Multi-Objective Optimal Sizing and Operation Control of Microgrid-Connected Batteryless Energy System Under Time of Use Tari

Abstract

Many countries are exponentially faced with high demand of electricity due to factors such as high living standards, industrialization, population growth, among other factors. As a result, the integration of renewable energy sources at distribution power level systems has increased, mainly because of global warming concerns. Modern power systems are geared towards distributed energy resources (DERs) at the distribution level networks. In this sense, microgrids are the main core infrastructure of today's modern power system. Microgrids consists of interconnection of distributed energy resources such as renewable energy sources, storage energy systems, diesel generators, and loads. In recent years, microgrids based on photovoltaic energy system are becoming popular for electri cation both for grid-connected mode and isolated mode. However, the deployment of a microgrid photovoltaic system brings many challenges due to stochastic nature of photovoltaic energy source and variability of load demand. This is because power generation must always match the load demand. Therefore, the main issues related to microgrid-connected photovoltaic systems are the optimization design parameters, planning and operation control schemes which seek to minimize or maximize prede ned objective functions subject to technical and operational constraints. Importantly, designing a cost-e ective microgrid-connected photovoltaic system alongside the dispatching and scheduling the power ows is very crucial in modern power system. Accordingly, this thesis is mainly divided into two major parts namely, optimal sizing of grid-connected photovoltaic batteryless energy system and optimal operation control of microgrid-connected photovoltaic-diesel generator backup energy system. In the rst part, a multi-objective optimal sizing grid-connected photovoltaic batteryless system that seeks to determine the most cost-e ective photovoltaic (PV) system size that maximizes the reliability requirements while lowering the power sold to the grid utility. This is because there is no power purchase agreement (PPA) within this jurisdiction. The economic analysis is expressed in terms of total life cycle cost and the microgrid reliability is measured by the loss of power supply probability (LPSP). On the other hand, optimal operation control in conjunction with demand side model (DSM) is carried out to improve the operational e ciency and resilience of microgrid-connected photovoltaic-diesel generator energy system. Time of use tari (TOU) is the type of DSM strategy considered in this research. Essentially, an open loop optimal and a closed-loop control are suitably designed. The open loop scheme takes into consideration the non-linearity of the diesel generator fuel consumption; and, the FMINCON algorithm in MATLAB is used to carry out the optimization problem. The closed-loop system is based on economic model predictive control (EMPC) solved using linear programming (LP) in OPTI Toolbox. In both open loop and closed-loop strategies, operational e ciency and energy e ciency are considerably improved. Notwithstanding, each of the two control schemes exhibits its pros and cons. On one hand, the open loop optimal control strategy is not complex and is stable. Thus, it is easier and cheaper to implement but cannot handle uncertainties and disturbances within the microgrid energy systems. On the other hand, the closed-loop EMPC strategy is complex and expensive to implement but present great robustness against uncertainties and disturbances. From the results obtained, the optimal number is 354 photovoltaic panels and the total life cycle cost of the system is found to be 191630$ over 25 years lifespan of the project. From the optimal operation control point of view, the daily energy saving is increased up to 52.1 % in intermittent connected mode (ICM) while the diesel energy not delivered increases to 84.8 % in intermittent mode (IM). The results of this thesis are evidently crucial for designers, decision-makers, performances analyzers, and control agents who are struggling with multiple objectives to make appropriate trade-o s for grid-connected photovoltaic systems.