论文部分内容阅读
卫星遥测参数的趋势变化状态对卫星故障预测的判断起到重要的指导作用,立足于遥测参数特性,提出等高线预测方法,从特征数据出现的时刻对参数进行趋势预测。为了提高预测精度,引入EMA(增加期望模式)误差分析模式,结合之前提出的新息灰预测模型,根据熵组合基本原理建立基于EMA熵组合预测模型。通过工程数据仿真计算,并利用预测有效度对各种算法进行评估,结果验证了基于EMA的熵组合预测模型的正确性和可靠性。
The trends of the satellite telemetry parameters play an important guiding role in judging the satellite fault prediction. Based on the characteristics of the telemetry parameters, a contour line prediction method is proposed to forecast the trend of the parameters from the appearance of the characteristic data. In order to improve the prediction accuracy, the error analysis mode of EMA (increase expected mode) is introduced. Combined with the new forecasting gray forecast model proposed earlier, the EMA entropy combined forecasting model is established based on the principle of entropy combination. Through the simulation calculation of engineering data and using the predictive validity to evaluate various algorithms, the results verify the correctness and reliability of the EMA-based entropy combination forecasting model.