The time series data of silicon content in hot metal were identified to have the chaotic feature because of the positive maximum Lyapunov exponent, and then the time scales to predict future were estimated. Finally a ...The time series data of silicon content in hot metal were identified to have the chaotic feature because of the positive maximum Lyapunov exponent, and then the time scales to predict future were estimated. Finally a chaotic local-region model was constructed to predict silicon content in hot metal with good performance due to high hitting rate.展开更多
The problem of linear parameter varying (LPV) system identification is considered based on the locally weighted technique which provides estimation of the LPV model parameters at each distinct data time point by giv...The problem of linear parameter varying (LPV) system identification is considered based on the locally weighted technique which provides estimation of the LPV model parameters at each distinct data time point by giving large weights to measurements that are "close" to the current time point and small weights to measurements "far" from the current time point. Issues such as choice of distance function, weighting function and bandwidth selection are discussed. The developed method is easy to implement and simulation results illustrate its efficiency.展开更多
文摘The time series data of silicon content in hot metal were identified to have the chaotic feature because of the positive maximum Lyapunov exponent, and then the time scales to predict future were estimated. Finally a chaotic local-region model was constructed to predict silicon content in hot metal with good performance due to high hitting rate.
基金Supported by the National Natural Science Foundation of China(10826100, 10901139 and 60964005)
文摘The problem of linear parameter varying (LPV) system identification is considered based on the locally weighted technique which provides estimation of the LPV model parameters at each distinct data time point by giving large weights to measurements that are "close" to the current time point and small weights to measurements "far" from the current time point. Issues such as choice of distance function, weighting function and bandwidth selection are discussed. The developed method is easy to implement and simulation results illustrate its efficiency.