Lifepo4 battery manufacturer Specialized in Energy Storage Battery since 2009.

# Improvement of Uninterruptible Power Supply (UPS) SOC Estimation Algorithm

by：dcfpower     2021-04-01
Since the Ah measurement method is prone to cause current errors, the open circuit voltage method is not convenient to measure the open circuit voltage, but the extended Kalman filter method can make up for the shortcomings of the two and play a corrective role. Therefore, in this article, we use the Ah measurement method, The combined algorithm of the open circuit voltage method and the extended Kalman filter method gives full play to the respective advantages of the three and overcomes the shortcomings of each other. In terms of models, although the model with hysteresis effect and the model with polarization effect can better reflect the characteristics of the battery, it is obvious that the structure of the model is complex and the calculation cost is high. The simplified model integrates the open circuit voltage into one variable. Calculate the open circuit voltage, so we use a composite model, which can reflect the main characteristics of the battery, improve the accuracy, and reduce the calculation time at the same time. Ah measurement method is currently the simplest and most commonly used estimation method. According to the discrete-time idea of u200bu200bKalman filter algorithm, we rewrite its basic principle formula into the following form: In formula (2-2), SOCk represents the k-th moment SOC state value, the SOC state value in the traditional Ah measurement method is quantified, and it can be effectively corrected by the extended Kalman filter algorithm. In this formula, there are two factors that we must solve, namely the charge-discharge rate influence coefficient ηi and the temperature influence coefficient ηT. Among them, the charge-discharge rate influence coefficient ηi can be obtained from the Peukert equation. It can be seen from the charge and discharge characteristics of the battery that different charge and discharge rates will cause the battery to have different charge and discharge capacities. At the same time, different charge and discharge rates also have an impact on the changes in battery voltage and SOC state values. After using the relationship between the open circuit voltage and the SOC state value to obtain the initial value of the SOC at each sampling time, and use the Ah measurement method to estimate the SOC state value at the next time, it is necessary to extend the correction function of the Kalman filter algorithm to the open circuit voltage method The initial value of and the error caused by the accumulation of current of Ah measurement method over time are corrected to improve the accuracy of calculation. The specific correction steps of the extended Kalman filter algorithm are as follows: (1) Model selection: In order to reduce the cost of calculation while ensuring the accuracy of the estimation results, we adopt a composite model. (2) Calculate the equation matching coefficient of the Kalman filter. (3) Initialization of state variables. (4) Use the extended Kalman filter algorithm for correction. The initial value of SOC state SOC0 can be calculated based on the previous remaining power and the current battery open circuit voltage. The initial value of noise error Dw, Dv and mean square estimation error P0+ depends on the noise interference of different battery models and data collection. set. u003cpu003eu003c/pu003e The improved SOC estimation algorithm combines the advantages of the Ah measurement method, the open circuit voltage method and the extended Kalman filter algorithm. First, use the open circuit voltage method to provide the system with a relatively accurate initial value of the SOC state, and then repeatedly use the Ah measurement method to calculate the SOC state value, make a preliminary estimate of the current SOC state value, and then use the Kalman filter algorithm to modify Therefore, the errors of Ah measurement method and open circuit voltage method are eliminated, and an optimal estimation value of SOC state at the current moment is obtained. The algorithm not only reduces the calculation cost, but also improves the calculation accuracy, making the entire system stable and effective. u003c/pu003e
Custom message
Chat Online 编辑模式下无法使用
Chat Online inputting...
We will get back to you asap