CHAGSKODE Algorithm for Solving Real World Constrained Optimization Problems

Abstract

Real-world optimization problems are often gov- erned by one or more constraints. Over the last few decades, extensive research has been performed in Constrained Opti- mization Problems (COPs) fueled by advances in computational power. In particular, Evolutionary Algorithms (EAs) are a preferred tool for practitioners for solving these COPs within practicable time limits. We propose a novel hybrid Evolutionary Algorithm based on the Differential Evolution algorithm and Adaptive Parameter Gaining Sharing Knowledge-based algo- rithm to solve global real-world constrained parameter space. The proposed CHAGSKODE algorithm leverages the power of multiple adaptation strategies concerning the control parameters, search mechanisms, as well as uses knowledge sharing between junior and senior phases. We test our method on the benchmark functions taken from the CEC2020 special session & competition on real-world constrained optimization. Experimental results indicate that CHAGSKODE is able to achieve state-of-the- art performance on real-world constrained global optimization when compared against other well-known real-world constrained optimizers.

A QUBO model-based approach for solving the computational resource allocation problem and its application analysis

“…The quadratic unconstrained binary optimization approach improves the quality of the solution and the running time by converting the constraints of the problem into a cost function through preprocessing. A major advantage of quadratic unconstrained binary optimization is that it provides a unified algorithmic framework suitable for solving many types of problems [3].…”