| Since the consume of fossil fuels is increasing, along with the increasing proportion of the new energy in the power system, the traditional AC(Alternating Current) grid can not meet the demands in admitting large-scale new energy into power system. While the DC(Direct Current) grid can solve the new energy’s admission problem effectively.DC grid’s development is incomplete at present, its topology and control objectives are uncertain. In this thesis, the tree topology with six ports is used as the topology of DC grid, the stability control problem of DC voltage is studied, the power loss optimization problem is studied. DC grid’s mathematical model is established with the bond graph method. A state feedback controller and a linear quadratic optimal controller are designed to control the DC voltage, particle swarm optimization algorithm is used to optimize the network power loss, these control algorithms are simulated in Matlab, and the effectiveness of these control methods is verified. The main work is done as follows:1. It has been researched that six ports tree topology can describe the DC grid topology well. In this thesis, so six ports tree topology is used as the DC grid topology, the topology can be simplified as an equivalent circuit, DC grid mathematical model is established with the bond graph method, the model is simulated in Matlab/Simulink, and the effectiveness of bond graph method modeling is verified.2. For DC voltage stabilization control problem, a state feedback controller is designed to make pole assignment feasible; since the adjusting amplitude is large, it makes the power loss larger, consider these problems, the linear optimal quadratic index J=1/2(?) [xTQx+uTRu] dt is made up by control variable u and state variable x, a linear quadratic optimal state regulator is designed to keep linear optimal quadratic index optimal, these two control methods are simulated in Matlab/Simulink, and the effectiveness is verified.3. For the network power loss optimizing problem, the comprehensive index is made up of network power loss and the mean value of DC voltage steady-state error, and its fitness function is established with Particle Swarm Optimization (PSO) algorithm, which is used to search for the optimal value. As PSO algorithm is easy to fall into local optimum, an improved PSO algorithm based on fuzzy control theory is designed. These two PSO algorithms are simulated in Matlab, the effectiveness of these two PSO algorithms is verified. |
