Study On Some Issues In PV Microgrid Control And Optimization

Forced by the pressure of the environment and energy, the use of renewable energy is attracting more and more attention. When large-scale renewable energy generation is connected to the power grid, the security and stability of the power grid will be affected because of renewable energy with the characteristics of intermittent and uncertainty, such as photovoltaic and wind power generation. Microgrid, consisting of a variety of distributed generations and loads, realizes coordinated operation between distributed generation and distributed generation and distributed generation and distritubtion grid and reduces the impact on distritubtion grid after the renewable energy generation integrated into grid.. Meanwhile, microgrid can output highly reliable supply and multiple sorts of energy for loads as well and it is effective way to achieve an active distribution grid. In this thesis, a number of key issues concerning on power control and optimization are analyzed and studied in a photovoltaic (PV) microgrid, which are related with PV power forecasting, distributed generation configuration, power flow tracking control at the point of common couple (PCC) in connected-grid, and economical optimization operation.Contributions of this thesis can be summarized as follows:(1) Renewable energy generation forecast plays an important role in maintaining micro-grid power balance and the economy. Taking PV generation for an example, two statistical data based forecasting methods are proposed in this paper. One is based on the numerical class weather forecasting and uses the system identification method to establish a PV power prediction model; another way is based on the language class weather forecast and uses the principle of adaptive neural fuzzy inference system (ANFIS) to establish a PV power prediction model. Since both methods are based on weather forecasts, with the development of the weather forecasting technology, the PV power forecast accuracy and timeliness will also continue to improve.(2) The rational configuration of distributed generations in microgrid is an important prerequisite to maximize the use of renewable energy. In this paper, based on different frequency characteristics of distributed generations and loads, a frequency domain based method is proposed to configure distributed generations in a microgrid. In this method, the spectrums of the load and renewable energy analysis are analyzed, and as a basis the type and capacity configuration of the distributed generations are determined. Although the method is presented as an example of PV microgrid, it can be applied to all kinds of microgrid with other renewable energy. (3) One of the functions is to coordinate power grid scheduling and achieve an active distribution grid. Hence the power flow tracking control of PCC is the core in the microgrid design and operation. A decouple control system architecture based on the frequency domain is proposed. With the role of the controller, each distributed generation bears the load only within a certain range of fluctuation in the frequency domain and the integrated frequency domain characteristics curve of all distributed generations covers the entire load band. Control parameter tuning is an important part to achieve control objectives. A nonlinear diffusion particle swarm optimization (NDPSO) method is applied to tune the control parameters. Frequency domain analysis is taken in the PV microgrid. Frequency domain analysis and time-domain simulation results show that the proposed control structure and strategies can realize the power flow tracking control of PCC and the nonlinear diffusion particle swarm algorithm applied to the control parameter tuning can achieve the desired goals. Empirical studies show that the proposed method can be realized, and verify the effect of different distributed generations in the microgrid.(4) For the sake of the economic optimization in the microgrid operation, an operation schedule is applied to the power flow tracking control of PCC. Based on the renewable energy generation and load forecasting, the operation schedule is mapped up with the objective of minimum operation cost and by genetic algorithm (GA). Microgrid operates abide by the schedule. When the predict value and the actual one do not match, the power flow tracking control system of PCC will carry out the real-time tracking adjustment. Simulation results show that, under the guidance of operational schedules, besides ensuring the power flow tracking, the operating economy has been improved.