A new method for model reduction and controller design of large-scale dynamical systems

This study presents a novel approach for the simplification and controller design of expansive dynamic linear time-invariant plants. This approach applies to both single input single output and multiple input multiple output models on a wide scale. The diminution approach is a straightforward method that guarantees the stability of the lower-order plant, given that the higher-order system is stable. Implementing the balanced truncation approach determines the denominator polynomial of the required system. The coefficients of the numerator polynomial are computed using a simple mathematical procedure, as outlined in the suggested scenario. The proposed method overcomes the constraints of the balanced truncation scheme while maintaining its crucial attributes, including stability, controllability, observability, passivity, etc. The strategy that has been developed guarantees the preservation of stability, time moments, Markov parameters, and other features of the higher-order plant in the reduced model. A two-input, two-output, single-machine infinite bus real-time power system model and a ninth-order system are utilized to test the accuracy and effectiveness of the proposed method. The findings of the suggested technique are compared against other popular algorithms. Furthermore, controllers are derived using the moment-matching process using the recommended lower-order plant for two real-time case studies. The controller’s design is shown, and its efficacy is confirmed using a real-time system described in the literature.