抄録
DC motor drive systems are still widely used in many industrial applications because of their simpler controllers compared with AC motor systems. To realize the desired control performances, system parameters are important for modeling and analysis. Usually, some parameters of a motor are given by manufactures, and others can be measured or calculated by users. But from the standpoint of the entire systems, it is hard to measure the parameters collectively. Especially, since carbon brushes are used in DC motors, their contact resistance and voltage drop vary depending on running condition. In this paper, we propose a parameter estimation method on the basis of the theory of least mean square method. In this method, we use a model of DC motor which takes account of the voltage drop, and introduce block pulse functions which is one of the ways to approximate function as a linear combination of a set of orthogonal basis functions. The parameters of the entire system can be estimated on-line. We use a Transputer to realize on-line estimation. Experimental results shows the validity of the estimation method.
