Method for measuring residual capacity of battery in online manner on basis of particle swarm optimization

Abstract

The invention discloses a method for measuring the residual capacity of a battery in an online manner on the basis of particle swarm optimization. The method includes fully charging the lead-acid storage battery, cooling the charged lead-acid storage battery until the temperature of the lead-acid storage battery reaches the room temperature, discharging the lead-acid storage battery by a low constant current, and sampling and recording output voltages of a sensor and the residual capacity of the battery at fixed intervals; using data recorded in experiments as input data of a support vector machine and training and creating an SVR (support vector regression) model; solving parameters in the model by the aid of a particle swarm optimization algorithm to acquire a mathematical relation among the residual capacity of the battery and the output voltages of the sensor; combining the obtained relation among the residual capacity of the battery and the output voltages of the sensor with a currently measured output voltage of the sensor to acquire the residual capacity of the storage battery in the online manner. The method has the advantages that experiential knowledge and priori knowledge of designers are omitted, the residual capacity of the storage battery can be accurately and quickly acquired from the output voltages of the sensor by the aid of the few experimental data, the efficiency and the precision are high, and the method is high in practicality.

Description

technical field [0001] The invention belongs to Liang Yu of lead-acid storage battery technology, and in particular relates to an online measurement technology for the remaining capacity of a lead-acid storage battery. Background technique [0002] Most batteries have insignificant or irregular changes in their electrochemical properties as their state of charge changes and cannot be used as a direct indicator of SOC. At present, the methods commonly used to detect the remaining capacity of batteries at home and abroad include internal resistance method, power accumulation method, open circuit voltage method, ampere-hour (AH) measurement method, linear model method, Kalman filter method, neural network method, etc. However, in these technologies, such as open circuit voltage and internal resistance methods, there are shortcomings such as poor accuracy and lack of real-time performance. There are cumulative errors in current measurement in the ampere-hour (AH) measurement met...

AI Technical Summary

Problems solved by technology

However, in these technologies, such as open circuit voltage and internal resistance methods, there are shortcomings such as poor accuracy and lack of real-time performance. There are cumulative errors in current measurement in the ampere-hour (AH) measurement method, which affects the measurement accuracy, while methods such as fuzzy neural networks. There are disadvantages of relying too much on the structure, training data and training methods of the neural network

Images