Optimal Power Flow Of Distribution Network With Distributed Generation Based On Modified Firefly Algorithm

With the exhaustion of fossil energy and the deterioration of environment, distributed generation has been paid more attention in recent years. Distributed generation has many advantages, such as reducing pollution of the environment, improving the reliability and quality of power supply, decreasing the loss of power system, etc. However, when distributed generation units are introduced to distribution network, the power flow and the direction of sub-circuit will be changed, which makes the wastage of system not only relates to the load, but also correlates to the location and constant volume of distributed generation. Therefore, studying the reasonable planning of the distributed generation in distribution network is of great significance. Moreover, efficient intelligent optimization algorithm is an important means of solving the problem that how to make distributed generation in distribution network optimize configuration model. The main research contents of this paper are as follows:Firstly, based on the introduction of the basic principle of Firefly Algorithm as well as its shortcomings of easy premature, slow speed and low precision of convergence and over-reliance on control parameters, this thesis blends the idea of holistic optimal and chaos search strategies into the Firefly Algorithm, proposing a newly improved Firefly Algorithm. Through testing several typical kinds of the optimization function, this article verifies the feasibility and effectiveness of this improved algorithm proposed by the author. Besides, the author holds that compared with Particle Swarm Algorithm and conventional Firefly Algorithm, new algorithm has better optimization speed and optimization precision.Secondly, based on the power flow calculation of distribution network, this thesis makes a short term plan of the location and constant volume of distributed generation in distribution network. Based on the power flow of distribution network with distributed generation, a multi-objective programming model has been established, which costs the minimum investment and has the least network loss, and an improved firefly optimization algorithm to the multi-objective programming model has been applied to analyze and calculate the certain capacity, then the grid data of network loss and voltage fluctuation can be obtained. Utilizing of the simulation analysis of IEEE33 node distribution network system, this thesis verifies the advantages of the newly improved algorithm in solving the traditional typical complex optimization problems as well as compares with the optimization results of single objective and multi-objective.Lastly, Changes of load will influence the voltage of the power system and the loss of network, which results in the variations of load influencing the planning and control of the distributed generation supply. Moreover, optimization of distributed generation supply will be changed with load of the continuous. Based on the analysis the third chapter, it added a new load variations model. This paper sets the minimum loss of active network as the optimal allocation model for a long-term optimal allocation of distributed generation considering load variations and DG type. Through the simulation analysis of IEEE33 node in distribution network system, this thesis verifies the rationality and practicability of the model and compares with the analyzed results of the third chapter.In addition, this paper establishes the model of long term scheduling for optimal allocation and sizing of DG unit considering load variations and DG type, with the minimum of active power loss as the objective function. Finally, this paper combined with an example to analysis and verification.