Date of Award

12-5-2018

Document Type

Dissertation

Degree Name

Doctor of Philosophy (PhD)

Department

Systems Engineering

First Advisor

Jing Zhang

Abstract

In existing long-established power systems, the grid control and balancing are implemented on the generation side in centralized schemes. As the use of renewable, non-dispatchable, inverter-based (inertia-less) energy sources is becoming increasingly pervasive in modern power systems, this paradigm will change and the need for decentralized demand-side control will be imperative to ameliorate the grid’s dynamic performance and stability. Battery energy storage (BES) is one of the essential constituents that will empower this transition. Battery energy storage systems (BESS) are eminent for their features of scalability, mobility, high energy density, and rapid response in tracking abrupt power fluctuations. These features make BESS suitable not only for grid-scale systems but also for load-scale applications under the frameworks of dynamic demand control (DDC) and demand response control (DRC). Situated on the load side, load-scale BESS have recently been receiving a great deal of attention in a wide range of functions for the independent system operators (ISOs), utilities, and end users. The overarching goal of this work is to present a battery-buffered smart load (BBSL) to realize a combination of demand-side participation functions. The proposed BBSL uses a new control strategy to adjust the load power consumption from the grid based on two control signals: the line frequency deviation and time-of-use (ToU) rates. Therefore, the BBSL can increase the battery utilization by combining and synchronizing primary frequency control (PFC) and ToU energy arbitrage (EA) services for the ISO and end users, respectively. To investigate the realization and technical usefulness of the proposed BBSL concept, analysis and experimental tests of different smart load (SL) techniques are conducted. To this end, a scaled-down single-phase experimental system of the BBSL using Nickel-Metal Hydride (NiMH) batteries was built. The system works at a low-voltage level (24V AC and 48V DC) with a fixed-point microcontroller and a LabVIEW-supported test bench. Measurement and analysis of practical line frequency data are carried out and the battery utilization is evaluated under PFC reliability standards of the four interconnections of the North American grid. Mathematical formulation, economic feasibility, and practical implementation of different scenarios of the proposed SL strategy are discussed and compared to hourly, daily, and monthly simulations. The investigation in this work also entails the study of the battery dynamic performance under pulsed charge and discharge profiles. Therefore, experimental extraction and validation of a Thevenin equivalent circuit model of single and multiple series-connected NiMH cells are presented. In the BBSL application, the battery is connected as an energy buffer between the grid and load to control energy flow between them continually. Such charge-discharge energy transactions between the battery and load are always accompanied by energy loss. Therefore, the battery energy efficiency is a critical parameter for performance evaluation, sizing, and control design of the BBSL. This work presents a new electrical model of the battery to evaluate its energy efficiency. The model is derived from the equivalent circuit of the battery which consists of a current-dependent nonlinear resistance to model the total polarizations in the battery. The model parameters are determined by considering the effects of the battery current and state of charge. Then, a method is presented to practically evaluate the energy and power loss in the BBSL system under different PFC operational conditions. Finally, the proposed efficiency model is used to verify the experimental measurements of the battery energy loss.

Share

COinS