charge value is lower than the preset SOC threshold and the internal battery temperature data is lower than the temperature threshold, the pre-charge mode is triggered to obtain the pre-charge start condition judgment result. The battery edge computing module dynamically sets the time constant based on the pre-charge start-up condition judgment result and the battery internal temperature data to obtain the pre-charge time constant parameter; The battery edge computing module calculates the pre-charge current change trajectory data based on the pre-charge time constant parameter; Based on the pre-charge current change trajectory data, the battery edge computing module sets the upper limit of the current and divides the timing for the BMS discharge switch in the emergency start scenario to obtain the target current limit parameters.

[0010] In conjunction with the first aspect, in a fifth implementation of the first aspect of the present invention, the battery edge computing module calculates the pre-charging current change trajectory data based on the pre-charging time constant parameter, including: The battery edge computing module initializes the exponential function parameters based on the pre-charge time constant parameters to obtain the basic parameters for exponential function calculation. The battery edge computing module performs time series discretization calculations based on the basic parameters calculated by the exponential function, and generates a pre-charging time node array according to the sampling interval starting from the pre-charging start time. The battery edge computing module substitutes each time node in the pre-charging time node array into the exponential function formula for iterative calculation to obtain the pre-charging current change trajectory data.

[0011] In conjunction with the first aspect, in the sixth implementation of the first aspect of the present invention, the battery edge computing module performs power level classification and time period scheduling of the BMS charging and discharging switch of the emergency start battery in the parking air conditioning scenario based on the SOC state of charge value, and obtains voltage output control parameters, including: The battery edge computing module determines the threshold range based on the SOC state of charge value to obtain the battery state level identifier. The battery edge computing module calculates the state transition probability by analyzing the historical power consumption data and the usage time and frequency of the parking air conditioner, and obtains the power consumption behavior prediction results. The battery edge computing module performs time-of-use power scheduling based on the battery status level identifier and the power consumption behavior prediction results to obtain a time-of-use power scheduling strategy. The battery edge computing module performs BMS charge and discharge switch control parameter formatting based on the time-sharing power scheduling strategy, converting the power limit value into the corresponding voltage output control command and switch conduction sequence to obtain the voltage output control parameters.

[0012] In conjunction with the first aspect, in the seventh implementation of the first aspect of the present invention, the battery edge computing module, based on historical electricity consumption data, calculates the state transition probability by analyzing the historical usage duration and frequency of the parking air conditioner, and obtains the electricity consumption behavior prediction result, including: The battery edge computing module divides historical electricity consumption data into time windows to obtain structured historical electricity consumption sequence data; The battery edge computing module performs statistical analysis on the parking air conditioner usage pattern based on the structured historical power consumption sequence data to obtain statistical parameters of parking air conditioner usage characteristics; The battery edge computing module discretizes the state space of the parking air conditioner usage characteristic statistical parameters according to usage duration and frequency to obtain a state transition probability matrix; The battery edge computing module performs multi-step predictions based on the state transition probability matrix and the current parking air conditioning usage status to obtain the electricity consumption behavior prediction results.

[0013] In conjunction with the first aspect, in the eighth implementation of the first aspect of the present invention, the remote control method for the emergency start parking air conditioner battery further includes: The background management system encapsulates the SOC state of charge value, BMS switch status data and battery status parameters in real time to obtain a battery status data packet. The background management system remotely transmits and stores the battery status data packet through a 5G communication module and performs multi-parameter anomaly detection and fault warning generation to obtain battery status monitoring data. The back-end management system performs digital signature verification and secure encapsulation of the OTA firmware upgrade package, and sends it to the BMS controller of the battery edge computing module through the 5G communication module to obtain the OTA upgrade package sending confirmation signal; The backend management system tracks the firmware installation progress and verifies the completion status based on the confirmation signal issued by the OTA upgrade package. When it receives the installation success response returned by the battery edge computing module, it obtains the firmware update completion flag.

[0014] In conjunction with the first aspect, in the ninth implementation of the first aspect of the present invention, the background management system tracks the firmware installation progress and verifies the completion status based on the confirmation signal issued by the OTA upgrade package. When it receives the installation success response returned by the battery edge computing module, it obtains a firmware update completion flag, including: The background management system sends a confirmation signal based on the OTA upgrade package and combines it with the geographical location data of the built-in GPS module to perform geographical security verification of the upgrade process. When an abnormal change in the geographical location of the emergency start battery is detected, the upgrade process is automatically interrupted and rolled back to the original firmware version, and a geographical security verification upgrade control command is obtained. The background management system intelligently schedules the upgrade timing based on the geographic security verification upgrade control command and the current power consumption mode classification results to obtain the upgrade process control strategy. The background management system performs real-time backup and dynamic updates of the BMS charging and discharging switch during the upgrade process based on the upgrade process control strategy, and obtains the parameter synchronization results of the upgrade process. The backend management system performs post-upgrade function verification and geographic environment adaptation testing based on the synchronization results of the upgrade process parameters. When it receives a successful installation response from the battery edge computing module, it obtains a firmware update completion flag.

[0015] The technical solution provided by this invention, through the Fast Fourier Transform algorithm and current slope detection technology, can accurately identify the dual-mode load characteristics of emergency start-up high current impact and parking air conditioning low current continuous discharge, solving the problem of inaccurate load identification in traditional systems. The use of dynamic weighting factors and improved Bayesian algorithms significantly improves the estimation accuracy of SOC (State of Charge) values, and temperature-compensated exponential function pre-charge control effectively avoids damage to the battery caused by low-temperature high current. The established five-level battery state management strategy, combined with a Markov prediction model, achieves intelligent power scheduling, improving battery utilization efficiency. A triple security authentication mechanism, constructed through GPS geofence matching and RSA-2048-bit digital certificate verification, effectively prevents malicious remote control attacks, ensuring system security. The OTA (Over-The-Air) upgrade mechanism, combined with GPS geolocation verification and intelligent power consumption mode scheduling, ensures the security and reliability of the firmware upgrade process.

[0016] Other features and advantages of the invention will be set forth in the description which follows, and will be apparent in part from the description, or may be learned by practicing the invention. The objects and other advantages of the invention are realized and obtained in accordance with the structures particularly pointed out in the description, claims and drawings.

[0017] To make the above-mentioned objects, features and advantages of the present invention more apparent and understandable, preferred embodiments are described below in detail with reference to the accompanying drawings. Attached Figure Description

[0018] Figure 1 This is a schematic diagram of an embodiment of the remote control method for emergency start parking air conditioner battery in this invention. Detailed Implementation

[0019] To make the objectives, technical solutions, and advantages of the embodiments of the present invention clearer, the technical solutions 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, 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.

[0020] The terms "comprising" and "having," and any variations thereof, used in the embodiments of this invention are intended to cover non-exclusive inclusion. For example, a process, method, system, product, or device that includes a series of steps or units is not limited to the steps or units listed, but may optionally include other steps or units not listed, or may optionally include other steps or units inherent to these processes, methods, products, or devices.

[0021] To facilitate understanding of this embodiment, a remote control method for an emergency start parking air conditioner battery disclosed in this embodiment of the invention will first be described in detail. For example... Figure 1 As shown, this method includes the following steps: 101. The back-end management system performs geofence matching and digital certificate verification on the geographic coordinate information obtained by the built-in GPS module of the emergency start-up battery to obtain the remote control permission authentication status. It is understood that the executing entity of this invention can be a remote control device for an emergency start parking air conditioner battery, or it can be a terminal or a server; no specific limitation is made here. This embodiment of the invention will be described using a server as an example.

[0022] Specifically, the built-in GPS module continuously transmits satellite positioning signals to the backend management system. After receiving the original NMEA format or equivalent satellite data, the backend system performs coordinate decoding, signal-to-noise ratio evaluation, and calibration on the data to obtain standardized geographic coordinate data in latitude and longitude form. It then performs validity judgment to eliminate data errors caused by drift anomalies or multipath interference. Based on this, the backend management system calls the preset set of geofence parameters. The geofence is defined in the form of an ellipse, and the boundary parameters include the coordinates of the ellipse's center point, major axis radius, minor axis radius, and azimuth angle. By substituting the current battery location coordinates into the ellipse boundary judgment model, Euclidean spatial distance calculation is performed, and a rotation matrix is ​​introduced for direction correction to adapt to the judgment accuracy of the tilted geofence. A geographic location matching verification result is generated, which is used to determine whether the current battery is within the authorized control range. Meanwhile, the backend system performs digital identity verification on the received remote control commands. Specifically, it parses the RSA-2048-bit digital certificate field in the command, extracts the public key information and signature digest contained therein, and calculates a local digest of the command content using a hash algorithm based on a stored list of legitimate user certificates. The calculated hash value is then compared with the certificate signature content to obtain a certificate verification status code, indicating whether the current request originates from a registered and authenticated user. The backend management system performs a logical AND operation on the geolocation matching verification result and the certificate verification status code, ensuring both are passed before proceeding to the next stage of control judgment. Simultaneously, the system reads the timestamp field attached to the control command and performs a time window verification with the current system time. If the verification time does not exceed the preset maximum response delay threshold (e.g., 15 seconds), the timestamp of the request is deemed valid. Finally, the system jointly judges the location legitimacy, identity legitimacy, and timeliness, outputting the remote control authorization authentication status in a logically comprehensive manner. If all conditions are met, an authorization authentication pass flag is returned, allowing the control command to be sent to the battery-side BMS; otherwise, the control process is immediately terminated and a security event log is recorded.

[0023] 102. The battery edge computing module identifies the load mode of the emergency start-up high current impact of the emergency start battery and the low current continuous discharge of the parking air conditioner, and obtains the current power consumption mode classification result. Specifically, a current sensor performs high-speed sampling of the real-time charging and discharging current signal at the battery port, with the sampling frequency consistently set above 100Hz to ensure the ability to capture short-term pulse characteristics. The edge computing module then receives this raw current signal and performs digital filtering to remove measurement noise and environmental interference. The filtering algorithm used employs a finite impulse response filter or a Kalman filter to ensure that the time-domain signal retains its rapid changing trend while removing high-frequency noise, thus obtaining denoised current waveform data. The denoised current waveform data is then input into a fast Fourier transform module, which uses frequency domain transformation to map the current waveform from the time domain to the frequency domain. This allows for the extraction of its main characteristic frequencies and amplitude distributions. Statistical analysis of the concentrated energy segments of the spectrum generates a set of frequency domain characteristic parameters for the current waveform. Based on these parameters, the first derivative of the current change process, i.e., the current slope, is calculated to measure the rate of current rise or fall. Simultaneously, peak detection is performed to determine the maximum current value in the waveform and record its corresponding moment. After obtaining the two key indicators, slope and peak current, classification judgment is carried out according to the preset recognition logic. If the current slope exceeds the defined slope threshold (e.g., 50A / s) and the peak current is within the first current range (i.e., between 400A and 800A), the current load is determined to have short-term high current surge characteristics, and the output power consumption mode classification result is emergency start mode. Conversely, when the current slope is much lower than the threshold and the waveform is stable, the peak current is always within the second current range (e.g., between 15A and 35A), and the current is maintained in a continuous output state, the current load is determined to have low current long-term characteristics, and the output power consumption mode classification result is parking air conditioning mode.

[0024] 103. The battery edge computing module performs battery state analysis based on the current power consumption pattern classification results to obtain the SOC state of charge value; Specifically, based on the load identification module, the typical power consumption scenario of the current battery operation is determined. When the classification result is an emergency start-up high current impact, it indicates that the current load is in a high-power, short-term discharge state. In this case, the weight factor in the state estimation process is set as the first weight factor parameter to enhance the attention to rapid dynamic response parameters such as instantaneous voltage drop and internal resistance change of the battery. When the classification result is a parking air conditioner low current continuous discharge, it means that the battery is in a long-term, low-power state. The corresponding weight factor is adjusted to the second weight factor parameter to strengthen the identification ability of long-term voltage drop trend and capacity change. Then, based on the weight settings of these two typical operating conditions, a dynamic weight factor parameter is constructed. This parameter can be switched or interpolated in real time with the change of operating conditions to adapt to the continuous change of actual operating state, thereby improving the robustness and sensitivity of state estimation. Based on the constructed dynamic weighting factor parameters, the battery's internal state variables (including terminal voltage, current, temperature, internal resistance, historical SOC estimates, aging factors, etc.) are fed into the extended Kalman filter model as state inputs. This filter predicts the battery's behavior through a nonlinear state transition function and recursively updates its error covariance matrix, forming a prior distribution estimate of the battery's current operating state. Subsequently, the system incorporates actual acquired observations using a measurement update mechanism, corrects the prior distribution for posterior probability through Bayesian inference, and combines the observation residuals with Kalman gain to adjust prediction bias, generating posterior distribution data containing key states such as voltage, SOC, and SOH. To extract an actionable SOC estimate from uncertainty modeling, an importance resampling operation is performed based on this posterior distribution. This involves selecting several representative state instances from the posterior samples, removing samples with low weights, assigning normalized weights to valid samples, and finally performing a weighted average calculation on the selected posterior samples to output a real-time representative SOC state of charge value.

[0025] 104. The battery edge computing module performs pre-charging timing control on the BMS discharge switch of the emergency start battery in the emergency start scenario based on the SOC state of charge value to obtain the target current limit parameters. Specifically, a dual-condition triggering mechanism including SOC threshold and temperature threshold is established. The battery edge computing module collects temperature distribution data inside the battery module in real time through embedded multi-point temperature sensors and compares this data with the set temperature threshold. At the same time, it retrieves the current SOC state value and the preset minimum SOC operation guarantee value for joint judgment. When the SOC value is lower than the threshold and the internal temperature of the battery is also lower than the temperature critical value, such as 5°C, the pre-charge control logic is triggered and the pre-charge start condition judgment result is output. This result indicates that the current environment and battery state have entered a high-risk low temperature and low power condition range, and a soft start mechanism is required to prevent large current impact from causing structural damage to the battery. Based on the judgment results and combined with the real-time temperature values ​​collected by the temperature sensor, the battery edge computing module establishes a dynamic adjustment model according to the relationship between temperature and time response. This model calculates and generates a pre-charge time constant parameter, which controls the current growth rate. The value of this time constant increases as the temperature decreases to ensure sufficient recovery of ion activity. For example, when the temperature is below 0℃, τ is set to 8 seconds; when the temperature is between 0 and 10℃, τ is set to 5 seconds; and when the temperature is above 10℃, τ is shortened to 3 seconds. This time constant parameter is then substituted into the current growth function model to generate pre-charge current change trajectory data in the form of an exponential curve. This current trajectory exhibits a slow initial increase followed by gradual stabilization, which helps to improve electrode temperature and internal resistance characteristics in a short period while avoiding surge shocks caused by excessively rapid conduction. Based on this trajectory data, a discharge control timing structure is formulated and the maximum current limit value of the current BMS discharge path is set. During the entire pre-charging phase, the battery edge computing module will dynamically adjust the conduction width of the discharge channel or the PWM duty cycle according to the trajectory to achieve the step-by-step release from current-limited conduction to the target discharge capacity, and finally form the target current limit parameters that match the current battery state.

[0026] 105. The battery edge computing module performs power level classification and time period scheduling of the BMS charging and discharging switch of the emergency start battery in the parking air conditioning scenario based on the SOC state of charge value, and obtains voltage output control parameters.

[0027] Specifically, based on the real-time acquired State of Charge (SOC) value, a threshold range judgment is performed. By comparing the SOC value with the five-level zoning standard, the current battery status is divided into five levels: green (safe), blue (warning), yellow (alarm), orange (dangerous), and red (emergency), forming a battery status level identifier. Simultaneously, the edge computing module retrieves historical parking air conditioning power consumption data from local data cache or cloud-synchronized databases. It statistically analyzes and categorizes information such as usage time, frequency, and SOC change trends, and constructs a power consumption state transition matrix based on Markov chain modeling. This matrix then calculates the probability of user power consumption behavior transitions at different time periods, obtaining a power consumption behavior prediction result for the next few hours. This result is used to assess the air conditioning start-up time and load duration in parking scenarios. The battery edge computing module performs time-segmented power scheduling based on the battery status level identifier and the power consumption behavior prediction result. This stage integrates two input dimensions: the current State of Charge (SOC) level as the power limit and the future air conditioning demand window provided by the behavioral prediction model as the load constraint. It dynamically divides time intervals and assigns a maximum power output limit to each interval. For example, if the air conditioning is predicted to be used continuously for 4 hours while the SOC is at the yellow warning level, the system automatically sets the first 2 hours as a high-power limit segment and the last 2 hours as a power-saving segment, forming a time-segmented power scheduling strategy. After the strategy is formulated, the battery edge computing module converts the power scheduling strategy into hardware-executable control instructions. This involves parameter formatting, converting the power limit values ​​for each time period into target voltage values ​​using the voltage-current-power conversion formula, and generating PWM duty cycle data or a fixed on-time schedule based on the periodic control law of switch conduction, thereby obtaining voltage output control parameters adapted to the current BMS control unit.

[0028] In one specific embodiment, the process of performing step 101 may specifically include the following steps: The background management system performs coordinate analysis and accuracy verification on the raw satellite positioning signal transmitted by the built-in GPS module of the emergency start-up battery to obtain standardized geographic coordinate data; The back-end management system calculates Euclidean distance and determines boundaries based on standardized geographic coordinate data and preset elliptical geofence boundary parameters to obtain geographic location matching verification results. The back-end management system parses and compares the hash value of the RSA-2048-bit digital certificate in the remote control command to obtain the certificate verification status code; The backend management system performs logical AND operations and timestamp validity checks based on the geolocation verification result and certificate verification status code to obtain the remote control permission authentication status.

[0029] Specifically, the backend management system receives raw satellite positioning data from the built-in GPS module in the emergency start-up battery. This positioning data is a standard message in NMEA 0183 format, including various statement formats such as GGA and RMC. The system uses a professional parsing module to extract parameters such as UTC time, latitude, longitude, altitude, positioning accuracy factor, and number of satellites. After extraction, the backend performs coordinate system unification on the received coordinate information, converting the extracted WGS-84 format latitude and longitude coordinates into a unified plane coordinate system (such as the UTM coordinate system or the CGCS2000 system commonly used in geographic information systems) using a projection transformation algorithm. Accuracy verification is then performed by checking the "positioning status field" and HDOP (horizontal accuracy factor) to determine whether the current coordinates meet the system's position accuracy requirements. If the HDOP value exceeds the set value, the system will not accept the data. If the threshold or location status is invalid, the coordinate data is deemed unreliable, and the system will terminate the subsequent judgment process. Once the coordinate data passes the accuracy verification, the backend management system calls the preset geofence model. This model describes the physical geographic boundary that the battery can control using an elliptical region. Its boundary parameters include the coordinates of the ellipse's center point, the major axis radius, the minor axis radius, and the ellipse's direction angle θ. The system performs Euclidean distance calculation by substituting the battery's current coordinates into the elliptical region boundary formula, translates the current coordinate point with the ellipse's center as the origin, and then performs coordinate axis transformation based on the ellipse's rotation angle to convert it into standard position data in the ellipse's principal axis aligned coordinate system. This data is then substituted into the ellipse boundary equation to determine whether the spatial position constraint (x / a)² + (y / b)² ≤ 1 is satisfied, where a and b are the lengths of the major and minor axes, respectively. If the judgment result is valid, the current geographic location is considered to be within the fenced area, and the geographic location matching verification result is output as "passed". The backend system performs identity security verification on the remote control commands issued by the user. The control commands contain digital certificate information generated based on the RSA-2048 encryption algorithm. The certificate includes information such as the signing subject, encryption public key, timestamp, digest algorithm description, and signature content. The backend system calls the local trusted CA certificate store to parse the digital certificate item by item. First, it extracts the digest of the control command and calculates the hash value of the actual message. Then, it uses the public key to decrypt the signature content to recover the original digest value. If the digest value calculated by the system is consistent with the digest value obtained by decryption, it proves that the command content has not been tampered with and the certificate comes from a trusted source. At this time, the certificate verification status code "valid" is generated.Once both the geolocation verification result and the certificate verification status code are generated, the backend management system performs a logical AND operation to determine whether both verifications pass simultaneously. If either result is invalid, the remote control permission is considered invalid. To prevent control commands from being maliciously copied or replayed with delays, the system introduces a timestamp validity judgment logic. This involves comparing the standard timestamp field attached to the remote control command with the time received by the backend. If the interval between the two timestamps does not exceed a set valid window (e.g., 15 seconds), the timestamp is considered valid. Combining the aforementioned location and identity verification results, the system ultimately determines the remote control permission authentication status. If all three judgment conditions are met, a "Control Allowed" status is generated and fed back to the BMS control module, while relevant control logs are recorded. If any condition is not met, the system terminates command forwarding and sends a security warning message to the user and management platform.

[0030] In one specific embodiment, the process of performing step 102 may specifically include the following steps: The battery edge computing module performs digital filtering on the raw charging and discharging signals collected by the current sensor to obtain denoised current waveform data. The battery edge computing module inputs the denoised current waveform data into the fast Fourier transform algorithm for frequency domain decomposition and feature frequency extraction to obtain the frequency domain feature parameters of the current waveform. The battery edge computing module calculates the current slope and detects the peak value based on the frequency domain feature parameters of the current waveform to obtain the current slope and peak current. When the current slope exceeds the slope threshold and the peak current reaches the first current range, the current power consumption mode classification result is output as emergency start large current impact. When the current slope is less than the slope threshold and the peak current is stable in the second current range, the current power consumption mode classification result is output as parking air conditioner small current continuous discharge.

[0031] Specifically, digital filtering is performed on the original charging and discharging signal to remove interference and restore its intrinsic characteristics. Algorithms such as finite impulse response filters, unbiased moving average filters, or wavelet denoising methods are selected. By setting the sliding window width or frequency cutoff value, high-frequency noise components are filtered out, while the true current response represented by fast rising edges and step transitions is retained, thus generating denoised current waveform data. The current waveform is then mapped from the time domain to the frequency domain to extract its potential structural features. This step is accomplished using a fast Fourier transform algorithm. The specific process includes performing a discrete Fourier expansion on the current data within a finite window, obtaining the amplitude spectrum by squaring the real and imaginary parts, and then extracting frequency domain characteristic parameters such as the dominant frequency peak position, energy distribution range, bandwidth index, and frequency density distribution. These parameters are used to identify whether there is a large oscillation behavior in a short period of time or a continuous, slowly changing, stable trend. The system calls the slope detection module to calculate the slope of the current change, that is, to perform a first-order difference operation on the current curve at each sampling point, analyze the current change amplitude per unit time, and compare it with the set threshold. The slope calculation result serves as an important basis for identifying sudden high current behavior. At the same time, the system performs peak detection processing, that is, to extract the maximum current value from the entire waveform and determine whether it is within a specific range. This range is defined as the high current output range required for emergency start-up scenarios, between 400A and 800A. The system uses two key indicators, current slope and peak current, as joint criteria for logical judgment. If the detection result shows that the slope value has exceeded the preset threshold (e.g., 50A / s) and the peak current is within the first current range mentioned above, it is confirmed that the current load behavior is an emergency start mode, which is a typical high current surge process. At this time, the battery edge computing module immediately sets the current power consumption mode classification result as "emergency start". Conversely, if the current slope calculation result remains at a low level for a long time, it indicates that the current output is stable and changes slowly. At the same time, the peak current in the waveform is stable in the second current range for a long time, such as between 15A and 35A, and the duration meets the set threshold. If it is maintained for more than 300 seconds, it indicates that the current load behavior is a parking air conditioning mode, which is a typical small current continuous discharge process. At this time, the edge computing module outputs the classification result as "parking air conditioning".

[0032] In one specific embodiment, the process of performing step 103 may specifically include the following steps: When the current power consumption mode classification result is emergency start high current impact, the weight factor of the battery edge computing module is set to the first weight factor parameter; when the current power consumption mode classification result is parking air conditioner low current continuous discharge, the weight factor is set to the second weight factor parameter. Dynamic weight factor parameters are constructed based on the first and second weight factor parameters. The battery edge computing module inputs the dynamic weight factor parameters and the state variables of the emergency start battery into the extended Kalman filter for state prediction and covariance matrix update, and obtains the prior distribution parameters of the battery state. The battery edge computing module updates the posterior probability based on the prior distribution parameters of the battery state to obtain the posterior distribution data of the battery state. The battery edge computing module resamples and weights the battery state posterior distribution data to generate the SOC state of charge value.

[0033] Specifically, the power consumption pattern recognition result is directly used as the conditional input for state estimation. Within the edge computing module, when the current power consumption pattern classification result is "emergency start," it indicates that the system is in a state of short-term high load, high discharge current, and high power density. Under this condition, the influence of physical characteristics reflecting the battery's strong impulse response capability, such as terminal voltage, internal resistance, and voltage recovery speed, on state estimation is significantly enhanced. The system then sets the weight factor participating in the state update calculation as the first weight factor parameter. This parameter is set to a high value to prioritize the prior sensitivity of voltage drop amplitude and response rate. Conversely, when the power consumption pattern recognition result is "parking air conditioning," it indicates that the battery is in a state of long-term low-current stable discharge. In the case of a slow discharge, the system switches the weighting factor to a second weighting factor parameter. This parameter is used to highlight the importance of capacity stability, temperature control level, and the slow decrease trend of SOC during the slow discharge process, thus forming a weighting factor decision structure driven by load type. In actual operation, since the operating conditions are often in a transitional state between the two modes, the system does not use a single weighting factor. Instead, it uses additional judgment mechanisms such as pattern recognition reliability, waveform similarity, and timing state switching window to perform linear interpolation, exponential weighting, or proportional modulation on the first and second weighting factors to construct a dynamic weighting factor parameter with continuity and gradual change, thereby eliminating the impact of sudden changes in operating conditions on the convergence stability of the filter state. The battery edge computing module inputs the dynamic weighting factor and the collected battery operating state variables into an extended Kalman filter for nonlinear state recursive estimation. These state variables include battery terminal voltage, current, temperature, internal resistance, estimated state of charge, historical charge / discharge power density, and temperature slope. The filter performs state prediction on these variables. Under the control of the state transition model, it performs forward extrapolation of the current battery state based on the system input and prior state, while updating the error covariance matrix and correcting the measurement noise propagation range. This outputs an uncertain prior estimate describing the state distribution, obtaining the prior distribution parameters of the battery state. After completing the prior update, the system calculates the Kalman gain based on the deviation difference (residual) between the current observation and the predicted value. This gain is then multiplied by the residual and used to correct the state vector, thus completing the a posteriori update process. In other words, based on the Bayesian inference mechanism, the predicted and observed information are fused to generate the posterior distribution data of the battery state, where each state vector sample contains a state value and its corresponding weight. Since the output of the extended Kalman filter is a weighted and uncertain sample distribution, direct use may result in multimodality, skewness, or variance drift. Therefore, the system introduces a resampling mechanism, which extracts several high-weight samples from the posterior distribution for reconstruction and renormalizes them to remove abnormal weights or over-aggregated regions, thus avoiding sample degradation. Finally, the resampled sample groups are summed and solved according to the weighted average principle to generate the SOC value corresponding to the current time.

[0034] In one specific embodiment, the process of performing step 104 may specifically include the following steps: The battery edge computing module collects internal battery temperature data through a temperature sensor. When the SOC state of charge value is lower than the preset SOC threshold and the internal battery temperature data is lower than the temperature threshold, the pre-charge mode is triggered, and the pre-charge start condition judgment result is obtained. The battery edge computing module dynamically sets the time constant based on the pre-charge start-up condition judgment result and the battery internal temperature data to obtain the pre-charge time constant parameter. The battery edge computing module calculates the pre-charge current change trajectory data based on the pre-charge time constant parameter. Based on the pre-charge current change trajectory data, the battery edge computing module sets the current upper limit and divides the timing of the BMS discharge switch in the emergency start scenario to obtain the target current limit parameters.

[0035] Specifically, multiple temperature acquisition points are embedded in the battery pack structure, and a thermal coupling model is used to ensure that the local temperature of each cell module can be sensed in real time and accurately. During normal operation, the battery edge computing module continuously reads the output data of these temperature sensors and performs validity verification and temperature consistency analysis. At the same time, it executes the SOC state of charge estimation logic in parallel to obtain the current remaining usable power index of the battery. When the detection result shows that the current SOC is lower than the system's preset minimum state of charge threshold (e.g., 25%), and the temperature value of any one or more temperature acquisition points is lower than the system's set low temperature threshold (e.g., 5℃), it is determined that the battery environment has entered a high-risk operating condition area of ​​low power + low temperature. At this time, the edge computing module sends a pre-charge mode trigger signal and forms a pre-charge start condition judgment result. Based on this judgment, the system initiates the pre-charge control branch logic and constructs a time constant setting model with the currently detected lowest battery temperature as the main input variable. This model dynamically correlates the temperature value with the time constant τ through a mapping function; that is, the lower the temperature, the larger the time constant, to achieve the slow-start characteristic of the current release process. Specifically, a piecewise function is used. For example, when the temperature is below 0℃, τ is set to 8 seconds; when the temperature is between 0 and 10℃, τ is 5 seconds; and when the temperature is above 10℃, τ is compressed to 3 seconds, thus ensuring that the system can maintain a reasonable charging preheating cycle under different thermal environments. After obtaining the time constant parameters, the system establishes a pre-charge current change trajectory according to a predefined current growth model. This trajectory presents a non-linear, monotonically increasing curve. In the initial stage, the current increases slowly, then gradually rises exponentially, approaching the rated current. This ensures that the cell material activity is gradually activated by a small current at low temperatures, reducing internal resistance and increasing the ion diffusion rate, avoiding irreversible damage caused by the instantaneous impact of a large current during cold starts. The battery edge computing module uses the current trajectory as a timing control reference input to the BMS discharge control logic. Based on the trajectory data, it programs the discharge switch conduction behavior, that is, it sets a target range in which the upper limit of the current gradually increases over time within the control cycle, and dynamically adjusts the conduction ratio through high-frequency PWM modulation to adapt to changes in the target current value, or sets the current threshold boundary through a simulated current source current limiting mechanism, thereby forming a control mechanism in which the upper limit of the current changes automatically over time. The entire conduction cycle is divided into several current release segments according to the trajectory growth trend. Within each segment, a conduction strategy is set by jointly setting a fixed upper limit of current value and duration parameters to ensure that the current release process is gradual and controllable.With the combined effect of trajectory control and timing partitioning, the battery edge computing module generates target current limiting parameters. This parameter set includes the initial conduction current value, maximum limiting current, rise time, conduction rhythm, and cutoff condition. These parameters are written into the BMS driver layer control command channel in real time to directly drive the duty cycle modulation and soft conduction response of the discharge power switch. This enables a gradual pre-charge process before emergency start-up discharge, thereby reducing the risk of lithium dendrite puncture, electrode material structure collapse, and system thermal imbalance caused by cold-state impact current.

[0036] In one specific embodiment, the process by which the battery edge computing module calculates the pre-charging current change trajectory data based on the pre-charging time constant parameter can specifically include the following steps: The battery edge computing module initializes the exponential function parameters based on the pre-charge time constant parameter to obtain the basic parameters for exponential function calculation. The battery edge computing module performs time series discretization calculations based on the basic parameters calculated by the exponential function, and generates an array of pre-charging time nodes according to the sampling interval starting from the start of pre-charging. The battery edge computing module substitutes each time node in the pre-charging time node array into the exponential function formula for iterative calculation to obtain the pre-charging current change trajectory data.

[0037] Specifically, the battery edge computing module calls the pre-charge time constant parameter τ output by the temperature judgment and SOC evaluation modules from the previous stage. This parameter directly reflects the battery's thermal response capability under low-temperature conditions. Based on the pre-charge time constant parameter, the exponential function parameter initialization settings are performed. The edge computing module automatically sets the starting value, target upper limit value, and maximum allowable time range of the entire pre-charge process according to the value of τ. The edge computing module establishes a time discretization model aligned with the control cycle. This model generates an array of time nodes starting from the beginning of pre-charge. Assuming the system sampling period is set to 0.1 seconds, 900 time nodes will be generated within a 90-second pre-charge cycle. The construction of this array also needs to consider synchronization requirements, i.e., it needs to match the execution cycle of the BMS's PWM controller main loop or current limiter to avoid current abrupt changes or response lag caused by control cycle misalignment. After the time node array is constructed, the battery edge computing module iterates through each time point sequentially, calling the exponential function iterator to substitute the current node into the preset current growth model for point-by-point calculation to obtain the theoretical current value at the corresponding time. The current trajectory data generated in this process constitutes a one-dimensional array of current changes. This current change curve is used in actual control to guide the dynamic current limiting behavior of the discharge circuit. Based on this trajectory, the system executes a current limiting command on the discharge switch in each sampling period, ensuring its conduction capacity matches the current target current. This prevents the battery from experiencing large current surges at low temperatures, thereby reducing the probability of high-risk events such as electrode desorption, lithium dendrite growth, or localized electrolyte crystallization. The slope trend of the current trajectory data is used to determine whether the battery temperature rise is normal. If the current rise is significantly lower than the model curve within the expected time period, it indicates abnormal thermal resistance or uneven electrode cooling within the battery, triggering a protection mechanism to extend the pre-charge time or reduce the target current limit. Conversely, if the current rise is too rapid and the temperature rise exceeds the safety threshold, a temperature-controlled power limiting command is triggered, achieving active derating control. Through these steps, the final current change trajectory data is generated.

[0038] In one specific embodiment, the process of performing step 105 may specifically include the following steps: The battery edge computing module determines the threshold range based on the SOC state of charge value to obtain the battery state level identifier; The battery edge computing module calculates the state transition probability by analyzing the historical power consumption data and the usage time and frequency of the parking air conditioner, and obtains the power consumption behavior prediction results. The battery edge computing module performs time-of-use power scheduling based on battery status level identification and power consumption behavior prediction results to obtain a time-of-use power scheduling strategy. The battery edge computing module formats the BMS charging and discharging switch control parameters based on the time-sharing power scheduling strategy, converting the power limit value into the corresponding voltage output control command and switch conduction sequence to obtain the voltage output control parameters.

[0039] Specifically, the battery edge computing module obtains the current SOC (State of Charge) value of the battery in real time and compares it with multiple preset power threshold ranges in the system. The threshold ranges are divided into five levels: green (safe) above 80%, blue (warning) between 60% and 80%, yellow (warning) between 40% and 60%, orange (danger) between 20% and 40%, and red (emergency) below or equal to 20%. After completing the comparison, the system will output the current battery status level. The edge computing module calls upon parking air conditioner usage behavior data recorded in the local historical data cache and performs statistical analysis based on the historical cumulative on-time, single usage duration, activation frequency, and correlation with geographical location and ambient temperature. Based on this, a Markov state transition model is introduced to predict future air conditioner usage behavior. Specifically, it calculates the probability transition path from "unused" to "potentially used" and "definitely used" states within each time period, forming a 24-hour electricity consumption behavior prediction matrix. The prediction probability for each hour represents the risk level of the air conditioner load being activated within that time period, thus obtaining hourly electricity consumption behavior prediction results. The battery edge computing module combines the battery status level identifier and the electricity consumption behavior prediction results as scheduling conditions input to the time-sharing power management logic. The battery status level identifier is used to constrain the maximum power limit within a certain time period, while the behavior prediction results are used to construct the electricity demand intensity identifier for that time period. Therefore, during the scheduling strategy generation process, the system divides 24 hours into multiple management cycles, for example, each cycle is 1 hour, and sets three parameters for each cycle: maximum allowable power, priority scheduling flag, and power limit correction factor. If the predicted probability of behavior is high during a certain period but the SOC status level is "red emergency level", then that period will be set to "output prohibited" state. If the predicted probability is moderate during a certain period but the SOC status is "blue warning level", then the system will output limited power, for example, only allowing 80% of the rated power to be output. When the predicted probability is high and the SOC level is "green safety level", then full power output is allowed, thus forming a time-sharing power scheduling strategy that takes into account both load response capability and remaining power guarantee. After the scheduling strategy is formed, the battery edge computing module converts it into specific hardware execution parameters and starts the BMS control parameter formatting processing module. This module converts the maximum power limit value in each period into the expected output target of voltage and current. Through the conversion relationship between power, voltage, and current, the corresponding target voltage range is obtained. Then, based on this voltage range, conduction timing data for controlling the on and off behavior of the power switch is generated.Simultaneously, the system constructs a redundancy management mechanism to continuously monitor whether the current output conforms to the preset trajectory during the actual conduction cycle. If a deviation exceeds the safety tolerance range, such as the output current exceeding the set upper limit by more than 5%, the conduction reduction mechanism is automatically triggered to reduce the duty cycle or shorten the on / off cycle, thereby maintaining consistency between output behavior and scheduling strategy during the execution of the strategy. The final generated voltage output control parameters include the target voltage value, as well as conduction rhythm, duration, redundancy judgment logic, and abnormal protection strategy, covering the power management requirements required at different load stages during the use of the parking air conditioner.

[0040] In one specific embodiment, the process by which the battery edge computing module calculates the state transition probability based on historical electricity consumption data and analyzes the historical usage duration and frequency of the parking air conditioner to obtain the electricity consumption behavior prediction result can specifically include the following steps: The battery edge computing module divides historical electricity consumption data into time windows to obtain structured historical electricity consumption sequence data; The battery edge computing module performs statistical analysis on the usage patterns of the parking air conditioner based on structured historical power consumption sequence data, and obtains statistical parameters of the usage characteristics of the parking air conditioner. The battery edge computing module discretizes the state space of the parking air conditioner usage characteristic statistical parameters according to usage duration and frequency to obtain the state transition probability matrix; The battery edge computing module performs multi-step predictions based on the state transition probability matrix and the current parking air conditioning usage status to obtain the electricity consumption behavior prediction results.

[0041] Specifically, the battery edge computing module archives all charging and discharging behaviors related to the parking air conditioner recorded during battery operation and organizes them on a daily, weekly, or monthly basis. The raw power consumption data includes start time, end time, corresponding current value, power level, ambient temperature, and location tags. The system segments this data according to a unified time window, such as by hour, half-hour, or 15-minute segment, and labels each time window with whether the air conditioner is on, its duration, its state of charge (SOC) level, and the ambient temperature, forming a structured historical power consumption sequence dataset composed of timestamps, active power, load status, and environmental variables. The battery edge computing module initiates the parking air conditioner usage behavior modeling process, extracting typical usage pattern parameters through statistical analysis of the aforementioned historical sequence data. Key analysis indicators include: average daily air conditioner usage duration, distribution of typical activation time periods (e.g., usage frequency from 0:00 to 6:00 AM), high-frequency usage days within the week (e.g., weekends or holidays), average duration of a single use, maximum continuous usage duration, high-frequency usage locations, and the correlation between temperature and usage probability. The system uses sliding window technology to overlay and analyze behaviors across multiple time periods, thereby obtaining global and local usage trend distribution statistics and constructing a set of characteristic statistical parameters. These parameters include statistical descriptive quantities such as air conditioning usage probability curves, continuous usage interval distribution, peak daily usage periods, and daytime fluctuation amplitudes, reflecting user operating habits and typical scenario changes related to parking air conditioning. The battery edge computing module converts these parking air conditioning usage characteristic statistical parameters into discrete state space input variables. The system uses usage duration and frequency as core variables for spatial discretization. For example, single usage duration is divided into four intervals: 0-1 hour, 1-2 hours, 2-4 hours, and more than 4 hours; usage frequency is divided into three levels: low frequency (less than once a day), medium frequency (12 times a day), and high frequency (more than twice a day). This constructs a 12-grid state space model, where each grid point represents a typical usage state. The system then analyzes historical behavior to record state transition frequencies between these grid points, noting the number of times a state transitions from one point to another within adjacent time periods or dates. A state transition probability matrix is ​​then calculated through normalization. Each element in this matrix represents the probability of transitioning from one usage state to another, reflecting the behavioral evolution of the parking air conditioner under different usage conditions. This probability matrix is ​​constructed using the Markov chain concept, assuming that the current state is sufficient to describe the conditional distribution of future states, rather than relying on more distant historical states. This model transforms the entire usage behavior evolution into a probabilistic transition process between state sequences.The battery edge computing module reads the current usage status of the parking air conditioner as the initial condition for prediction, determines the position of the current state in the discrete state space, and uses this as the starting state to call the state transition probability matrix for forward evolution calculation. The system sets the prediction step size to 3, 6, or 24 steps, corresponding to the behavior prediction range of the next 1 hour, 3 hours, or 24 hours. In each iteration, the system propagates the current state to all possible subsequent states and updates the prediction results according to the weighted transition probabilities, finally obtaining a time series-form electricity consumption behavior prediction array. Each element in the array represents the probability of the air conditioner load being activated in a certain period of the future.

[0042] In one specific embodiment, the method for remotely controlling the parking air conditioner battery for emergency start-up further includes the following steps: The background management system encapsulates real-time data of SOC state of charge value, BMS switch status data and battery status parameters to obtain battery status data package; The back-end management system remotely transmits and stores battery status data packets through a 5G communication module, while simultaneously performing multi-parameter anomaly detection and fault warning generation to obtain battery status monitoring data. The back-end management system performs digital signature verification and secure encapsulation of the OTA firmware upgrade package, and sends it to the BMS controller of the battery edge computing module through the 5G communication module to obtain the OTA upgrade package sending confirmation signal; The backend management system tracks the firmware installation progress and verifies the completion status based on the confirmation signal sent by the OTA upgrade package. When it receives the installation success response from the battery edge computing module, it obtains the firmware update completion flag.

[0043] Specifically, the battery edge computing module periodically pushes comprehensive operational information to the backend management system, including SOC state of charge (SOC) values, BMS switch status data, and other key battery operating parameters. This information encompasses multiple dimensions such as terminal voltage, battery current, temperature distribution, internal resistance changes, historical cycle count, current load power, cell balancing status, and battery location and ambient temperature. Upon receiving this raw data stream, the backend management system performs field validation, timestamp alignment, communication number binding, and redundant field removal according to a preset data frame structure. All valid fields are packaged and encapsulated in a unified format to form a battery status data packet. The backend management system then transmits this battery status data packet uplink via a built-in 5G communication module. This communication link offers advantages such as low latency, high bandwidth, and connection stability, supporting data reporting at a rate of one second. After data transmission, the system simultaneously writes the data packet to the cloud-based battery operation database and performs structured indexing and storage based on fields such as vehicle ID, battery number, upload time, and operating environment. Simultaneously, the backend system inputs the received battery status data packets into the multi-parameter anomaly detection module. It employs statistical models, rule matching, and machine learning algorithms to perform cross-analysis of key indicators, such as detecting abnormal SOC decline rates, current surges exceeding thresholds, voltage fluctuations exceeding limits, localized temperature changes, and internal resistance increases conforming to the aging curve. If any single or combined indicator triggers a risk rule, a corresponding fault warning is immediately generated and recorded in the alarm system using a multi-level alarm mechanism. Simultaneously, the alarm information is pushed to remote monitoring terminals or maintenance personnel's mobile devices, achieving full lifecycle monitoring and early warning management of the battery's remote operating status. The final output is battery status monitoring data. After ensuring stable battery operation and meeting upgrade requirements, the backend management system initiates the OTA firmware update process. The OTA firmware upgrade package to be distributed undergoes digital signature processing. After generating a digest using the SHA256 hash algorithm, it is encrypted and signed using the system's private key. The signature file and the upgrade package are then combined into a secure encapsulation to ensure that the data is not tampered with or forged during transmission. The backend management system sends the packaged OTA upgrade package to the edge computing module of the target battery through the 5G communication channel and monitors the communication port of its BMS controller in real time. After successful delivery, the system waits for the BMS to return an upgrade package delivery confirmation signal. This signal includes a reception completion flag, file integrity verification result, and installation preparation status. Through this signal, the backend confirms that the upgrade package has been successfully delivered to the target control terminal and is ready to enter the installation process.Upon receiving the confirmation signal, the backend management system enters the OTA installation tracking phase. Based on the status data fed back by the BMS controller, the system records the firmware installation progress and verifies periodic receipts. For example, the system can require a periodic receipt or installation status code to be uploaded every 10% of the progress, and monitor the receipt intervals to ensure that the controller does not lose connection, the upgrade logic is not interrupted, and storage resources are not exceeded during the upgrade process. It also captures any abnormal states that occur during installation, such as installation failures, verification errors, or rollback triggers. Once the BMS controller completes all firmware writing, restart verification, and returns an installation success response data frame, the backend system immediately records the completion status of this upgrade task, generates a firmware update completion flag, and records the update event information along with the previous and subsequent firmware version numbers in the system upgrade log and battery maintenance history, achieving closed-loop control of the entire remote OTA upgrade task process.

[0044] In one specific embodiment, the execution step of the background management system to track firmware installation progress and verify completion status based on the confirmation signal issued by the OTA upgrade package, and the process of obtaining the firmware update completion flag when receiving the installation success response returned by the battery edge computing module can specifically include the following steps: The back-end management system performs geographic security verification of the upgrade process based on the confirmation signal sent by the OTA upgrade package and the geographic location data of the built-in GPS module. When an abnormal change in the geographic location of the emergency start battery is detected, the upgrade process is automatically interrupted and rolled back to the original firmware version, and a geographic security verification upgrade control command is obtained. The back-end management system intelligently schedules the upgrade timing based on the geographical security verification upgrade control command and the current power consumption mode classification results, thus obtaining the upgrade process control strategy. The background management system performs real-time backup and dynamic updates of the BMS charging and discharging switch during the upgrade process based on the upgrade process control strategy, and obtains the parameter synchronization results of the upgrade process. The backend management system performs post-upgrade function verification and geographic environment adaptation tests based on the parameter synchronization results during the upgrade process. When it receives a successful installation response from the battery edge computing module, it obtains a firmware update completion flag.

[0045] Specifically, upon successful OTA upgrade package delivery and receipt of confirmation response from the BMS controller, a geographic security verification mechanism is initiated to ensure that the current spatial location of the battery meets the requirements of the safe upgrade area. This mechanism reads the location information uploaded in real time by the battery's built-in GPS module and dynamically compares the current coordinates with the legal geofence range set in the background. The geofence boundary can be circular, rectangular, or elliptical, and can be flexibly set according to the needs of the scenario. When the system determines that the battery's current geographic location is stable within the authorized geofence, the upgrade process continues. However, if the system detects that the battery is moving at high speed (e.g., the vehicle is in operation) or its location crosses the boundary into an unauthorized area, such as crossing borders, entering a communication blind spot, the factory test area, or approaching a restricted area, the background system will immediately trigger the upgrade interruption mechanism. The system will stop the BMS installation process, close the upgrade execution thread, clear any incompletely written data, and automatically roll back the BMS to the previously recorded original firmware version, thereby minimizing abnormal states caused by power outages, environmental changes, or control failures during the upgrade process. The output is a geographic security verification upgrade control command. The backend management system logically integrates the geographic security verification upgrade control command with the power consumption mode classification results uploaded by the current edge computing module to identify the current battery load status, including whether an emergency start is underway, whether the air conditioning is continuously running while the vehicle is parked, whether it is in standby mode, or whether it is in the communication interruption recovery phase. Based on these operating modes, the system dynamically plans the OTA upgrade window, forming a safe, reliable upgrade scheduling strategy that does not affect the vehicle's core functions. For example, when the current power consumption mode is detected as "emergency start" or a high-frequency start-up period is predicted within 2 hours, the system postpones the upgrade plan and sets a delay time limit; if the vehicle is detected to be parked for a long time and under the conditions of air conditioning off and stable SOC, the upgrade plan is immediately triggered, thereby generating an intelligent decision-driven upgrade process control strategy. Based on the upgrade process control strategy, the background management system performs real-time backup of key BMS parameters and prepares dynamic control commands to ensure the consistency and dynamic protection of the BMS charging and discharging switch logic throughout the upgrade process. The system uniformly backs up the original BMS charging limits, discharging thresholds, voltage closed-loop control parameters, temperature control current limiting parameters, and soft power-off response strategies, and synchronizes the current state in real time during the upgrade writing process. Whenever there is a jump segment or upgrade partition switching in the control strategy, the system automatically calls the corresponding parameter set to quickly restore the control state, avoiding control abnormalities or false triggering of the fuse mechanism due to parameter mismatch, logic interruption, or protection failure during the upgrade process.The system also supports dynamic updates for certain parameters on a per-item basis. For example, if the upgrade package contains BMS soft-start logic or voltage closed-loop regulation strategy optimization, the system will map the old parameters to the new parameters and use sliding synchronization, field replacement or function overriding to complete the transition update of key control parameters, forming an upgrade process parameter synchronization result that maintains BMS control stability while gradually completing strategy replacement. Once the upgrade package is written and a phased feedback receipt is received from the BMS controller, the backend management system enters the functional verification phase. Based on the synchronization results of the upgrade process parameters, the system calls the upgraded function activation interface to perform static functional verification. This involves checking whether the new firmware has a complete file structure, whether all preset service modules are loaded normally, whether the logical channel is successfully established, and whether the command response latency is within the threshold range. Then, the system enters the dynamic geographic environment adaptation test. The system will read the GPS module data again and match the strategy branch according to the current environment of the battery. For example, if the vehicle is stationary in the city, it verifies whether the air conditioning load adjustment strategy is effective; if it is in a mountainous or low-temperature environment, it verifies whether the temperature control current limiting logic is successfully enabled. During the verification process, the system sends instruction packets to the BMS controller, such as reading the SOC estimation logic, monitoring the discharge response, current jump protection behavior, and voltage callback strategy to see if they are consistent with the upgrade target. When all function calls return valid responses and key indicators such as voltage, current, and temperature are within the control range, the system considers the upgrade to be successful. Finally, upon receiving the installation success response instruction from the battery edge computing module, the system generates and records the firmware update completion flag.

[0046] In this embodiment of the invention, frequency domain decomposition and feature frequency extraction are performed using a Fast Fourier Transform algorithm. Combined with current slope calculation and peak detection technology, this method can accurately distinguish between two different load modes: emergency start-up high current surge and parking air conditioning low current continuous discharge. This solves the technical problem that traditional systems cannot accurately identify dual-mode load characteristics. Dynamic weighting factors and an improved Bayesian algorithm are employed to adaptively adjust weight parameters based on the classification results of different power consumption modes. Extended Kalman filters and importance resampling techniques significantly improve the estimation accuracy and reliability of the State of Charge (SOC) value. A pre-charging timing control mechanism based on temperature sensor data is established. An exponential function current control algorithm is used to achieve gradual current growth, effectively avoiding damage to the battery caused by high emergency start-up current in low-temperature environments and extending battery life. A five-level battery state management strategy is established based on the SOC value. Combined with a Markov state transition probability model to predict power consumption behavior, time-sharing power scheduling and voltage output level control are achieved in the parking air conditioning scenario, improving battery utilization efficiency. A triple security authentication mechanism, employing GPS geofence matching, RSA-2048-bit digital certificate verification, and timestamp validity checks, effectively prevents malicious remote control attacks and ensures the security of remote battery management. A security verification mechanism for the upgrade process based on GPS geographic location data is established, and intelligent scheduling of upgrade timing is performed in conjunction with power consumption pattern classification results. This ensures that firmware upgrades do not affect the normal functioning of the battery, improving the system's intelligent management level. Through a 5G communication module and data compression algorithms, the effective monitoring cycle is controlled within 5 seconds, enabling real-time monitoring of battery status and fault early warning, thus improving the system's response speed and reliability.

[0047] Those skilled in the art will clearly understand that, for the sake of convenience and brevity, the specific working processes of the systems and units described above can be referred to the corresponding processes in the foregoing method embodiments, and will not be repeated here.

[0048] If the integrated unit is implemented as a software functional unit and sold or used as an independent product, it can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of the present invention, in essence, or the part that contributes to the prior art, or all or part of the technical solution, can be embodied in the form of a software product. This computer software product is stored in a storage medium and includes several instructions to cause a computer device (which may be a personal computer, server, or network device, etc.) to execute all or part of the steps of the methods described in the various embodiments of the present invention. The aforementioned storage medium includes various media capable of storing program code, such as USB flash drives, portable hard drives, read-only memory (ROM), random access memory (RAM), magnetic disks, or optical disks.

[0049] The above-described embodiments are only used to illustrate the technical solutions of the present invention, and are not intended to limit it. Although the present invention has been described in detail with reference to the foregoing embodiments, those skilled in the art should understand that modifications can still be made to the technical solutions described in the foregoing embodiments, or equivalent substitutions can be made to some of the technical features. Such modifications or substitutions do not cause the essence of the corresponding technical solutions to deviate from the spirit and scope of the technical solutions of the embodiments of the present invention.

Claims

1. A remote control method for a parking air conditioner battery used for emergency starting, characterized in that, include: The back-end management system uses the geographic coordinate information obtained by the built-in GPS module of the emergency start-up battery to perform geofence matching and digital certificate verification to obtain the remote control permission authentication status. The battery edge computing module identifies the load mode of the emergency start-up high current impact of the emergency start-up battery and the low current continuous discharge of the parking air conditioner, and obtains the current power consumption mode classification result. The battery edge computing module performs battery state analysis based on the current power consumption pattern classification results to obtain the SOC state of charge value. The battery edge computing module performs pre-charge timing control on the BMS discharge switch of the emergency start battery in the emergency start scenario based on the SOC state of charge value to obtain the target current limit parameters. The battery edge computing module performs power level classification and time period scheduling of the BMS charging and discharging switch of the emergency start battery in the parking air conditioning scenario based on the SOC state of charge value, and obtains voltage output control parameters.

2. The remote control method for the emergency start parking air conditioner battery according to claim 1, characterized in that, The backend management system performs geofence matching and digital certificate verification on the geographic coordinate information obtained by the built-in GPS module of the emergency start-up battery to obtain the remote control permission authentication status, including: The background management system performs coordinate analysis and accuracy verification on the raw satellite positioning signal transmitted by the built-in GPS module of the emergency start-up battery to obtain standardized geographic coordinate data; The backend management system performs Euclidean distance calculation and boundary judgment based on the standardized geographic coordinate data and the preset elliptical geofence boundary parameters to obtain the geographic location matching verification result; The backend management system parses and compares the hash value of the RSA-2048-bit digital certificate in the remote control command to obtain the certificate verification status code; The backend management system performs logical AND operations and timestamp validity checks based on the geographic location matching verification result and the certificate verification status code to obtain the remote control permission authentication status.

3. The remote control method for the emergency start parking air conditioner battery according to claim 1, characterized in that, The battery edge computing module performs load mode identification on the emergency start-up high current surge of the emergency start-up battery and the low current continuous discharge of the parking air conditioner, and obtains the current power consumption mode classification result, including: The battery edge computing module performs digital filtering on the raw charging and discharging signals collected by the current sensor to obtain denoised current waveform data. The battery edge computing module inputs the denoised current waveform data into the fast Fourier transform algorithm for frequency domain decomposition and feature frequency extraction to obtain the frequency domain feature parameters of the current waveform. The battery edge computing module performs current slope calculation and peak detection based on the frequency domain feature parameters of the current waveform to obtain the current slope and peak current. When the current slope exceeds the slope threshold and the peak current reaches the first current range, the current power consumption mode classification result is output as emergency start large current impact. When the current slope is less than the slope threshold and the peak current is stable in the second current range, the current power consumption mode classification result is output as parking air conditioner small current continuous discharge.

4. The remote control method for the emergency start parking air conditioner battery according to claim 1, characterized in that, The battery edge computing module performs battery state analysis based on the current power consumption pattern classification results to obtain the SOC (State of Charge) value, including: When the current power consumption mode classification result is an emergency start high current impact, the weight factor of the battery edge computing module is set to the first weight factor parameter; when the current power consumption mode classification result is a parking air conditioner low current continuous discharge, the weight factor is set to the second weight factor parameter. Dynamic weight factor parameters are constructed based on the first weight factor parameter and the second weight factor parameter. The battery edge computing module inputs the dynamic weight factor parameters and the state variables of the emergency start battery into the extended Kalman filter for state prediction and covariance matrix update to obtain the prior distribution parameters of the battery state. The battery edge computing module updates the posterior probability based on the prior distribution parameters of the battery state to obtain the posterior distribution data of the battery state. The battery edge computing module resamples and weights the battery state posterior distribution data to generate the SOC state of charge value.

5. The remote control method for the emergency start parking air conditioner battery according to claim 1, characterized in that, The battery edge computing module performs pre-charge timing control on the BMS discharge switch of the emergency start battery in an emergency start scenario based on the SOC state of charge value, and obtains target current limit parameters, including: The battery edge computing module collects internal battery temperature data through a temperature sensor. When the SOC state of charge value is lower than the preset SOC threshold and the internal battery temperature data is lower than the temperature threshold, the pre-charge mode is triggered to obtain the pre-charge start condition judgment result. The battery edge computing module dynamically sets the time constant based on the pre-charge start-up condition judgment result and the battery internal temperature data to obtain the pre-charge time constant parameter; The battery edge computing module calculates the pre-charge current change trajectory data based on the pre-charge time constant parameter; Based on the pre-charge current change trajectory data, the battery edge computing module sets the upper limit of the current and divides the timing for the BMS discharge switch in the emergency start scenario to obtain the target current limit parameters.

6. The remote control method for the emergency start parking air conditioner battery according to claim 5, characterized in that, The battery edge computing module calculates the pre-charge current change trajectory data based on the pre-charge time constant parameter, including: The battery edge computing module initializes the exponential function parameters based on the pre-charge time constant parameters to obtain the basic parameters for exponential function calculation. The battery edge computing module performs time series discretization calculations based on the basic parameters calculated by the exponential function, and generates a pre-charging time node array according to the sampling interval starting from the pre-charging start time. The battery edge computing module substitutes each time node in the pre-charging time node array into the exponential function formula for iterative calculation to obtain the pre-charging current change trajectory data.

7. The remote control method for the emergency start parking air conditioner battery according to claim 1, characterized in that, The battery edge computing module performs power level classification and time period scheduling of the BMS charging and discharging switch of the emergency start battery in the parking air conditioning scenario based on the SOC state of charge value, and obtains voltage output control parameters, including: The battery edge computing module determines the threshold range based on the SOC state of charge value to obtain the battery state level identifier. The battery edge computing module calculates the state transition probability by analyzing the historical power consumption data and the usage time and frequency of the parking air conditioner, and obtains the power consumption behavior prediction results. The battery edge computing module performs time-of-use power scheduling based on the battery status level identifier and the power consumption behavior prediction results to obtain a time-of-use power scheduling strategy. The battery edge computing module performs BMS charge and discharge switch control parameter formatting based on the time-sharing power scheduling strategy, converting the power limit value into the corresponding voltage output control command and switch conduction sequence to obtain the voltage output control parameters.

8. The remote control method for the parking air conditioner battery for emergency starting according to claim 7, characterized in that, The battery edge computing module, based on historical electricity consumption data, calculates state transition probabilities by analyzing historical parking air conditioning usage duration and frequency, and obtains electricity consumption behavior prediction results, including: The battery edge computing module divides historical electricity consumption data into time windows to obtain structured historical electricity consumption sequence data; The battery edge computing module performs statistical analysis on the parking air conditioner usage pattern based on the structured historical power consumption sequence data to obtain statistical parameters of parking air conditioner usage characteristics; The battery edge computing module discretizes the state space of the parking air conditioner usage characteristic statistical parameters according to usage duration and frequency to obtain a state transition probability matrix; The battery edge computing module performs multi-step predictions based on the state transition probability matrix and the current parking air conditioning usage status to obtain the electricity consumption behavior prediction results.

9. The remote control method for an emergency start parking air conditioner battery according to claim 1, characterized in that, The remote control method for the emergency start parking air conditioner battery also includes: The background management system encapsulates the SOC state of charge value, BMS switch status data and battery status parameters in real time to obtain a battery status data packet. The background management system remotely transmits and stores the battery status data packet through a 5G communication module and performs multi-parameter anomaly detection and fault warning generation to obtain battery status monitoring data. The back-end management system performs digital signature verification and secure encapsulation of the OTA firmware upgrade package, and sends it to the BMS controller of the battery edge computing module through the 5G communication module to obtain the OTA upgrade package sending confirmation signal; The backend management system tracks the firmware installation progress and verifies the completion status based on the confirmation signal issued by the OTA upgrade package. When it receives the installation success response returned by the battery edge computing module, it obtains the firmware update completion flag.

10. The remote control method for an emergency start parking air conditioner battery according to claim 9, characterized in that, The backend management system tracks the firmware installation progress and verifies the completion status based on the confirmation signal issued by the OTA upgrade package. When it receives a successful installation response from the battery edge computing module, it obtains a firmware update completion flag, including: The background management system sends a confirmation signal based on the OTA upgrade package and combines it with the geographical location data of the built-in GPS module to perform geographical security verification of the upgrade process. When an abnormal change in the geographical location of the emergency start battery is detected, the upgrade process is automatically interrupted and rolled back to the original firmware version, and a geographical security verification upgrade control command is obtained. The background management system intelligently schedules the upgrade timing based on the geographic security verification upgrade control command and the current power consumption mode classification results to obtain the upgrade process control strategy. The background management system performs real-time backup and dynamic updates of the BMS charging and discharging switch during the upgrade process based on the upgrade process control strategy, and obtains the parameter synchronization results of the upgrade process. The backend management system performs post-upgrade function verification and geographic environment adaptation testing based on the synchronization results of the upgrade process parameters. When it receives a successful installation response from the battery edge computing module, it obtains a firmware update completion flag.

Images