多目标离散控制:一种对策方法

Multiple Objective Discrete Control——A Game Method

在多目标控制中，线性加权法常常被用于把多目标问题转化为单目标问题的研究中。若各子目标与总体目标存在非凸关系时，该方法难以有效。本文构造了一种对策模型来解决这类多目标控制问题，提出了利用对策理论求解这类问题的方法。该方法可将决策人对目标的偏好加入模型中，从而放宽了问题的凸性限制。

In multiple objective control fields, linear weighted method is often used to convert the multiple objective problem to the single objective problem. If nonconvex relation between total objective and its performance indices, the LWM is not effective. In this paper, the gaming model of the multiple objective control problem is built. The method solving the problem is presented by using dynamic game theory. The objective perforences of decision maker is considered in the model, which helps to overcome the difficulty incurred by nonconvexity.