Date of Award

May 2018

Degree Type


Degree Name

Doctor of Philosophy



First Advisor

Chris Yuan

Committee Members

Benjamin Church, John Reisel, Ryoichi Amano, Wilkistar Otieno


Battery energy storage, Daylighting, Demand side management, Energy efficiency, Peak demand reduction, Sharing


This dissertation aims to investigate demand side management (DSM) to improve building energy efficiency and reduce on-peak electricity demand. Daylighting device and battery energy storage system (BESS) for peak shaving are proposed and studied. To reduce the cost and improve the efficiency of building lighting, an innovative parabolic trough solar lighting and thermal (PTL/T) system is designed and analyzed. In the proposed system, a parabolic trough collector (PTC) controlled by two-axis solar tracking system is used as solar collector. The collected sunlight is split by a cold mirror into visible light and infrared light. The visible light is reflected by the cold mirror, re-concentrated by a second-stage Fresnel lens, and then delivered by plastic optical fiber to the buildings for daylighting. The infrared light goes through the cold mirror, reaches the thermal system, and is used for heat generation. An economic and efficiency model of the PTL/T system is built to optimize the system parameters. A case study is conducted to get a specific optimized illumination area and PTC area. Its maximum energy savings, and simple payback period in the US are calculated to demonstrate its economic feasibility. The results show the proposed PTL/T system is competitive compared with traditional solar energy systems.

Peak demand charge is a significant portion of building utility cost, and battery energy storage system (BESS) is recognized as an effective technology for peak demand reduction. In this research, a mathematical model is developed to analyze the economic benefit of using lithium-ion BESS for peak shaving. The impacts of four key factors including BESS capacity, battery degradation, operation temperature, and utilization rate are quantified and compared for the first time. The simulation results show that the Net Present Value of Unit BESS (NPVU) decreases with the increase of Normalized Size Percentage (NSP). The benefit from optimizing the BESS capacity is limited by relatively low NPV because only around 4% NSP can be chosen. Due to the extremely low Utilization Rate (UR), the final cycling loss is at a super low level. The average NPVU can be improved by around 7.8% through the optimization of operation temperature in the US. In contrast, a 69.9% increase of NPVU is obtained through the increase of UR by peak-time charging. Through the comparison of all the analyzed factors, the UR is recognized as the most significant factor.

After, two new DSMs based on sharing economy are proposed to maximize the utilization and savings of BESS. Peak shaving is the base function of all proposed sharing network. The simple sharing, which shares one BESS among multiple customers, is capable to work as a cloud service. The demand reduction for each customer is optimized using genetic algorithm (GA). Simulation results show the sharing among six customers can increase max NSP, optimized NSP, NPVU and UR by 97.9%, 102.5%, 515.3% and 180.6% respectively without significant battery life impact. The comprehensive sharing, which shares one BESS for different services, works in modularized method. All the services that can be conveniently applied on the customer side are evaluated, which include TOU peak load shifting, integration of PTL/T, power factor correction and combine meter effect. Economic analysis shows all the modules can provide considerable additional savings to the DSM system.