一种降低计算负荷的快速预测控制算法（英文）

A fast algorithm for reducing the computation burden of model predictive control

大多数模型预测控制算法因求解优化问题时在规定的采样时间内需要在线处理矩阵乘法操作而存在繁重的计算负荷问题。为了解决这一问题,本文提出了一种降低计算负荷的快速预测控制算法,将实时观测信号调制成正负一的比特流序列来简化传统的乘法运算。同时,将控制时域内未来时刻的控制律序列通过线性反馈控制律离线递推,用当前时刻控制律近似替代,从而降低了求解优化问题时的矩阵维数。

Most of the industrial model-based predictive control algorithms are suffering from heavy computation burden when solving the QR optimizations and on-line matrixes multiplication within a limited sampling step. In order to optimize this problem, a fast algorithm is proposed, which the real-time values are modulated into bit streams to simplify the multiplication as the bit based operation could extremely decrease the compute consumption. In addition, the control variables are deduced from the prediction horizon to the current control actuation approximately by a recursive relation instead of figuring all of the control actuations out strictly to reduce the matrix dimension.