Study on a new metaheuristic algorithm – Tribal intelligent evolution optimization and its application in optimal control of cooling plants

This study proposes a new metaheuristic algorithm called Tribal Intelligent Evolution Optimization (TIEO). This algorithm is inspired by the process of continuous division and unification of tribes in history. The fundamental theory and structure of the TIEO are illustrated in detail. The superiority of the algorithm has been verified against a set of standard test functions through experiments and comparisons with the other optimization algorithms such as PSO, DE, WPS, SSA, SBO, SCA, GSA, etc. Through the comprehensive evaluation of optimization accuracy, standard deviation and calculation time, the proposed TIEO algorithm exhibits the best performance among the typical optimization algorithms. Furthermore, a sensitivity analysis of key hyperparameters in the new algorithm has been conducted, and the recommended ranges for each hyperparameter are provided. The iterative formula of the algorithm has been discussed from a mathematical perspective, and the probability of improving parameter fitness has been tested. Finally, the proposed TIEO has been successfully applied to the optimal control of a central HVAC cooling plant, achieving energy-saving rates of 19.43 %, 12.94 %, and 17.9 % under low, medium, and high load conditions, respectively, which are higher than those based on the other standard optimization algorithms. This work contributes to the development of the metaheuristic optimization algorithm.