CN106054081A - A lithium battery modeling method for electric vehicle power battery SOC estimation - Google Patents

Abstract

The invention discloses a lithium battery modeling method for SOC estimation of a power battery of an electric automobile, which comprehensively considers the influence of SOC environmental factors, defines each electrical parameter in a lithium battery Thevenin model as a function of an environmental variable, obtains model parameters through a hybrid power pulse capability characteristic HPPC experiment, obtains actual parameter values of the battery model through testing and calculation, and determines a parameter fitting method of the model according to the actual parameter values. Based on the defects of the prior art, the method provided by the invention is used for establishing the variable parameter Thevenin model of the lithium battery by measuring and evaluating the internal parameters of the battery at different temperatures and SOC and analyzing the environmental factors influencing the parameter change.

Description

A lithium battery modeling method for electric vehicle power battery SOC estimation

technical field

The invention relates to the field of battery SOC modeling methods, in particular to a lithium battery modeling method for estimating the SOC of an electric vehicle power battery.

Background technique

SOC, that is, state of charge, refers to the remaining state of charge of the battery, generally defined as the ratio of the current remaining capacity of the battery to the nominal capacity, which directly affects the effectiveness of the balance control of the power battery pack and the reliability of the mileage judgment of the electric vehicle. However, the complex operating conditions of electric vehicles have brought great difficulties to the accurate estimation of SOC. Establishing an accurate lithium battery model is the key to improving the accuracy of SOC estimation. Currently, the commonly used lithium battery equivalent circuit models include Rint model, RC Among them, Thevenin model has the advantages of clear physical meaning and easy implementation of model parameter identification experiments. It is widely used in mathematical modeling of batteries, but its parameters are fixed and cannot reflect the dynamic characteristics of batteries.

In the existing SOC estimation methods, the consideration of environmental conditions, especially the influence of temperature is not comprehensive. In fact, the parameters of the battery model change continuously with the change of ambient temperature, and its available capacity varies greatly under different temperature conditions. Therefore, the temperature To a large extent, it affects the accuracy of the battery model, which in turn affects the estimation accuracy of SOC. With seasonal changes and latitude differences, there are large temperature changes during the operation of electric vehicles, so simply using a fixed model to estimate the remaining capacity will have a large error. Some algorithms propose to correct the SOC through the temperature compensation coefficient, or The estimated SOC at different temperatures is converted directly through the ampere-hour integration results. For the above method, the influence of temperature directly acts on the SOC estimation result as a correction factor, and its influence on the battery model still needs further study.

Contents of the invention

The purpose of the present invention is to provide a lithium battery modeling method for SOC estimation of electric vehicle power batteries, so as to solve the problem of the limitation of the scope of application of the model in the algorithm of the prior art.

In order to achieve the above object, the technical scheme adopted in the present invention is:

A lithium battery modeling method for electric vehicle power battery SOC estimation, characterized in that: comprehensively considering the influence of SOC environmental factors including temperature, defining each electrical parameter in the lithium battery Thevenin model as a function of environmental variables, And the model parameters are obtained through the HPPC experiment of the hybrid pulse capability characteristics. First, the actual parameter values of the battery model are obtained through testing and calculation, and the parameter fitting method of the model is determined based on this;

According to the Thevenin model, there is a polarization capacitance Cpol and a polarization resistance Rpol in the lithium battery, which are connected in parallel, and are combined in series with the DC voltage source Uoc representing the open circuit voltage, the ohmic internal resistance Rohm and other internal resistances R0, where the polarization capacitance Cpol, pole The chemical resistance Rpol and the ohmic internal resistance Rohm are variable parameters, which are related to the battery SOC environmental factors. The electrical parameters in the model can be calculated through the parameters that can be actually detected, and the model is obtained through the HPPC experimental calculation of the hybrid power pulse capability characteristic. parameter;

Analyze the experimental data to obtain the electrical parameter values of different environmental factors, and use the discrete data curve fitting method to obtain the functional relationship between each electrical parameter and environmental factors; it can be seen that in the range where the SOC is less than 30%, the ohmic internal resistance Rohm It suddenly rises, and when the SOC is greater than 30%, the influence of SOC on the ohmic internal resistance Rohm is not obvious, and the influence of temperature on SOC is a continuous process, so after analyzing the influencing factors, the battery model parameters are segmented;

The mathematical model of the lithium-ion power battery is the basis for estimating the SOC of the battery. The observation equation of the battery can be obtained through the established model. The observation equation describes the functional relationship between the SOC, charge and discharge current, temperature factors and the battery terminal voltage; the actual capacity of the battery It is set as a quantity that changes with temperature, and by studying the change law of the actual battery capacity at different temperatures, the temperature compensation coefficient _ηT is introduced to correct the error between the actual capacity of the battery and the rated capacity of the battery at different temperatures.

Compared with the prior art, the beneficial effects of the present invention are reflected in:

(1) In the Thevenin model of lithium batteries, parameters such as ohmic internal resistance, polarization internal resistance, and polarization capacitance are defined as functions of environmental variables, and parameter fitting is performed to improve the accuracy of the battery model and expand the model scope of application.

(2) The parameter identification method adopted takes into account the physical characteristics of lithium batteries, the measurement results are accurate, and the identified parameters are closer to the actual values.

Description of drawings

Fig. 1 is the relationship between Rohm and environmental factors measured by the present invention.

FIG. 2 is a structural diagram of the hardware system of the sampling link in a specific embodiment of the present invention.

Fig. 3 is a circuit diagram of voltage sampling for a multi-series battery pack in a specific embodiment of the present invention.

Fig. 4 is a temperature sampling circuit diagram in a specific embodiment of the present invention.

detailed description

A lithium battery modeling method for SOC estimation of electric vehicle power batteries, comprehensively considering the influence of SOC environmental factors including temperature, defining each electrical parameter in the lithium battery Thevenin model as a function of environmental variables, and through hybrid The model parameters are obtained from the HPPC experiment of pulse capability characteristics. First, the actual parameter values of the battery model are obtained through testing and calculation, and the parameter fitting method of the model is determined based on this;

According to the Thevenin model, there is a polarization capacitance Cpol and a polarization resistance Rpol in the lithium battery, which are connected in parallel, and are combined in series with the DC voltage source Uoc representing the open circuit voltage, the ohmic internal resistance Rohm and other internal resistances R0, where the polarization capacitance Cpol, pole The chemical resistance Rpol and the ohmic internal resistance Rohm are variable parameters, which are related to the battery SOC environmental factors. The electrical parameters in the model can be calculated through the parameters that can be actually detected, and the model is obtained through the HPPC experimental calculation of the hybrid power pulse capability characteristic. parameter;

Analyze the experimental data to obtain the electrical parameter values of different environmental factors, and use the discrete data curve fitting method to obtain the functional relationship between each electrical parameter and environmental factors; it can be seen that in the range where the SOC is less than 30%, the ohmic internal resistance Rohm It suddenly rises, and when the SOC is greater than 30%, the influence of SOC on the ohmic internal resistance Rohm is not obvious, and the influence of temperature on SOC is a continuous process, so after analyzing the influencing factors, the battery model parameters are segmented;

The mathematical model of the lithium-ion power battery is the basis for estimating the SOC of the battery. The observation equation of the battery can be obtained through the established model. The observation equation describes the functional relationship between SOC, charge and discharge current, temperature factors and the battery terminal voltage;

In the traditional EKF-based SOC estimation method, the total available capacity Qreal of the battery is regarded as a constant value. In fact, the temperature in the environmental factors will affect Qreal, and then affect the accuracy of the SOC estimation value.

The actual capacity of the battery is set as a quantity that changes with temperature. In order to explore the influence of temperature on Qreal, by studying the change law of the actual capacity of the battery at different temperatures, the temperature compensation coefficient _ηT is introduced to correct the actual capacity of the battery at different temperatures. Error between rated capacities.

Due to the existence of the thermal management system and the heat generated by battery discharge, the battery temperature will change during the operation of the electric vehicle. The power of the battery is constantly increasing, so the actual capacity of the battery is set as an amount that changes with temperature.

The battery SOC estimation is divided into two parts, the first part is the establishment of the battery mathematical model, and the second part is the design of the hardware system.

The model formula of the modeling link is as follows:

U _oc ＝U ₁ +R _ohm I _B +U _B (1)

In the formula, IB is the charging and discharging current of the battery, and U1 is the voltage across Rpol and Cpol. The relationship between battery internal resistance, temperature and SOC can be calculated through the group hybrid pulse capability test, and the battery internal resistance expression can be divided into two sections according to the closeness of battery internal resistance and SOC changes:

C _pol =τ/R pol (4)

In the formula, Rohm represents the ohmic internal resistance of the power battery, Rpol represents the internal resistance of the battery polarization, Cpol represents the polarization capacitance of the battery, T represents the working environment temperature of the battery, and SOC represents the remaining power of the battery. When the remaining capacity of the battery is greater than 30%, the battery model parameters are less affected by the change of battery SOC, which can be simplified as a function of the battery model to the ambient temperature to reduce the complexity of the algorithm; when the remaining capacity of the battery is less than 30%, the parameters of the battery model vary with the battery The change of SOC will change significantly, so the battery model parameters are binary relational expressions related to battery SOC and ambient temperature.

The SOC state equation during the discharge process is

Q _real = Q _full /η _T (6)

in SOC(t) is the instantaneous SOC value at time t; SOC(0) is the initial SOC value; i(t) is the instantaneous current value at time t; Qreal is the total available capacity of the battery. Qfull is the nominal total capacity of the battery, and ηT is a proportional coefficient, which is used to compensate the influence of influencing factors on the total capacity of the battery. This is to directly integrate the ampere-hour of battery discharge, but the integration error will continue to accumulate over time, resulting in a large SOC estimation error. Therefore, the extended Kalman filter algorithm (EKF) is used to eliminate the accumulated error and improve the accuracy of the algorithm.

U _B (SOC, T) = U _oc (SOC, T) - U ₁ (SOC, T) - R _ohm (SOC, T) · I _B (7)

y _k ＝U _oc (x _k )-U ₁ (x _k , T)-R _ohm (x _k , T)·μ _k +v _k (9)

Equation (7) is the observation equation of the battery nonlinear state space model obtained according to the battery model established in Fig. 1. In the EKF algorithm, the state variable xk of the battery model is the only variable, representing the SOC obtained by the kth calculation, the observed variable yk represents the battery terminal voltage UB obtained by the kth calculation of the battery model, and the input variable μk represents the kth calculation When the battery charge and discharge current IB. The discrete state space model of the battery can be obtained by discretizing equations (5) and (7) as equations (8) and (9). Among them, Δt is the sampling period, wk is the system noise, vk is the observation noise, the two are uncorrelated zero-mean Gauss white noise, and Uoc is the open circuit voltage of the battery, which is a variable only related to SOC.

Sampling link hardware system structure diagram

In order to realize the data acquisition of the battery model, the quantities to be collected include battery voltage, temperature and charge and discharge current. The hardware structure diagram is shown in Figure 2.

Since the electric vehicle power battery generally has a large number of series cells, the voltage acquisition adopts Linear's LTC6803-4 chip to detect the voltage of 12 series-connected battery cells, and the stacking structure can realize the voltage sampling of multi-cell series battery packs. The circuit structure is shown in Figure 3.

The temperature collection adopts 100kΩ NTC temperature sensor to collect the battery temperature. Use the voltage reference to supply power to the series circuit of the resistor and the NTC temperature sensor, collect the series voltage division of the temperature sensor and the resistor, and send it directly to the ADC built in the MCU for voltage acquisition through the signal conditioning circuit, calculate the temperature sensor resistance value, and then calculate the temperature sensor value according to the temperature sensor The temperature value can be obtained from the comparison table of temperature and resistance value. The sampling circuit is shown in Figure 4.

The current acquisition adopts an external Hall sensor. By matching the Hall sensor, the current is obtained, and the charging and discharging current is collected after the signal conditioning of the Hall sensor.

Claims (1)

1. the lithium battery modeling method estimated for electric automobile power battery SOC, it is characterised in that: consider and include Each electric parameter in lithium battery Thevenin model, in the impact of interior SOC environmental factors, is defined as environmental variable by temperature Function, and obtain model parameter by the experiment of hybrid power pulse ability characteristics HPPC, first pass through test and be calculated electricity Pool model actual parameter value, and determine the parameter fitness method of model on this basis；

According to Thevenin model, there is polarization capacity Cpol in lithium battery interior and polarization resistance Rpol is in parallel, and with sign The direct voltage source Uoc of open-circuit voltage, ohmic internal resistance Rohm and other internal resistances R0 tandem compound, wherein polarization capacity Cpol, pole Change resistance Rpol, ohmic internal resistance Rohm be variable element, relevant with battery SOC environmental factors, in model each electrically Parameter can by can actually detected to parameter calculate, obtained by hybrid power pulse ability characteristics HPPC experimental calculation To model parameter；

Analysis experimental data acquisition varying environment, because of each electrical parameters of vegetarian refreshments, utilizes discrete data curve-fitting method, obtains Each electric parameter and the functional relationship of environmental factors；It can be seen that be less than in the range of 30% at SOC, ohmic internal resistance Rohm can dash forward So raising, and when SOC is more than 30%, SOC is on the impact of ohmic internal resistance Rohm inconspicuous, and temperature is one on the impact of SOC Individual continuous print process, therefore carries out segment processing to battery model parameter after analyzing influence factor；

The mathematical model of lithium-ion-power cell is the basis of estimation battery SOC, passes through set up model and can obtain battery Observational equation, observational equation describes the functional relationship of SOC, charging and discharging currents, temperature factor and battery terminal voltage；By battery Actual capacity is set as an amount varied with temperature, by research battery actual capacity Changing Pattern at different temperatures, Introduce temperature compensation coefficient η_TRevise the error between battery actual capacity and battery rated capacity under different temperatures.