Multi-Verse Optimization for Two Stage CMOS Operational Amplifier Circuit Design

This research work proposes two variants of Multi-verse optimization (MVO) for optimal design of two stage CMOS opamp circuit. MVO is a nature-inspired global optimization algorithm, which uses the mathematical modelling of multi-verse (many universes) theory. Exploitation capability of MVO algorithm is enhanced by Brownian movement and cross-entropy (CE) in the two proposed versions of MVO namely MVO-BR and MVO-CE. Also, other popular metaheuristic optimization algorithms like differential evolution (DE), particle swarm optimization (PSO) and artificial bee colony (ABC) are applied to the selected design problem for verifying the validity of the proposed MVO based algorithms. Performance of the base MVO and the two enhanced versions are rigorously tested for optimality, robustness and convergence on the design of two stage opamp circuit. To compare the effectiveness of various algorithms, a new performance metric called “overall figure of merit” is applied along with existing metrics. By comparing the numerical results, it is demonstrated that the proposed MVO-CE algorithm provides high-quality design of two-stage opamp reliably with faster convergence than other algorithms.