Lithium battery capacity online estimation method based on convolutional neural network

Abstract

The invention discloses a lithium battery capacity online estimation method based on a convolutional neural network, and the method comprises the steps: firstly carrying out the processing through employing the charging and discharging circulation experiment data of a reference battery, obtaining a source data set comprising a plurality of samples and corresponding capacity labels, wherein each sample is composed of a charging voltage, a charging voltage first-order differential and charging current data; furthermore, establishing a convolutional neural network for estimating the capacity of the lithium battery, and optimizing hyper-parameters of the neural network by using an optimization algorithm; taking charging voltage and charging current data with the same charging capacity intervallength as the source data set sample in the latest charging process of the tested battery; and obtaining and inputting charging voltage, charging voltage first-order differential and charging currentdata into the convolutional neural network which has been subjected to optimization training, and the output value of the convolutional neural network is the capacity estimation value of the tested battery. The method is suitable for lithium battery capacity online estimation under the actual use condition, any section of charging voltage and current data meeting the interval length requirement can be used as input data, the data requirement is low, the calculation resource consumption is low, the estimation precision is high, and the method has very high actual application value.

Description

[0001] Technical field [0002] The invention proposes an online lithium battery capacity estimation method based on a convolutional neural network, which belongs to the technical field of fault prediction and health management (PHM). Background technique [0003] Due to its excellent characteristics (such as high energy density, long life, etc.), lithium batteries have been more and more widely used in electric vehicles, mobile phones and other fields. During the use of lithium batteries, the estimation of their capacity is very important. In fact, in actual use, the lithium battery will not be discharged from 100% state of charge to 0% according to the standard discharge current, and its capacity cannot be directly measured, so other methods are needed to estimate the capacity of the battery. [0004] Commonly used battery capacity estimation methods mainly include two categories: model-based and data-driven. Specifically, the model-based battery capacity estimation method...

AI Technical Summary

Problems solved by technology

However, such methods usually require that the state of charge (SOC) range of the battery during charging or discharging is large enough to pass through some specific SOC points, which is difficult to achieve in actual use.

In fact, in the actual use of the battery, the initial state of charge and the final state of charge of the charging or discharging process are uncertain, and even in the discharging process, the magnitude of the discharge current and the temperature of the environment, etc. All of them are uncertain and dynamically changing. Therefore, the above-mentioned models have certain limitations in the application of state monitoring and fault diagnosis of lithium batteries under actual working conditions.

Images