Some control aspects of distributed power electronics equipment

Abstract

Instantaneous balance of electric “power generation” and “power demand” is a fundamental requirement for power system stability. In the emerging power grids with increasing penetration of distributed renewable energy generation of intermittent nature, the total power generation/supply becomes difficult to predict and control in real-time. This situation will be worse when the penetration of intermittent renewable energy becomes substantial. Under such a situation, the supply side management may be insufficient to tackle this new challenge. An alternative control paradigm called demand following generation (demand-side management) could be an effective supplement to the supply side management. Various demand response methods have been proposed, such as flexible pricing, direct load control, and load scheduling. Electric spring (ES) has been proposed as a new demand response technology with fast dynamic. The previous research has explored the function of ES in voltage regulation, frequency recovery, power quality improvement, and power imbalance compensation. Most of the previous work focuses on the modeling and control of a single ES circuit. However, the future application scenario of ES may require accurate coordination among distributed ESs. There is a research gap in the control paradigm shift from individual control to group control. Though droop control is proven to be effective in improving the performance of distributed ESs without communication network, accurate power sharing among ESs is not guaranteed. In the future smart grid, the development of communication technologies makes the information exchange between different power electronics units possible. There are new features that have not been explored by information-free control. Consensus control is a cooperative control method developed for ensuring a global agreement regarding a common quantity of interest through limited information exchange between neighbors. The communication links can be very sparse. Each agent has access to its local measurement and neighbors’ information only. Compared with centralized control, consensus control is robust against single-point failure because no global information is required. In this thesis, the research line starts with the practical evaluation of the consensus control of distributed ESs. In chapter 2, a consensus control strategy is proposed to coordinate distributed ESs for voltage/frequency regulation and reactive/active power sharing purposes. Successful tests on the hardware platform serve as the basis for the future large scale simulation study. Chapter 3 is a comparison study between consensus control and droop control for ES. Droop control can serve as the last defense when communication network breaks down. Then in chapter 4, we expand the small scale test to a system-level study. An estimated dynamic model of ES is firstly constructed with order-reduce technology. This order-reduced model is suitable for system-level simulation with high accuracy and low computation cost. Based on the dynamic model of ES, a distributed optimized control for ES based domestic heating load is introduced. Every coin has two sides. The extra layer of communication exposes the system to potential cyber-attacks. In chapter 5, a general cyber-attack detection method is built based on the state observer. Case studies of both ES and DC microgrid are discussed in detail.