卡尔曼滤波算法在焦炭产量预测中的应用

Application Research on Coke Production Based on Kalman Filtering Algorithm

焦炭产量预测是进行焦炭市场分析的必要环节。本文选用焦炭产量作为唯一的状态变量,利用卡尔曼滤波算法,通过每一时刻的新息进行不间断的迭代,以获得焦炭产量最优的真实状态估计值。通过实证分析表明,卡尔曼滤波算法能够较准确地跟踪和估测焦炭产量。

The research about forecasting the coke production is significative.The kalman filtering algorithm is applied to study the coke production,and the only state variable coke production is picked out to construct the State Space Model.The algorithm makes use of new information composed of the difference between the observed variable and forecasted variable.Through the continuous updates and iteration,the algorithm can help us obtain the optimal estimated value.The empirical analysis shows that the algorithm is effective in tracking and forecasting the coke production.