Optimal switchable load sizing and scheduling for standalone renewable energy systems

• Published in: Solar energy Vol. 144; pp. 707 - 720
• Main Authors: Habib, Abdulelah H., Disfani, Vahid R., Kleissl, Jan, de Callafon, Raymond A.
• Format: Journal Article
• Language: English
• Published: New York Elsevier Ltd 01-03-2017; Pergamon Press Inc
• Subjects: Algorithms; Energy storage; Inequality-constrained least square; Load management; Mixed-integer linear programming; Off-grid solar energy; Optimization; Photovoltaics; Renewable energy; Scheduling; Solar energy; Solar power; Storage; Inequality-constrained least square; Off-grid solar energy; Load management; Mixed-integer linear programming
• ISSN: 0038-092X; 1471-1257
• DOI: 10.1016/j.solener.2017.01.065

•Loads are sized for planning at off-grid PV sites.•A finite number of discrete static (on or off) electric loads is sized.•The optimization maximizes the solar energy consumed by the loads.•Analytical solutions and computational algorithms are proposed and compared.•Analytical methods are limited to clear days and 3 or less loads.•Mixed-integer linear programming and other algorithms produce similar results but differ in flexibility and computational cost.
The variability of solar energy in off-grid systems dictates the sizing of energy storage systems along with the sizing and scheduling of loads present in the off-grid system. Unfortunately, energy storage may be costly, while frequent switching of loads in the absence of an energy storage system causes wear and tear and should be avoided. Yet, the amount of solar energy utilized should be maximized and the problem of finding the optimal static load size of a finite number of discrete electric loads on the basis of a load response optimization is considered in this paper. The objective of the optimization is to maximize solar energy utilization without the need for costly energy storage systems in an off-grid system. Conceptual and real data for solar photovoltaic power production provides the power input to the off-grid system. Given the number of units, the following analytical solutions and computational algorithms are proposed to compute the optimal load size of each unit: mixed-integer linear programming and constrained least squares. Based on the available solar power profile, the algorithms select the optimal on/off switch times and maximize solar energy utilization by computing the optimal static load sizes. The effectiveness of the algorithms is compared using one year of solar power data from San Diego, California and Thuwal, Saudi Arabia. It is shown that the annual system solar energy utilization is optimized to 73% when using two loads and can be boosted up to 98% using a six load configuration.