CN108572324A - Battery SOC Estimation Device Based on Immune Algorithm Optimizing BP Neural Network - Google Patents

Abstract

The present invention discloses the battery SOC estimation device based on immune algorithm Optimized BP Neural Network, including：Sensor group and the microprocessor for loading BP neural network model；Wherein, the following data of the sensor group acquisition battery：Charging and discharging currents, terminal voltage, environment temperature, charge and discharge number and previous battery charge state measured value, the microprocessor estimate the value of current time battery charge state according to the data acquired.The battery SOC estimation device based on immune algorithm Optimized BP Neural Network can establish power battery mathematical models, realize the accurate estimation to power battery nuclear power state.

Description

Battery SOC Estimation Device Based on Immune Algorithm Optimizing BP Neural Network

technical field

The present invention relates to a measuring device for estimating the State of Charge of a battery, in particular to a battery SOC estimating device based on an immune algorithm optimized BP neural network.

Background technique

With the depletion of fossil fuels and increasing environmental awareness, sales of electric vehicles are increasing day by day. Accurately predicting the nuclear power state of the power battery is of great significance to improving the energy utilization efficiency of the battery and prolonging the service life of the battery. However, ambient temperature, charge and discharge current, and charge and discharge times will all affect the internal parameters of the battery, and it is difficult to establish an accurate mathematical model for the battery, which increases the difficulty of predicting the nuclear power state of the power battery.

Contents of the invention

The purpose of the present invention is to provide a battery SOC estimation device based on immune algorithm optimization BP neural network, the battery SOC estimation device based on immune algorithm optimization BP neural network can establish an accurate mathematical model for the power battery, and realize the state of charge of the power battery. precise estimate of .

In order to achieve the above object, the present invention provides a battery SOC estimation device based on an immune algorithm to optimize the BP neural network, the battery SOC estimation device includes: a sensor group and a microprocessor loaded with a BP neural network model; wherein the sensor group collects The following data of the battery: charge and discharge current, terminal voltage, ambient temperature, charge and discharge times, and the previous battery state of charge measurement value. The microprocessor estimates the value of the battery state of charge at the current moment based on the collected data.

Preferably, the sensor group includes: a temperature sensor, wherein the temperature sensor is connected to the battery to collect the ambient temperature of the battery.

Preferably, the sensor group includes: a current sensor, wherein the current sensor is connected to the battery to collect charging and discharging current of the battery.

Preferably, the sensor group includes: a voltage sensor, wherein the voltage sensor is connected to the battery to collect the terminal voltage of the battery.

Preferably, the microprocessor pre-stores the previous measurement value of the battery state of charge, and collects the number of charging and discharging times of the battery.

Preferably, the microprocessor is connected to a host computer, and MATLAB software is used on the host computer to optimize the weight and threshold of the BP neural network based on the immune algorithm, and establish a mathematical model of the lithium battery.

Through the above technical scheme, the present invention adopts BP neural network model to fit battery state of charge and ambient temperature, charging and discharging current, sampling time, battery terminal voltage, battery state of charge value at the previous sampling moment and battery charging and discharging times to battery parameters The impact of fusion of information contained in a large amount of data, high accuracy and low cost. The invention has a wide range of application fields, and can be used in energy management of electric vehicle batteries, protection of battery packs, etc., and is especially suitable for high-power occasions.

Other features and advantages of the present invention will be described in detail in the detailed description that follows.

Description of drawings

The accompanying drawings are used to provide a further understanding of the present invention, and constitute a part of the description, together with the following specific embodiments, are used to explain the present invention, but do not constitute a limitation to the present invention. In the attached picture:

Fig. 1 illustrates a kind of BP neural network model of the battery SOC estimating device based on immune algorithm optimization BP neural network of the present invention;

Fig. 2 is the working flow diagram that illustrates a kind of immune algorithm optimization BP neural network of the present invention; And

FIG. 3 is a block diagram illustrating a battery SOC estimation device based on an immune algorithm optimized BP neural network according to the present invention.

Detailed ways

Specific embodiments of the present invention will be described in detail below in conjunction with the accompanying drawings. It should be understood that the specific embodiments described here are only used to illustrate and explain the present invention, and are not intended to limit the present invention.

The present invention provides a battery SOC estimation device based on an immune algorithm to optimize the BP neural network, the battery SOC estimation device includes: a sensor group and a microprocessor loaded with a BP neural network model; wherein, the sensor group collects the following data of the battery: The charging and discharging current, terminal voltage, ambient temperature, charging and discharging times, and the previous measured value of the battery state of charge, the microprocessor estimates the value of the battery state of charge at the current moment according to the collected data.

Through the above technical scheme, the present invention adopts BP neural network model to fit battery state of charge and ambient temperature, charging and discharging current, sampling time, battery terminal voltage, battery state of charge value at the previous sampling moment and battery charging and discharging times to battery parameters The impact of fusion of information contained in a large amount of data, high accuracy and low cost. The invention has a wide range of application fields, and can be used in energy management of electric vehicle batteries, protection of battery packs, etc., and is especially suitable for high-power occasions.

In a specific implementation manner of the present invention, the sensor group may include: a temperature sensor, wherein the temperature sensor is close to the battery to collect the ambient temperature of the battery.

In a specific implementation manner of the present invention, the sensor group may include: a current sensor, wherein the current sensor is connected to the battery to collect the charging and discharging current of the battery.

In a specific implementation manner of the present invention, the sensor group may include: a voltage sensor, wherein the voltage sensor is connected to the battery to collect the terminal voltage of the battery.

In a specific implementation manner of the present invention, the microprocessor prestores the previous measurement value of the state of charge of the battery, and collects the number of charging and discharging times of the battery.

In a specific embodiment of the present invention, the microprocessor is connected to a host computer, and MATLAB software is used on the host computer to optimize BP neural network weights and thresholds based on an immune algorithm, and to establish a mathematical model of a lithium battery.

The invention constructs a mathematical model based on the BP neural network to describe the nonlinear relationship between the ambient temperature, the charging and discharging current, the charging and discharging times and the state of charge of the power battery. Construct the BP neural network as shown in Figure 1, with the ambient temperature t, the charge and discharge current I, the sampling time T, the battery terminal voltage U, the charge and discharge times N of the power battery, and the state of charge of the battery at the previous sampling time SOC (k-1 ) as input, and the battery state of charge SOC(k) at the current moment as output, the BP neural network model established. The immune optimization algorithm is used to optimize the weight and threshold of BP neural network to obtain the best performance of BP neural network. The optimized process is shown in Figure 2:

Determine the topology of the BP neural network;

Encode the weights and thresholds of the neural network to form an initial antigen population;

Decode the initial antigen and assign it to BP to extend into the network;

Use the sample data to train the BP neural network;

Use the test sample to test the BP neural network, and obtain the error data between the output value of the BP neural network and the real value;

According to the above error data, evaluate the fitness of the antigen individual;

Judging whether the conditions are met, if yes, output the result; if not, according to the fitness of the antigen individual, promote and suppress the antigen population to form a new antigen population;

Return to (3) and continue iterative operations until the requirements are met.

In practical application, as shown in Figure 3, after using the MATLAB neural network toolbox to construct the BP neural network model, use the immune algorithm to determine its weight and threshold. Then realize the above-mentioned BP neural network model based on microprocessor programming. Data is sampled every sampling time T, the charge and discharge current I(k) of the battery is obtained through the current sensor, the terminal voltage U(k) of the battery is obtained through the voltage sensor, the ambient temperature t(k) is obtained through the temperature sensor and combined with the previous time The measured data of the battery state of charge is substituted into the BP neural network for calculation, and the value of the battery state of charge at the current moment is estimated.

The invention uses MATLAB software on the host computer to optimize the weight and threshold of the BP neural network based on the immune algorithm, establishes the mathematical model of the lithium battery, and improves the accuracy of the model. On this basis, the optimized BP neural network model is implemented based on the microprocessor, which reduces the cost of the equipment and improves the portability and practicability of the equipment.

When estimating the state of charge of the power battery, the present invention considers the sampling time, the charging and discharging current of the battery, the terminal voltage of the battery, the ambient temperature t(k), the number of charge and discharge times, and the previous battery state of charge measurement as The input quantity of the BP neural network estimates the value of the state of charge of the battery at the current moment, which improves the accuracy of the measuring device.

The preferred embodiment of the present invention has been described in detail above in conjunction with the accompanying drawings, but the present invention is not limited to the specific details of the above embodiment, within the scope of the technical concept of the present invention, various simple modifications can be made to the technical solution of the present invention, These simple modifications all belong to the protection scope of the present invention.

In addition, it should be noted that the various specific technical features described in the above specific embodiments can be combined in any suitable way if there is no contradiction. The combination method will not be described separately.

In addition, various combinations of different embodiments of the present invention can also be combined arbitrarily, as long as they do not violate the idea of the present invention, they should also be regarded as the disclosed content of the present invention.

Claims (6)

1. A battery SOC estimating device based on immune algorithm optimization BP neural network, characterized in that the battery SOC estimating device comprises: a sensor group and a microprocessor loaded with a BP neural network model; wherein the sensor group collects the battery SOC The following data: charging and discharging current, terminal voltage, ambient temperature, charging and discharging times, and the previous measured value of the battery state of charge, the microprocessor estimates the value of the battery state of charge at the current moment according to the collected data. 2. The battery SOC estimating device based on immune algorithm optimization BP neural network according to claim 1, wherein the sensor group comprises: a temperature sensor, wherein the temperature sensor is connected to the battery to collect battery ambient temperature. 3. The battery SOC estimation device based on immune algorithm optimization BP neural network according to claim 1, wherein the sensor group comprises: a current sensor, wherein the current sensor is connected to the battery to collect battery charge and discharge current. 4. The battery SOC estimation device based on immune algorithm optimization BP neural network according to claim 1, wherein the sensor group comprises: a voltage sensor, wherein the voltage sensor is connected to the battery to collect battery the terminal voltage. 5. the battery SOC estimating device based on immune algorithm optimization BP neural network according to claim 1, is characterized in that, described microprocessor prestores the battery state of charge measurement value of last time, and collects the charge and discharge of battery frequency. 6. the battery SOC estimation device based on immune algorithm optimization BP neural network according to claim 1, is characterized in that, described microprocessor is connected with upper computer, utilizes MATLAB software to optimize BP neural network based on immune algorithm on described upper computer Network weights and thresholds, and establish a mathematical model for lithium batteries.