Energy management system for PV-BESS-diesel generator islanded microgrid using model predictive control

Abstract

Islanded microgrid is identified as a cost-effective solution to provide electricity to the off-grid areas which are far from the main electricity grid. However, intermittent nature of renewable energy source (RES) and unpredictable nature of load demand can cause an imbalance in the islanded microgrid. In islanded microgrid, components such as battery energy storage system (BESS) and diesel generator can compensate this problem. Diesel generator, especially, can provide auxiliary energy and act as the main grid in the islanded microgrid. Nonetheless, the inefficient utilisation of diesel generator can lead to unnecessary operation cost, shorten the life span and most importantly, contributing to the environmental pollution. Hence, energy management system (EMS) is employed to operate and manage the microgrid efficiently. Model predictive control (MPC) which has grown in popularity in recent years will be integrated with the EMS. The focus of this research is to reduce the use of diesel generator through efficient microgrid management utilising MPC-based EMS. It is proposed that load curtailment and greenhouse gas emissions cost function are included in the EMS to reduce the cost of the diesel generator operation. Two test cases are designed. The first test case compares the performance MPC-based EMS with and without load curtailment to evaluate the diesel generator cost savings. The second test case will observe the impact of the cost functions especially diesel fuel consumption cost and greenhouse gas emission cost influence the cost saving of diesel generator. The MPC-based EMS is simulated using MATLAB software. The optimisation problem of the MPC-based EMS is formulated using mixed integer quadratic programming (MIQP) and is solved using CPLEX. The results shows that the reduction of diesel fuel cost and consumption by 52.21% in MPC-based EMS utilising load curtailment compared to without load curtailment. In the second test case, it is found the second order polynomial diesel fuel consumption cost function is sufficed in reducing fuel cost by 34% compared to combining both diesel fuel consumption cost function and greenhouse gas emission cost function. In contrast, diesel fuel cost and consumption increased dramatically by 178%, despite a trivial improvement in greenhouse gas emission cost. Hence, it can be concluded that MPC-based EMS with load curtailment is able to improve the cost saving of the diesel generator. In addition, second order polynomial diesel fuel consumption cost function is sufficed to drive down the usage of diesel generator. However, load curtailment, particularly demand side management strategies, must be enhanced for islanded or standalone microgrids where incentive-based method cannot be applied.