Intelligent charging management system and management method based on battery health state

Abstract

The invention discloses an intelligent charging management system and management method based on a battery health state, which is used for managing a vehicle-mounted battery on a new energy electric vehicle and comprises a battery health monitoring module, an intelligent decision module, a charging execution module and a heat management module, the battery health monitoring module is used for correcting the SOH evaluation value of the battery and is matched with the intelligent decision-making module based on the correction, the optimal charging curve is dynamically output, pulse charging is integrated through the charging execution module so as to reduce polarization and recover charging waste heat, the battery equalization function can be started in time, and the service life of the battery is prolonged. According to the method, the attenuation speed of the service life of the battery is delayed, the thermal runaway risk caused by fast charging is reduced, the charging loss is reduced so as to improve the energy utilization rate, and the effect of optimizing the charging efficiency, the safety and the service life of the battery is achieved through real-time monitoring and dynamic regulation and control.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of vehicle-mounted battery charging control, in particular to an intelligent charging management system based on battery state of health, and also relates to a management method for battery intelligent charging management using the system BACKGROUND

[0002] New energy vehicles include pure electric vehicles, extended-range electric vehicles, hybrid electric vehicles, fuel cell electric vehicles, hydrogen engine vehicles, etc., wherein the new energy electric vehicles are all provided with vehicle-mounted batteries for storing and releasing electric energy, which need to be charged and discharged under the cooperation of corresponding battery management systems. At present, the existing new energy vehicle charging systems generally adopt the mode of fixed current or fixed voltage to implement charging, and cannot dynamically adjust the charging strategy according to the battery state of health (SOH), which will lead to problems such as too fast attenuation of the service life of the battery pack in the vehicle-mounted battery, increased risk of fast charging thermal runaway, and reduced energy utilization efficiency. Specifically, if the charging strategy cannot be dynamically adjusted according to the battery state of health (SOH), the annual average capacity loss of the battery pack will be more than 15%, which will cause the service life to quickly attenuate; the accident rate of battery thermal runaway caused by fast charging will increase to 43% of the battery charging failure; and the energy loss caused by charging will be as high as 20%-30%.

[0003] Therefore, it is necessary to further improve the charging management of the vehicle-mounted battery of the new energy vehicle to eliminate the above-mentioned defects. SUMMARY

[0004] The technical problem to be solved by the present application is to provide an intelligent charging management system based on battery state of health, which can realize triple optimization of charging efficiency, safety and battery life through real-time monitoring and dynamic regulation.

[0005] To solve the above technical problems, the technical scheme of the present application is: an intelligent charging management system based on battery state of health, which is used for managing the vehicle-mounted battery on a new energy electric vehicle, comprising a battery health monitoring module, an intelligent decision module, a charging execution module and a thermal management module, The battery health monitoring module is provided with a multi-physical field sensor array for signal acquisition and an electrochemical impedance spectroscopy detection unit, the multi-physical field sensor array and the electrochemical impedance spectroscopy detection unit are respectively signal connected to a master control unit MCU, the master control unit MCU is pre-provided with a battery state of health evaluation model, wherein the multi-physical field sensor array is used for detecting temperature, voltage, current, vibration and pressure signals of the vehicle-mounted battery; the electrochemical impedance spectroscopy detection unit performs regular internal resistance scanning on the vehicle-mounted battery; the battery state of health evaluation model corrects the battery state of health SOH evaluation value by fusing the ampere-hour integration method and the Kalman filtering algorithm; The intelligent decision module is connected with the main control unit MCU, and is provided with a charging strategy generation unit and a multi-objective optimization algorithm unit, wherein the charging strategy generation unit dynamically outputs an optimal charging curve by constructing a battery aging prediction model; and the multi-objective optimization algorithm unit makes a Pareto optimal decision based on charging speed, battery life and energy efficiency. The charging execution module is connected with the main control unit MCU, and is provided with a bidirectional DC-DC converter, a pulse charging controller and a waste heat recovery unit, for implementing voltage balancing, charging and waste heat recovery of the vehicle-mounted battery under the cooperation control of the battery health monitoring module and the intelligent decision module. The thermal management module is provided with a liquid cooling / air cooling composite heat dissipation unit and a temperature field balancing control unit, the liquid cooling / air cooling composite heat dissipation unit automatically switches between air cooling and liquid cooling modes according to the charging rate, and implements the voltage balancing and heat dissipation of the vehicle-mounted battery during charging; and the temperature field balancing control unit cooperates with the liquid cooling / air cooling composite heat dissipation unit to control the temperature difference between the battery monomers in the vehicle-mounted battery.

[0006] As a preferred technical solution, the electrochemical impedance spectrum detection unit performs internal resistance scanning of the vehicle-mounted battery once every 10 seconds, and the scanning accuracy reaches 0.1 mΩ.

[0007] As a preferred technical solution, the bidirectional DC-DC converter supports a wide range of current regulation of 0-150 A. The pulse output of the pulse charging controller is 5-500 Hz. The waste heat recovery unit is a phase change material (PCM) coupled thermoelectric generator module (TEG).

[0008] As a preferred technical solution, the battery aging prediction model constructed by the charging strategy generation unit is based on an LSTM neural network.

[0009] As a preferred technical solution, the liquid cooling / air cooling composite heat dissipation unit includes an air cooling subunit and a liquid cooling subunit. The air cooling subunit includes two ports of a flow guide air duct respectively arranged in communication with the inside of the vehicle-mounted battery, a centrifugal fan and an electric air duct baffle are connected in series on the flow guide air duct, and a PWM speed regulation module is installed on the centrifugal fan for cooperation. The liquid cooling subunit includes a plate heat exchanger used in cooperation with the vehicle-mounted battery, and a liquid cooling pipeline is in communication with the plate heat exchanger, a refrigerant storage device, a centrifugal pump and an electromagnetic valve are sequentially connected in series on the liquid cooling pipeline. The temperature field equalization control unit comprises a heat dissipation controller, and the heat dissipation controller is signal connected with the master control unit MCU, and the centrifugal fan, the electric air duct baffle, the PWM speed regulation module, the centrifugal pump and the electromagnetic valve are connected to the heat dissipation controller respectively.

[0010] As a preferred technical solution, the refrigerant storage stores a 50% concentration ethylene glycol solution, which is used as the refrigerant of the vehicle-mounted battery.

[0011] The application also relates to a management method of the intelligent charging management system based on the battery state of health. Real-time monitoring The temperature, voltage, current, vibration and pressure signals of the vehicle-mounted battery are collected in real time by the multi-physical field sensor array, and the collected signals are transmitted to the master control unit MCU. The internal resistance of the vehicle-mounted battery is scanned by the electrochemical impedance spectrum detection unit every 10 seconds, and the scanning data are transmitted to the master control unit MCU. Health assessment The battery health monitoring module obtains the battery state of health assessment model for correcting the battery state of health SOH assessment value by means of the fusion of ampere-hour integration method and Kalman filtering algorithm according to the data of the real-time monitoring step, i.e. The initial battery state of health SOH is calculated by using the ampere-hour integration method, and the calculation formula is, SOH init =Q current / Q rated ×100% In the formula, Q current is the current maximum available capacity of the vehicle-mounted battery, which is calculated by ∫I·η·dt, I is the charging current, η is the dynamically corrected coulomb efficiency, and Q rated is the rated capacity of the vehicle-mounted battery. Then, the battery state of health SOH value is corrected by iteration based on the state equation and the observation equation through the Kalman filtering algorithm, and the error rate after correction is less than 3%, wherein the state equation is In the formula, x k is the battery state of health SOH assessment value at k time, x k-1 is the battery state of health SOH assessment value at k-1 time, I k is the charging current at k time, η k is the dynamically corrected coulomb efficiency, and ω k is the process noise of the state equation. The observation equation is z k=OCV(SOH k , T k ) + R k ·I k +v k , wherein z k is the battery terminal voltage observation value at time k, OCV is the open circuit voltage, SOH k is the battery state of health SOH value at time k, T k is the battery temperature at time k, R k is the battery internal resistance at time k, I k is the charging current at time k, and v k is the observation noise of the observation equation; The battery state of health evaluation model is trained based on historical data, and the battery remaining capacity (RC), equivalent circuit parameters R0, R ct , and C dl are dynamically calculated; Strategy generation A battery aging prediction model is constructed based on an LSTM neural network, the model input parameters include historical current, voltage, temperature, internal resistance, and cycle number, the cycle number is set to 50 times, i.e., a time window of 50 sampling points, a 1-layer 64-neuron LSTM layer is used to capture the time sequence features, and the battery state of health SOH prediction value after 50 future cycles is output; training uses 100 sets of battery state of health SOH, the battery state of health SOH range used is 50-100%, MSE is used as the loss function, the parameters are updated online every 50 times of charging and discharging, and the optimal charging curve is dynamically output; Pareto optimal decision is made based on charging speed, battery life, and energy efficiency as objective functions, NSGA-II algorithm is used to solve the Pareto solution set, weights are dynamically allocated according to the battery state of health SOH, charging parameters are dynamically adjusted, and the charging speed is controlled; wherein the charging speed formula is F1=1 / t charge , wherein t charge is the charging time; The battery life formula is wherein E a is the activation energy, reflecting the temperature sensitivity, unit: J / mol, R is the gas constant (8.314 J / (mol•K)), T+273 is the absolute temperature, and I is the battery current; The energy efficiency formula is F3= Q output / Q input ×100%, wherein Q output is the total energy output in the discharging stage, and Q input is the total energy input in the charging stage; Execution control The charging execution module is used to implement multi-mode cooperative charging and timely implement intelligent equalization; The multi-mode cooperative charging is, When the battery state of health SOH≥90%, the charging rate is limited to be greater than or equal to 0.8C; When 80%≤battery state of health SOH<90%, the charging rate is limited to be less than 0.8C and greater than or equal to 0.5C; When the battery state of health SOH<80%, the charging rate is limited to be less than or equal to 0.5C and intelligent equalization is implemented; The intelligent equalization is, When the voltage difference of a certain battery cell is >5mV, the high-voltage battery cell energy is transferred to the low-voltage battery cell within a response time of <10ms through the cooperation of the pulse charging controller and the bidirectional DC-DC converter to complete voltage equalization; when the battery state of health SOH<80%, the equalization current is controlled to be increased by 20%; The thermal management module is used to implement automatic switching between air cooling and liquid cooling modes according to the charging rate, namely When the charging rate is ≤0.5C and the temperature of the vehicle-mounted battery is 25-35℃, the liquid cooling / air cooling composite heat dissipation unit is controlled to implement low-gear air cooling heat dissipation; when the temperature is >35℃ or the temperature difference of the battery cells is >2℃, it is switched to high-gear air cooling heat dissipation; when the charging rate is >0.8C and the temperature of the vehicle-mounted battery is 25-35℃, the liquid cooling / air cooling composite heat dissipation unit is controlled to implement low-flow liquid cooling heat dissipation; when the temperature is >35℃ or the temperature difference of the battery cells is >2℃, it is switched to high-flow liquid cooling heat dissipation; the PID algorithm is used to control the liquid cooling / air cooling composite heat dissipation unit to make the battery temperature fluctuate <1℃; When 0.5C<charging rate≤0.8C and the temperature of the vehicle-mounted battery is 25-35℃, the liquid cooling / air cooling composite heat dissipation unit is controlled to implement low-gear air cooling heat dissipation and low-flow liquid cooling heat dissipation; when the temperature is >35℃ or the temperature difference of the battery cells is >2℃, it is switched to high-gear air cooling heat dissipation and high-flow liquid cooling heat dissipation; The temperature field equalization control unit cooperates with the liquid cooling / air cooling composite heat dissipation unit to control the temperature difference between the battery cells in the vehicle-mounted battery to be <2℃.

[0012] As an improvement to the above technical solution, the charging execution module receives the battery state of health SOH difference data between the battery cells obtained by the battery health monitoring module; when the difference is >10%, the equalization priority is increased; When the intelligent equalization is implemented simultaneously during the charging process of the charging execution module, the charging current is automatically reduced by 10%-20%; After the equalization is completed, the battery cell consistency data is fed back to the intelligent decision module to optimize the charging curve; The heat energy generated in the intelligent equalization process is synchronously dissipated by the heat management module.

[0013] As an improvement of the above technical solution, the process of adding equalization priority is to suspend the current charging process, improve the response priority of intelligent equalization to the highest level, shorten the voltage difference detection interval, and increase the equalization current to 1.5 times of the regular value, until the SOH difference between the battery cells is ≤5%, and then restore the original charging process.

[0014] Due to the adoption of the above technical solution, the present application has the following beneficial effects: the battery health state SOH evaluation value is corrected by the battery health monitoring module, and based on this, the intelligent decision module is cooperated to dynamically output the optimal charging curve, the pulse charging is integrated by the charging execution module to reduce polarization and recover the charging waste heat, and the battery equalization function can be started in time to delay the decay speed of the battery life, reduce the risk of thermal runaway caused by fast charging, reduce the charging loss to improve the energy utilization rate, realize the real-time monitoring and dynamic regulation, and achieve the effect of optimizing the charging efficiency, safety and battery life. BRIEF DESCRIPTION OF DRAWINGS

[0015] The following drawings are only intended to illustrate and explain the present application, and do not limit the scope of the present application. Among them: Figure 1 is a simple block diagram of the liquid cooling / air cooling composite heat dissipation unit of the embodiment of the present application; Figure 2 is a simple flowchart of the multi-mode cooperative charging of the embodiment of the present application; In the figure: 1-vehicle-mounted battery; 2-flow guide air duct; 3-centrifugal fan; 4-electric air duct baffle; 5-PWM speed regulation module; 6-plate heat exchanger; 7-liquid cooling pipe; 8-refrigerant storage; 9-centrifugal pump; 10-solenoid valve; 11-heat dissipation controller. DETAILED DESCRIPTION

[0016] The present application will be further described below in conjunction with the drawings and embodiments. In the following detailed description, only certain exemplary embodiments of the present application are described by way of illustration. It is self-evident that those skilled in the art can make modifications to the described embodiments in various ways without departing from the spirit and scope of the present application. Therefore, the drawings and description are illustrative in nature and are not intended to limit the scope of protection of the claims.

[0017] The application discloses an intelligent charging management system based on a battery state of health, which is used for managing a vehicle-mounted battery on a new energy electric vehicle, and optimizes charging efficiency, safety and battery life through real-time monitoring and dynamic regulation, so that safe and energy-saving use of the vehicle-mounted battery is realized.

[0018] The battery health monitoring module is based on a battery state of health quantitative evaluation technology, and is provided with a multi-physical field sensor array and an electrochemical impedance spectroscopy (EIS) detection unit for signal acquisition. The multi-physical field sensor array and the electrochemical impedance spectroscopy (EIS) detection unit are respectively signal-connected to a master control unit (MCU), and a battery state of health evaluation model is pre-set in the master control unit (MCU). The master control unit (MCU) obtains a battery state of health (SOH) through the battery state of health evaluation model according to detection signals of the multi-physical field sensor array and the electrochemical impedance spectroscopy (EIS) detection unit after processing, and corrects the battery state of health (SOH) so that an error rate is less than 3%. The specific obtaining process of the battery state of health (SOH) is well known to those skilled in the art, and will not be described in detail here.

[0019] The multi-physical field sensor array in the battery health monitoring module is used for detecting temperature, voltage, current, vibration and pressure signals of the vehicle-mounted battery, such as a temperature sensor (with an accuracy within ±0.1℃) of TE Connectivity, a MAX471 current sensor (with an accuracy within ±0.5%) manufactured by the MAXIM company and the like. Among them, the vibration signal is used for monitoring mechanical abnormal phenomena such as loose connection of battery monomers and shell deformation, and a threshold value can be pre-set in the master control unit (MCU) as a control reference parameter. The vibration signal is first introduced into the battery health monitoring module for use, and can be used as one of the input parameters of the model, and participates in fusion calculation together with the internal resistance data obtained by the temperature, voltage, current, pressure signal and electrochemical impedance spectroscopy detection unit.

[0020] Specifically, the amplitude and frequency characteristics of the vibration signal are quantized and input into the battery state of health evaluation model pre-set in the master control unit (MCU), which is used for correcting error coefficients in the fusion process of the ampere-hour integral method and the Kalman filtering algorithm. For example, when the vibration signal is in a normal range (such as amplitude ≤0.3g, g is the acceleration of gravity), the model gives it a lower weight; when the vibration signal abnormally fluctuates (such as amplitude >0.3g and duration >1s), the model enhances the weight of the signal and enhances the correction strength of the battery state of health (SOH) evaluation value to reflect the influence of mechanical vibration on the stability of the internal structure of the battery, so as to finally improve the accuracy of SOH evaluation.

[0021] Help to improve the early fault warning ability. The electrochemical impedance spectroscopy (EIS) detection unit performs regular resistance scanning on the vehicle-mounted battery, such as resistance scanning of the vehicle-mounted battery every 10 seconds, with scanning accuracy up to 0.1 mΩ, high accuracy, and error rate guarantee for battery state of health SOH correction evaluation. The battery state of health evaluation model corrects the battery state of health SOH evaluation value by fusing the ampere-hour integration method and the Kalman filter algorithm to solve the problem of large evaluation error of the traditional battery state of health SOH.

[0022] The intelligent decision module is connected with the master control unit MCU, and is used for dynamically outputting an optimal charging curve. The module is provided with a charging strategy generation unit and a multi-objective optimization algorithm unit, wherein the charging strategy generation unit dynamically outputs an optimal charging curve by constructing a battery aging prediction model. The multi-objective optimization algorithm unit makes a Pareto optimal decision based on charging speed, battery life and energy efficiency. The master control unit MCU is based on an NXP S32K3 series microcontroller, and is integrated with a neural network acceleration unit (NNA) for constructing a battery aging prediction model. The charging strategy generation unit is based on an LSTM neural network, and completes construction of the battery aging prediction model under cooperation of the neural network acceleration unit (NNA).

[0023] The charging execution module is connected with the master control unit MCU, cooperates with an Infineon HybridPACK™ Drive 2 module installed on the vehicle body, supports 1200V / 300A high power density, is provided with a bidirectional DC-DC converter, a pulse charging controller and a waste heat recovery unit, and is used for implementing voltage balancing of the vehicle-mounted battery, charging of the vehicle-mounted battery and recovery of waste heat in the charging process under cooperation of the battery health monitoring module and the intelligent decision module. The bidirectional DC-DC converter supports 0-150A wide-range current regulation, and the efficiency is ≥98.5%. The pulse output of the pulse charging controller is a high-frequency pulse of 5-500Hz, so as to reduce polarization effect.

[0024] The waste heat recovery unit recovers charging waste heat by coupling a phase change material (PCM) and a thermoelectric generator module (TEG), and the heat recovery efficiency is increased to more than 18%. The waste heat recovery unit can be applied in the following manner: in a low-temperature environment (battery temperature < 15℃), the heat stored by the phase change material (PCM) is used for preheating the battery, so as to reduce low-temperature charging loss; and the electric energy generated by the thermoelectric generator module (TEG) is converted by a DC-DC converter, and is used for supplementing power supply of a vehicle-mounted low-voltage system (such as vehicle lights and a central control screen), so as to reduce main battery energy consumption.

[0025] The thermal management module is equipped with a liquid-cooled / air-cooled composite heat dissipation unit and a temperature field equalization control unit. The liquid-cooled / air-cooled composite heat dissipation unit automatically switches between air-cooled and liquid-cooled heat dissipation modes according to the charging rate to balance the voltage of the vehicle battery and dissipate heat during charging. The temperature field equalization control unit works in conjunction with the liquid-cooled / air-cooled composite heat dissipation unit to control the temperature difference between the individual battery cells in the vehicle battery.

[0026] like Figure 1 As shown, the liquid-cooled / air-cooled composite heat dissipation unit includes an air-cooled subunit and a liquid-cooled subunit, which form two independent heat dissipation channels that can be used in combination.

[0027] The air-cooling subunit includes two air ducts 2, each with its own port connected to the interior of the vehicle battery 1. The air ducts 2 can be constructed of aluminum alloy tubing. A centrifugal fan 3 and an electric air duct baffle 4 are connected in series on the air ducts 2. A PWM speed control module 5 is installed on the centrifugal fan 3 for coordinated use. The electric air duct baffle 4 controls the flow and shut-off of the air ducts 2. The centrifugal fan 3 accelerates the airflow speed between the vehicle battery 1 and the air ducts 2 to adjust the heat dissipation efficiency. The PWM speed control module 5 controls the power of the centrifugal fan 3, allowing for real-time adjustment based on heat dissipation efficiency requirements. The centrifugal fan 3 can be adjusted within an airflow range of 0-1500 m³ / h, which is sufficient to meet the heat dissipation needs of the vehicle battery 1.

[0028] The liquid-cooled subunit includes a plate heat exchanger 6 used in conjunction with the vehicle battery 1. The plate heat exchanger 6 is located inside the vehicle battery 1 and surrounds the outer periphery of each battery cell; this is a conventional structure in this technical field and will not be described in detail here. A liquid-cooling pipe 7 is connected to the plate heat exchanger 6. A refrigerant storage tank 8, a centrifugal pump 9, and a solenoid valve 10 are sequentially connected in series on the liquid-cooling pipe 7. The centrifugal pump 9 and the solenoid valve 10 allow the refrigerant to circulate between the refrigerant storage tank 8, the liquid-cooling pipe 7, and the plate heat exchanger 6. A flow regulating valve can also be connected in series on the liquid-cooling pipe 7 to improve the flexibility of flow regulation. In this embodiment, heat dissipation of the vehicle battery 1 is achieved through heat exchange between the plate heat exchanger 6 and the battery cells, with a heat exchange efficiency ≥95%, and heat energy is recovered through the refrigerant. The refrigerant storage tank 8 stores a 50% ethylene glycol solution, which is used as the refrigerant for the vehicle battery 1.

[0029] The temperature field equalization control unit comprises a heat dissipation controller 11, and the heat dissipation controller 11 is in signal connection with the master control unit MCU, the centrifugal fan 3, the electric air duct baffle 4, the PWM speed regulation module 5, the centrifugal pump 9 and the electromagnetic valve 10 are connected to the heat dissipation controller 11 respectively, and the automatic start-stop or opening degree control of the above components can be controlled through the heat dissipation controller 11 to meet the automatic switching between the air cooling and liquid cooling modes and the cooperation control of the two, so that the heat dissipation mode of the vehicle-mounted battery 1 is flexible and convenient to adjust.

[0030] The embodiment also relates to a management method of the intelligent charging management system based on the battery state of health, comprising the following steps, Real-time monitoring The temperature, voltage, current, vibration and pressure signals of the vehicle-mounted battery are collected in real time by the multi-physical field sensor array, and the collected signals are transmitted to the master control unit MCU for obtaining the battery state of health SOH evaluation value.

[0031] The internal resistance of the vehicle-mounted battery is scanned by the electrochemical impedance spectroscopy (EIS) detection unit every 10 seconds, and the scanning data are transmitted to the master control unit MCU for obtaining the battery state of health SOH evaluation value.

[0032] Health evaluation According to the data of the real-time monitoring step, the battery health monitoring module obtains the battery state of health evaluation model for correcting the battery state of health SOH evaluation value by fusing the ampere-hour integration method and the Kalman filtering algorithm, that is, the initial battery state of health SOH is calculated by using the ampere-hour integration method, and the calculation formula is, SOH init =Q current / Q rated ×100% In the formula, Q current is the current maximum available capacity of the vehicle-mounted battery, which is calculated by ∫I·η·dt, I is a charging current, η is a dynamically corrected coulomb efficiency, which can be calibrated by the manufacturer of the vehicle-mounted battery, and Q rated is the rated capacity of the vehicle-mounted battery.

[0033] Then, the battery state of health SOH value is corrected by iteration based on the state equation and the observation equation through the Kalman filtering algorithm, and the error rate after correction is less than 3%, wherein the state equation is: In the formula, x k is the battery state of health SOH evaluation value at k time, and x k-1is the battery state of health SOH evaluation value at k-1 moment, I k is the charging current at k moment, η k is the dynamically corrected coulomb efficiency, ω k is the process noise of the state equation (subject to a Gaussian distribution with a mean of 0).

[0034] The observation equation is z k =OCV(SOH k ,T k )+R k ·I k +v k , wherein z k is the battery terminal voltage observation value at k moment, OCV is the open circuit voltage, SOH k is the battery state of health SOH value at k moment, T k is the battery temperature at k moment, R k is the battery internal resistance at k moment, I k is the charging current at k moment, and v k is the observation noise of the observation equation (subject to a Gaussian distribution with a mean of 0). The battery state of health evaluation model is trained based on historical data, and the battery remaining capacity (RC), equivalent circuit parameters R0, R ct and C dl are dynamically calculated.

[0035] Strategy generation A battery aging prediction model is constructed based on an LSTM neural network, the model input parameters include historical current, voltage, temperature, internal resistance and cycle number, the cycle number is set to 50 times, that is, a time window of 50 sampling points, a 64-neuron LSTM layer is used to capture the time sequence characteristics, and a battery state of health SOH prediction value after 50 future cycles is output; training uses 100 battery state of health SOHs, the battery state of health SOH range used is 50-100%, MSE is used as the loss function, the parameters are updated online every 50 charge-discharge, and the optimal charging curve is dynamically output. The specific model algorithm is: Python: class BatteryHealthModel(tf.keras.Model): def __init__(self): super().__init__() self.lstm=tf.keras.layers.LSTM(64,return_sequences=True) self.dense = tf.keras.layers.Dense(1) def call(self, inputs): x = self.lstm(inputs) return self.dense(x) Pareto optimal decision is made based on charging speed, battery life and energy efficiency as target function, NSGA-II algorithm is used to solve Pareto solution set, weight is dynamically allocated according to battery state of health SOH, charging parameters are dynamically adjusted, charging speed is controlled, for example, when battery state of health SOH is greater than or equal to 90%, F1 weight is set to 40%.

[0036] wherein, the charging speed formula is F1=1 / t charge , wherein t charge is the time used for charging, the shorter the time, the faster the charging speed.

[0037] The battery life formula is wherein, E a is the activation energy, reflecting temperature sensitivity, unit: J / mol, R is the gas constant (8.314 J / (mol•K)), T+273 is the absolute temperature, and I is the battery current.

[0038] The energy efficiency formula is F3= Q output / Q input ×100%, wherein Q output is the total energy (Wh) output in the discharging stage, and Q input is the total energy (Wh) input in the charging stage.

[0039] Perform control Implement multi-mode collaborative charging by using the charging execution module, and timely implement intelligent balancing.

[0040] As shown in Figure 2 , the multi-mode collaborative charging is that the charging rate and heat dissipation mode are determined by the battery state of health SOH. In this embodiment, when the battery state of health SOH is greater than or equal to 90%, the charging rate is limited to be greater than or equal to 0.8C, fast charging is implemented, and heat dissipation is implemented in combination with the heat management module; when 80%≤battery state of health SOH

[0041] The intelligent equalization is that when the voltage difference of a certain battery monomer is greater than 5mV, the energy of the high-voltage battery monomer is transferred to the low-voltage battery monomer within a response time of less than 10ms through the cooperation of the pulse charging controller and the bidirectional DC-DC converter, so as to complete the voltage equalization; when the state of health SOH of the battery is less than 80%, the equalization current is controlled to be increased by 20% to balance the voltage of each battery monomer, and the polarization is reduced through the integration of pulse charging.

[0042] The automatic switching between the air cooling and liquid cooling modes is implemented by the thermal management module according to the charging rate, so that the vehicle-mounted battery is always maintained at 25-35℃ during the charging process. Specifically: When the charging rate is less than or equal to 0.5C and the temperature of the vehicle-mounted battery is 25-35℃, the low-gear air cooling is controlled to be implemented by the liquid cooling / air cooling composite cooling unit, and when the temperature is greater than 35℃ or the temperature difference between the battery monomers is greater than 2℃, the high-gear air cooling is switched to, so as to realize the temperature reduction or temperature equalization; when the charging rate is greater than 0.8C and the temperature of the vehicle-mounted battery is 25-35℃, the low-flow liquid cooling is controlled to be implemented by the liquid cooling / air cooling composite cooling unit, and when the temperature is greater than 35℃ or the temperature difference between the battery monomers is greater than 2℃, the high-flow liquid cooling is switched to; the PID algorithm is set in the cooling controller, the liquid cooling / air cooling composite cooling unit is controlled through the PID algorithm, the speed of the centrifugal fan 3 and the flow of the electromagnetic valve 10 are actually controlled, the temperature fluctuation of the battery is less than 1℃ through the cooperation of the two cooling modes, and the gentle temperature reduction of the vehicle-mounted battery 1 is formed.

[0043] When the charging rate is greater than 0.5C and less than or equal to 0.8C and the temperature of the vehicle-mounted battery is 25-35℃, the low-gear air cooling and the low-flow liquid cooling are controlled to be implemented by the liquid cooling / air cooling composite cooling unit; when the temperature is greater than 35℃ or the temperature difference between the battery monomers is greater than 2℃, the high-gear air cooling and the high-flow liquid cooling are switched to for cooperative cooling.

[0044] The temperature field equalization control unit cooperates with the liquid cooling / air cooling composite cooling unit to control the temperature difference between the battery monomers in the vehicle-mounted battery to be less than 2℃, so as to realize the temperature equalization between the battery monomers.

[0045] The charging execution module of the embodiment can receive the battery health state SOH difference data between the battery monomers obtained through the battery health monitoring module, and when the difference is greater than 10%, the equalization priority is increased to maintain the consistency of the states of the battery monomers. Specifically, the process of increasing the equalization priority is as follows: the current charging process is paused, the response priority of the intelligent equalization is raised to the highest level, the voltage difference detection interval is shortened (from the regular 100ms to 50ms), and the equalization current is increased to 1.5 times of the regular value, and then the original charging process is restored after the SOH difference between the battery monomers is less than or equal to 5%.

[0046] When the intelligent balancing is implemented simultaneously with the charging process performed by the charging execution module, the charging current is automatically reduced by 10%-20% to facilitate the implementation of the balancing safely and smoothly. After the balancing is completed, the consistency data of the battery monomers is fed back to the intelligent decision module to optimize the charging curve, and based on the online learning algorithm of the intelligent decision module, the system can automatically update the charging curve every 50 times of charging and discharging. The heat energy generated in the intelligent balancing process is implemented by the heat management module to implement synchronous heat dissipation.

[0047] The description of the application is given for the purpose of illustration and description, and is not intended to be exhaustive or to limit the application to the disclosed form. Many modifications and variations will be apparent to those of ordinary skill in the art. The embodiments are chosen and described in order to best explain the principles of the application and its practical application, and to enable others skilled in the art to understand the application for various embodiments with various modifications as are suited to the particular use contemplated.

Claims

1. An intelligent charging management system based on battery health status, used for managing onboard batteries in new energy electric vehicles, characterized by: It includes a battery health monitoring module, an intelligent decision-making module, a charging execution module, and a thermal management module. The battery health monitoring module includes a multi-physics sensor array for signal acquisition and an electrochemical impedance spectroscopy (EIS) detection unit. Both the multi-physics sensor array and the EIS detection unit are connected to a main control unit (MCU). The MCU contains a pre-set battery health status assessment model. The multi-physics sensor array detects the temperature, voltage, current, vibration, and pressure signals of the vehicle battery. The EIS detection unit performs timed internal resistance scans of the vehicle battery. The battery health status assessment model corrects the SOH (State of Health) assessment value by fusing the ampere-hour integral method and the Kalman filter algorithm. The intelligent decision-making module is connected to the main control unit (MCU) and includes a charging strategy generation unit and a multi-objective optimization algorithm unit. The charging strategy generation unit dynamically outputs the optimal charging curve by constructing a battery aging prediction model. The multi-objective optimization algorithm unit performs Pareto optimal decision-making based on charging speed, battery life, and energy efficiency. The charging execution module is connected to the main control unit MCU and is equipped with a bidirectional DC-DC converter, a pulse charging controller and a waste heat recovery unit. Under the coordinated control of the battery health monitoring module and the intelligent decision module, it implements voltage balancing, charging and waste heat recovery of the vehicle battery during the charging process. The thermal management module is equipped with a liquid-cooled / air-cooled composite heat dissipation unit and a temperature field equalization control unit. The liquid-cooled / air-cooled composite heat dissipation unit automatically switches between air-cooled and liquid-cooled heat dissipation modes according to the charging rate to balance the voltage of the vehicle battery and dissipate heat during charging. The temperature field equalization control unit works in conjunction with the liquid-cooled / air-cooled composite heat dissipation unit to control the temperature difference between the individual battery cells in the vehicle battery.

2. The intelligent charging management system based on battery health status as described in claim 1, characterized in that: The electrochemical impedance spectroscopy detection unit performs an internal resistance scan of the vehicle battery every 10 seconds, with a scanning accuracy of 0.1 mΩ.

3. The intelligent charging management system based on battery health status as described in claim 1, characterized in that: The bidirectional DC-DC converter supports a wide current adjustment range of 0-150A; The pulse output of the pulse charging controller is 5-500Hz; The waste heat recovery unit is a phase change material (PCM) coupled thermoelectric generator (TEG).

4. The intelligent charging management system based on battery health status as described in claim 1, characterized in that: The battery aging prediction model constructed by the charging strategy generation unit is based on an LSTM neural network.

5. The intelligent charging management system based on battery health status as described in claim 1, characterized in that: The liquid-cooled / air-cooled composite heat dissipation unit includes an air-cooled subunit and a liquid-cooled subunit; The air-cooled subunit includes two air ducts that are respectively connected to the inside of the vehicle battery. A centrifugal fan and an electric air duct baffle are connected in series on the air duct. A PWM speed control module is installed on the centrifugal fan for use. The liquid cooling subunit includes a plate heat exchanger used in conjunction with the vehicle battery, and a liquid cooling pipe is connected to the plate heat exchanger. A refrigerant storage device, a centrifugal pump and a solenoid valve are sequentially connected in series on the liquid cooling pipe. The temperature field equalization control unit includes a heat dissipation controller, and the heat dissipation controller is connected to the main control unit MCU. The centrifugal fan, the electric air duct baffle, the PWM speed control module, the centrifugal pump and the solenoid valve are respectively connected to the heat dissipation controller.

6. The intelligent charging management system based on battery health status as described in claim 5, characterized in that: The refrigerant storage device contains a 50% ethylene glycol solution, which is used as the refrigerant for the vehicle battery.

7. The management method of the intelligent charging management system based on battery health status as described in claim 1, characterized in that: Includes the following steps, Real-time monitoring The multi-physics sensor array is used to collect the temperature, voltage, current, vibration and pressure signals of the vehicle battery in real time, and the collected signals are sent to the main control unit MCU. The internal resistance of the vehicle battery is scanned every 10 seconds using the electrochemical impedance spectroscopy detection unit, and the scan data is transmitted to the main control unit MCU. Health assessment The battery health monitoring module, based on data from real-time monitoring steps, uses a fusion of the ampere-hour integral method and the Kalman filter algorithm to obtain the battery health state assessment model used to correct the battery health state (SOH) assessment value. First, the initial state of health (SOH) of the battery is calculated using the ampere-hour integration method. The calculation formula is as follows: SOH init =Q current / Q rated ×100% In the formula, Q current The current maximum usable capacity of the vehicle battery is calculated using ∫I·η·dt, where I is the charging current, η is the dynamically corrected coulombic efficiency, and Q is the current. rated The rated capacity of the vehicle battery; Then, based on the state equation and the observation equation, the battery health state (SOH) value is iteratively corrected using a Kalman filter algorithm, with a corrected error rate of <3%. The state equation is... In the formula, x k Let x be the state of health (SOH) assessment value of the battery at time k. k-1 I represents the State of Health (SOH) assessment value of the battery at time k-1. k Let η be the charging current at time k. k For dynamically corrected Coulomb efficiency, ω k This refers to process noise in the state equations. The observation equation is z k =OCV(SOH) k T k )+R k ·I k +v k In the formula, z k Let OCV be the observed battery terminal voltage at time k, and SOH be the open-circuit voltage. k Let T be the state of health (SOH) value of the battery at time k. k Let R be the battery temperature at time k. k Let I be the internal resistance of the battery at time k. k Let v be the charging current at time k. k The observation noise in the observation equation; The battery health status assessment model is trained based on historical data to dynamically calculate the remaining battery capacity (RC), equivalent circuit parameters R0, and R... ct and C dl ; Strategy Generation A battery aging prediction model is constructed based on an LSTM neural network. The model input parameters include historical current, voltage, temperature, internal resistance, and cycle count. The cycle count is set to 50 times, i.e., 50 sampling points in the time window. A 64-neuron LSTM layer is used to capture time-series features, and the predicted state of health (SOH) of the battery after the next 50 cycles is output. The training uses 100 sets of battery SOH values, with the SOH range being 50-100%. MSE is used as the loss function, and the parameters are updated online every 50 charge-discharge cycles, dynamically outputting the optimal charging curve. Pareto optimal decision-making is performed based on charging speed, battery life and energy efficiency as objective functions. The NSGA-II algorithm is used to solve the Pareto solution set. The weights are dynamically allocated according to the battery health state (SOH), and the charging parameters are dynamically adjusted to control the charging speed. The charging speed formula is F1 = 1 / t charge In the formula t charge Time taken for charging; The battery life formula is: In the formula, E a The activation energy reflects temperature sensitivity, and the unit is J / mol. R is the gas constant (8.314 J / (mol•K)), T+273 is the absolute temperature, and I is the battery current. The energy efficiency formula is F3 = Q output / Q input ×100%, where Q output Q represents the total energy output during the discharge phase (Wh). input Total energy input during the charging phase (Wh); Implementation of regulation The charging execution module is used to implement multi-mode collaborative charging and intelligent balancing as needed; Multi-mode collaborative charging is, When the battery's state of health (SOH) is ≥ 90%, the charging rate is limited to 0.8C or higher. When 80% ≤ State of Health (SOH) < 90%, the charging rate is limited to less than 0.8 and greater than or equal to 0.5C. When the battery health status (SOH) is less than 80%, the charging rate is limited to less than 0.5C and intelligent balancing is implemented. Intelligent balancing is, When the voltage difference of a battery cell is greater than 5mV, the pulse charging controller and the bidirectional DC-DC converter work together to transfer energy from the high-voltage battery cell to the low-voltage battery cell within a response time of less than 10ms, thus completing voltage equalization; when the battery health state (SOH) is less than 80%, the equalization current is increased by 20%. The thermal management module automatically switches between air cooling and liquid cooling modes based on the charging rate. When the charging rate is ≤0.5C and the vehicle battery temperature is 25-35℃, the liquid-cooled / air-cooled composite heat dissipation unit is controlled to implement low-level air-cooling heat dissipation. When the temperature is >35℃ or the temperature difference between individual battery cells is >2℃, it switches to high-level air-cooling heat dissipation. When the charging rate is >0.8C and the vehicle battery temperature is 25-35℃, the liquid-cooled / air-cooled composite heat dissipation unit is controlled to implement low-flow liquid cooling heat dissipation. When the temperature is >35℃ or the temperature difference between individual battery cells is >2℃, it switches to high-flow liquid cooling heat dissipation. The liquid-cooled / air-cooled composite heat dissipation unit is controlled by a PID algorithm to ensure that the battery temperature fluctuation is <1℃. When 0.5C < charging rate ≤ 0.8C and the vehicle battery temperature is 25-35℃, the liquid cooling / air cooling composite heat dissipation unit is controlled to implement low-level air cooling and low-flow liquid cooling for coordinated heat dissipation; when the temperature is > 35℃ or the temperature difference of the battery cells is > 2℃, it switches to high-level air cooling and high-flow liquid cooling for coordinated heat dissipation. The temperature field equalization control unit works in conjunction with the liquid cooling / air cooling composite heat dissipation unit to control the temperature difference between individual battery cells in the vehicle battery to be less than 2℃.

8. The management method as described in claim 7, characterized in that: The charging execution module receives the battery health status (SOH) difference data between individual battery cells obtained through the battery health monitoring module. When the difference is greater than 10%, the equalization priority is increased. When the charging execution module performs intelligent balancing during the charging process, it automatically reduces the charging current by 10%-20%. After equalization is completed, the consistency data of individual battery cells is fed back to the intelligent decision-making module to optimize the charging curve; The heat generated during the intelligent balancing process is simultaneously dissipated by the thermal management module.

9. The management method as described in claim 8, characterized in that: The process of increasing the equalization priority is as follows: pause the current charging process, raise the response priority of intelligent equalization to the highest level, shorten the voltage difference detection interval, and increase the equalization current to 1.5 times the normal value until the SOH difference between battery cells is ≤5%, then resume the original charging process.

Images