Single-phase grid connected inverter with DC link voltage modulation

Abstract

Nowadays, the African continent is considered as a market with a lot of investment opportunities and it is characterized by prosperity and growing development. However, there is a serious corresponding sustainable energy need. The origin of this phenomenon is that the middle class in Africa, had a remarkable increase during a couple of decades ago. Moreover, people moved to the big cities. Consequently, more daily basic needs become on high demand. Most of them are food, clean water, infrastructure and energy. Developed countries such as China and European countries have started to invest in energy sector in Africa. So, the need is not limited to the other parts of the world where there is a strong tendency towards clean and affordable energy. The renewable energy sources are a key solution towards a substantial reduction of greenhouse gas emissions. The Sub-Saharan Africa has shown a high rate of use of wood for cooking and a lower number of people who have access to electricity. This has impact on the deforestation and the CO2 emission increases due not only on the cooking activity but also on transport of the wood. Many woods must be imported from neighboring countries, that import needs fuel for trucks and increases traffic. The exploitation of the solar energy can reduce the use of wood. This will affect the indoor air quality for people using the solar energy while cooking. Indeed, the indoor air quality is under discussion, while using charcoal or wood for cooking. The rate of access to electricity has increased a lot, but consequently it should be followed up with the use of renewable energy. This is especially needed in Rwanda. Many manufacturers from all over the world have made the PV (photovoltaic) modules affordable, as a lot of investments, research and development have been done to achieve it. However, the total cost of a PV system includes also the cost of the inverter, the cost of installing and grid connection. The grid connection cost, and installation cost are mainly depending on the local labor cost. The part of cost of ownership of the inverter becomes higher than the PV panels themselves due to the cost of the inverter combined with a typical lifetime of 10-12 years. This lifetime can be even less in tropical areas. Rwanda as well as in the other parts of the World experience a photovoltaic revolution in grid and standalone systems. Some countries use open land for the PV. Rwanda with small land and high population density should use the house roofs in order to save land and protect the vegetation. This aspect is somewhat comparable with Flanders. From the present research side, it is good to investigate if other inverter topologies and control strategies might have an advantage. This might give indeed opportunities, but in the meantime, almost the whole problem setting of gridconnected inverters is understood. If, for grid tied PV inverters, a longer life could be combined with a lower or equal cost, it will reduce the total cost of ownership. One could look at the possibilities in choice of topology, components and the necessary control to achieve this. Today most of inverters using classical topologies, contain many electrolytic capacitors. By changing the topology and gate drives, one could reduce the number of electrolytic capacitors to only one, at a less critical place. Most inverter designers use electrolytic capacitors in many places as they were just used to do so. However, this “habit” reduces the lifetime of the inverter. This is not the only lifetime limiting element. They are other elements such as automotive grade processors, contact corrosion protection, and lightening protection. Of course, a “good engineering” also remains a key element for a longer life design for an inverter. The present work aims at developing a single-phase gridconnected inverter without an electrolytic capacitor at DClink. The following objectives have been targets of the PhD work:  Propose an inverter topology that can withstand the harsh working environment in Rwanda: use less electrolytic capacitors and a design where an electrolytic capacitor can be easily replaced;  Study the stability of the topology;  A gate driver of IGBTs without electrolytic capacitors;  An adapted control of the inverter is needed. The thesis has eight chapters. The first chapter is an introductory part. It shows the situation of the use of the PV system as renewable energy source in Rwanda. It discusses the source of CO2 emission and the effect of using renewable energy on reducing it. It shows social and environmental impacts of integrating PV systems into traditional electrical network. The chapter also shows the parameters that can be based on in order to calculate the cost ownership of a PV system. The second chapter gives a literature review of a classical single-phase topology. It discusses different parts of a single-phase grid-connected inverter. It shows the different PWM schemes that are used for the single-phase inverter. It is in this chapter that the state of art is deeply identified. The third chapter proposes a topology that is based on a design of an inverter using a topology that replaces an electrolytic capacitor in DC-link by a much smaller polypropylene capacitor in the DC-link. It shows the benefit of using the polypropylene capacitors instead of electrolytic capacitors. The proposed topology can use a standard three-phase transistor bridge for a single-phase grid connected inverter. The bridge is used in a different way for single-phase injection with one leg for boost and two legs for an H-bridge. It is in this chapter that a DC voltage modulation technique is analyzed. Moreover, the calculation of RMS currents for the proposed topology compared to the classical topology is presented in this chapter. Chapter 4 analyses the stability of the DC-link of the proposed topology, partly using numerical inverse Laplace transform. The stability is not granted due to an inherent resonance of the topology. The validation of the stability was done by simulation as well as by a lab experiment. The fifth chapter proposes frequency synchronization using a Double Integration Method (DIM) after a brief discussion about drawbacks of the typical phase locked loop (PPL) control for single-phase inverters. The chapter presents some simulation, and lab results related to the DIM. The sixth chapter presents a design of an IGBT gate driver. It proposes a circuit with desaturation protection using the undervoltage protection of the gate driver. This chapter presents the simulation and lab results of the designed IGBT gate driver. The seventh chapter focuses on testing the whole system. In this chapter, the system was connected to the grid to test the inductor current control. Chapter 8 draws some conclusions and recommendations for the future work.