A comparison of reduced order model techniques

Four techniques for constructing reduced order models are compared. The techniques are based on: 1) truncating the balanced realization; 2) Laguerre series expansion; 3) Kautz series expansion; and 4) continued fraction expansion. These techniques was used to construct reduced order models for the same (very simple) plant transfer function. Each reduced order model is then used to design a controller to achieve the same design objectives. Finally, the performance of the closed loop control systems that result from using each controller with the full order plant are compared. Several model reduction techniques produce virtually identical models and hence virtually identical.