在分析模型参考自适应预估控制系统 (MRAPC)的基础上 ,提出了一种改进算法 ,通过将自适应结构中可调增益前移的办法克服原算法在自适应过程中存在滞后的问题。同时还给出了该改进算法的全局收敛性分析结果。这一算法已在大坝电厂 30 0MW机组的主汽温、再热汽温、燃烧及协调等主要控制系统中得到成功应用 ,取得了良好的控制效果。该应用首次在国内的分散控制系统中实现了自适应预估控制算法 Based on the analysis of the MRAPC model, an improved algorithm is proposed to overcome the problem of lag in the adaptive algorithm by advancing the adjustable gain in the adaptive structure. At the same time, the global convergence of the improved algorithm is also given. This algorithm has been successfully applied to the main control systems such as main steam temperature, reheat steam temperature, combustion and coordination of the 300MW unit of Daba Power Plant and has achieved good control effects. The application for the first time in the country’s decentralized control system to achieve adaptive predictive control algorithm