A robust and stochastic operational framework for a grid-connected biomass/photovoltaic hybrid system with battery storage using exploration-exploitation approach

In power system, optimal operation of grid-connected hybrid systems is a vital optimization problem in which the presence of multiple uncertain parameters is a challenging issue that affects the result of the operation. By modeling all uncertainties, operators can better predict extreme scenarios, optimize energy management and maintain higher reliability and cost-effectiveness during daily operation. This paper proposes a stochastic framework for the optimal operation of a biomass/photovoltaic (PV) hybrid system with battery storage in which load, solar radiation and electricity price uncertainties are simultaneously considered. To model these uncertainties, full-factorial design is employed as an efficient scenario generation approach. To optimally determine the power purchased from the grid, an objective function is defined that considers the cost of energy purchased from the grid, the cost of dumped energy, the cost of biomass generation and the revenue from selling excess PV power to the grid. For different reliability levels, defined by loss of power supply probability (LPSP), exploration-exploitation-based approach (E2), as a recently developed basic optimization method, is used to solve the operational problem and the results are compared with those obtained using particle swarm optimization (PSO) and crow search algorithm (CSA). Over the case study, simulated results reveal that increasing the desired LPSP from 0 to 0.02 will reduce the operating cost by approximately 16.7% while increasing the desired LPSP from 0.02 to 0.05 results in a reduction of about 3.8%. When the desired LPSP increases from 0 to 0.05, the average biomass cost obtained across different scenarios increases by 60.66%. Furthermore, the results of the two-tailed t-test indicate that E2 significantly outperforms both CSA and PSO.