Modelling and Optimisation of Energy Management in Grid-Connected Microgrid Systems for University Campuses in South East Nigeria

This study develops and optimises an energy management system model for grid-connected microgrid operation on university campuses in South East Nigeria, targeting the cost and reliability challenges inherent in the current dependence on diesel backup generation. South East Nigerian university campuses typically experience grid availability below 40 percent of the time, forcing prolonged diesel generation that dominates operational energy budgets. The microgrid architecture modelled comprises a rooftop PV array, a lithium-ion battery energy storage system, the existing diesel generators, and the grid connection point, all coordinated by a model predictive control energy management algorithm. The study site is modelled on University of Nigeria Nsukka's electrical infrastructure, with load profiles derived from campus substation metering data and solar resource derived from 10 years of NASA POWER satellite irradiance data. The energy management optimisation problem is formulated as a mixed-integer linear programme minimising operating cost subject to power balance, battery state-of-charge constraints, and minimum runtime constraints on diesel generators. The MILP is implemented in MATLAB using the CPLEX solver. Simulation results across a full year demonstrate that the optimised energy management system reduces annual diesel fuel consumption by 61 percent and total annual energy costs by 44 percent compared to the current unoptimised diesel-dominant operation. Battery degradation modelling is incorporated to ensure that the optimised dispatch strategy does not excessively cycle the battery, maintaining calendar life above 10 years. Sensitivity analysis shows robustness of savings to 20 percent variation in solar irradiance and electricity tariff inputs. Keywords: microgrid, energy management, model predictive control, optimisation, South East Nigeria.