Access Type

Open Access Dissertation

Date of Award

January 2016

Degree Type


Degree Name



Electrical and Computer Engineering

First Advisor

Caisheng Wang


Installation of photovoltaic (PV) units could lead to great challenges to the existing electrical systems. Issues such as voltage rise, protection coordination, islanding detection, harmonics, increased or changed short-circuit levels, etc., need to be carefully addressed before we can see a wide adoption of this environmentally friendly technology. Voltage rise or overvoltage issues are of particular importance to be addressed for deploying more PV systems to distribution networks. This dissertation proposes a comprehensive solution to deal with the voltage violations in distribution networks, from controlling PV power outputs and electricity consumption of smart appliances in real time to optimal placement of PVs at the planning stage. The dissertation is composed of three parts: the literature review, the work that has already been done and the future research tasks.

An overview on renewable energy generation and its challenges are given in Chapter 1. The overall literature survey, motivation and the scope of study are also outlined in the chapter. Detailed literature reviews are given in the rest of chapters. The overvoltage and undervoltage phenomena in typical distribution networks with integration of PVs are further explained in Chapter 2. Possible approaches for voltage quality control are also discussed in this chapter, followed by the discussion on the importance of the load management for PHEVs and appliances and its benefits to electric utilities and end users. A new real power capping method is presented in Chapter 3 to prevent overvoltage by adaptively setting the power caps for PV inverters in real time. The proposed method can maintain voltage profiles below a pre-set upper limit while maximizing the PV generation and fairly distributing the real power curtailments among all the PV systems in the network. As a result, each of the PV systems in the network has equal opportunity to generate electricity and shares the responsibility of voltage regulation. The method does not require global information and can be implemented either under a centralized supervisory control scheme or in a distributed way via consensus control. Chapter 4 investigates autonomous operation schedules for three types of intelligent appliances (or residential controllable loads) without receiving external signals for cost saving and for assisting the management of possible photovoltaic generation systems installed in the same distribution network. The three types of controllable loads studied in the chapter are electric water heaters, refrigerators deicing loads, and dishwashers, respectively. Chapter 5 investigates the method to mitigate overvoltage issues at the planning stage. A probabilistic method is presented in the chapter to evaluate the overvoltage risk in a distribution network with different PV capacity sizes under different load levels. Kolmogorov–Smirnov test (K–S test) is used to identify the most proper probability distributions for solar irradiance in different months. To increase accuracy, an iterative process is used to obtain the maximum allowable injection of active power from PVs. Conclusion and discussions on future work are given in Chapter 6.