Modelling and analysis of microgrid control techniques for grid stabilisation

Abstract

In recent times, renewable energy-based distributed generation (DG) has captivated the industrial sector and on a global scale this has become a leading research area. Distributed generation using wind, solar energy or biomass as a source of energy can produce electricity on a small scale. Therefore, there is a strong focus on using renewable energy as a safe alternative source of energy, especially because it can in future play a dominant role in the world’s energy production and help to tackle the increase of global warming caused by fossil energy. However, a major problem facing renewable energies is that they are highly dependent on weather conditions. Since the power generated by DG, as well as consumption, depends on the weather conditions, irregularity of production and consumption leads to frequency and voltage fluctuations, and it can become difficult to determine and monitor consumer usage at any given time. Distributed generation can then be subjected to discrepancies in consumer usage and this can lead to severe overloading. As a result, microgrids powered by DG, operating in a single, stand-alone controllable system mode, face new challenges in terms of balancing a cluster of loads. Balancing a cluster of loads by making sure at all times that the entire system operates without overloading, is an essential requirement for the proper operation of a power system. The microgrid load considered in this project is the sum of sensitive and non-sensitive loads, respectively 5 kW and 100 kW, which constitute load requirement of one village; this total load required by a number of villages is called a cluster load. Depending on the input power generated by a DG-based photovoltaic (PV) system, these loads can be controlled using a logic control switch (LCS). When the power produced is less than the minimum load required by a component of a cluster, overloading occurs. The purpose of using an LCS is to ensure that a stable system is maintained under various loads and resource conditions. An LCS is used to continuously monitor and adjust load through circuit breakers. It is a good alternative to load balancing for a cluster of villages in rural area where a microgrid is operating in stand-alone mode. The focus of this research is to design a photovoltaic system with a maximum capacity of 1 MW providing power to a cluster of rural villages, and operating in stand-alone mode, and then to apply different control techniques (droop control, dq0 reference frame + proportional integral (PI) controller, and PI controller alone) at the inverter terminal of the PV system, in order to evaluate the stability of the output voltage. Another goal of the research is to develop an energy management system (EMS) algorithm to support the PV system in reducing loads. Therefore, a iii stable system under various load and resource conditions, as well as suitable control mechanisms are required to model a PV system. There is a need for the modelling of a PV array using a physical modelling block in MATLAB (SIMULINK) software. The state flow provided by SIMULINK is used in this project to develop an algorithm for load balancing. The state flow gives possibilities of modelling complex algorithms by combining graphical and tabular representations to create sequential decision logic, derived from state transition diagrams and tables, flow charts and truth tables. Furthermore, the design of a microgrid using photovoltaic DG and an energy management system, has been developed. The present work mainly consists of a stand-alone microgrid operation, where the power generated must be equal to the load power. In addition, different control methods, consisting of a dq0 reference frame + PI controller, are analysed at the invertor terminal. Subsequently an LCS algorithm is developed; this is required to maintain the system within certain limits and prevents overloading. LCS algorithms are based on a flowchart and allow switching automatically selected loads, depending on the power (solar radiation) available. In addition, a flow chart provides an easy way of using a graphical transition state and state chart to establish a set of rules for the system. The simulation results show that both droop control and a dq0 reference frame + PI controller are much better than a PI controller alone; these results also compared well with similar studies found in the literature. Also, these results are further improved with an EMS in order to maintain the output voltage of the microgrid, by switching on and off certain loads depending on the input power. The modelling of the microgrid using DG, based on photovoltaic systems with a maximum capacity of 1 MW, supports and improves the PV system by reducing loads. Moreover, droop control, and dq0 transformation + PI control present a better result than PI controller alone.