Biogeography-based learning particle swarm optimization for combined heat and power economic dispatch problem

Combined heat and power economic dispatch (CHPED) is an important optimization task in the economic operation of power systems. The interdependence of heat and power outputs of cogeneration units and valve-point effects of thermal units impose non-convexity, nonlinearity and complication in the dispatch modeling and optimization. In this paper, a novel PSO algorithm called biogeography-based learning particle swarm optimization (BLPSO) is applied to solve the CHPED problem considering various constraints including power output balance, heat production balance, feasible operation area of cogeneration unit and prohibited operation zones. In BLPSO, based on a biogeography-based learning model, each particle uses a migration operator to update itself based on the personal best position of all particles. This updating strategy helps BLPSO overcome premature convergence and improve solution accuracy. Moreover, a repair technique is employed to handle the system constraints and guide the solutions toward feasible zones. The effectiveness of the proposed method is evaluated by testing on four CHPED problems containing 5, 7, 24, and 48 units. The experimental results show that BLPSO outperforms the state-of-the-art methods in terms of solution accuracy and stability. Therefore, BLPSO can be regarded as a promising alternative for the CHPED problem. • BLPSO algorithm is applied for solving CHPED problem with various constraints. • The interdependence of heat and power outputs of cogeneration units impose great complication. • Non-convex CHPED problems with/without prohibited operating zones are considered. • Comprehensive simulation results demonstrate the effectiveness of the BLPSO algorithm.