Multiobjective Optimization for PI-based Control Strategies for HVAC Systems in Coaches

Abstract

Thermal comfort in vehicles is crucial for several reasons, including safety, health and wellbeing of the passengers, energy efficiency and customer satisfaction. Heating, Ventilation and Air Conditioning (HVAC) systems and their control play an important role in achieving comfortable thermal conditions, so one of the most important steps to carry out is a correct tuning of the control loop parameters. In this work, two heating control strategies, specifically designed for coaches, are presented. The proposed schemes are based on PI controllers whose parameters are tuned using Multiobjective Optimization (MOO), where several cost functions related to cabin comfort and energy consumption (crucial in electric coaches) are used. For the tuning task, firstly, a multidimensional Pareto front approximation is obtained based on a set of metrics, such as the Integral Square Error (ISE) and the Total Variation (TV), among others. The set of optimal solutions is collected, and secondly, a Multiple Criteria Decision Making (MCDM) method is used to select an optimal solution from the Pareto set, specifically, the Technique for Order of Preference by Similarity to Ideal Solution (TOPSIS) method. The control schemes are compared by simulation using the Matlab/Simulink software. The results show good performance of both control strategies in reference tracking and the usefulness of the MOO approach in the tuning procedure.