Dynamical modeling least square support vector machine SOC estimation method

Abstract

The invention provides a dynamical modeling least square support vector machine soc estimation method belonging to the SOC estimation method technology field. The dynamical modeling least square support vector machine SOC estimation method comprises the steps that in a discharge initial stage, an LS-SVM is constantly matched with a sample most similar to a current discharge cycle, and is used to constantly build a new model to provide an SOC value corresponding to a current discharge moment, and after discharge is stable, no more new model is built; the current model is used to complete the SOC estimation of the subsequent discharge moment. During SOC estimation of a whole life cycle of a battery, the method provided by the invention is used for different batteries. The charging and the discharging of the battery is a complicated electrochemical reaction, and therefore the SOC is unable to be acquired by direct measurement. During the SOC estimation, the external measurable parameters of the battery are acquired, such as voltage, current, resistance, and temperature. The external parameters are in non-linear correlation with the SOC, and therefore the parameters are mapped to the SOC in a non-linear manner, and an object of measuring estimated SOC from outside is achieved.

Description

technical field [0001] The invention relates to a dynamic modeling least square support vector machine SOC estimation method, which belongs to the technical field of SOC estimation methods. Background technique [0002] The function of the battery is to realize the conversion between chemical energy and electrical energy, so its power is unpredictable. However, in practical applications, people need to make task planning and expected charging time based on the current state of charge of the battery, that is, the battery power. If the current battery power cannot be accurately estimated, taking electric vehicles as an example, it will cause power failure during driving and cannot complete the expected charging time. The driving route, or frequent charging wastes the owner's time and reduces the battery life and other hazards. Therefore, the State of Charge (SOC) is the most in-depth and extensive work carried out in the current battery state estimation research system. The ...

AI Technical Summary

Problems solved by technology

The open circuit voltage method needs to be left for a long time before it can be measured, so it is not suitable for online estimation of SOC

The internal resistance method needs to accurately measure the internal resistance of the battery, which requires extremely high accuracy of the measuring instrument, so it is not suitable for online estimation of SOC

The ampere-time method is a direct method for estimating SOC, which requires an initial value of SOC and a high-precision current measurement, and its errors will continue to accumulate, so it cannot be used alone

[0003] However, factors such as the battery's environment, operating conditions, and life decline will lead to an increase in the SOC estimation error, which has caused difficulties and challenges in accurate SOC estimation.

However, most of the current research does not focus on the SOC estimation of the battery's entire life cycle, and the errors of various models established correspondingly increase with the decline of battery life.

In the communication base station standby power state estimation project that the laboratory cooperated with enterprises in the early stage, many problems still exist in battery SOC estimation and the remaining requirements in practical applications were exposed.

Images