摘要
光伏发电系统中蓄电池状态是确定仅对本地负载供电或将电能上传电网的重要依据。为了改善光伏发电系统中蓄电池状态估计存在的精度低,跟踪速度慢,易积累误差和易发散等问题,提出了一种改进的Sigma-point卡尔曼滤波的状态估计方法。该方法基于Sigma-point卡尔曼滤波,在其滤波过程中引入渐消因子,对非当前滤波值进行渐消,以修正获取的当前滤波数据值,使之保持滤波的实时性。仿真对比研究显示,该方法的估计精度更高,跟踪速度更快,波形更平稳,其对蓄电池的荷电状态(SOC)和开路电压估计的相对误差最大值分别为3.256%和4.610%。仿真研究结果表明,所提出的方法应用于蓄电池状态的估计是可行的。
State estimate for battery in photovoltaic power generation system is an important basis that determines whether the power is supplied only for local load or uploading to the grid. In order to overcome the puzzles of being low in accuracy, slow in track speed, easy to accumulating error and divergence, the paper proposed a sort of state estimation method based on the improved Sigma-point Kalman filter. Based on principle of the Sigma-point Kalman filter, it introduced diminishing factor to fade the non-current filtering value, and modifying the current filtering value in filtering process for keeping filtering real-time. The simulation comparative experiments showed that the method would be higher in estimate accuracy, faster in track speed, more stable in waveform, and the relative errors of maximum estimation in SOC and open-circuit voltage would be respectively 3.256% and 4.610%. Simulation researches show that the proposed state estimate method is reasonable and feasible.
出处
《电气技术》
2015年第6期67-72,共6页
Electrical Engineering
关键词
光伏发电系统
状态估计
卡尔曼滤波
渐消因子
photovoltaic power generation system
state estimate
Kalman filtering
diminishing factor
