The Analysis And Study Of Harmonic Distortion In Distribution System Based On The Improving Genetic Algorithm

As early as the end of 19th century, people had discovered the distortion problem of voltage and current existing in the distribution system when the alternating current arisen as a newly motive form. With the gradually development of the modern industry, the problem of harmonic distortion in the electric power system has been more and more serious, especially with the increasing of the nonlinear loads' numbers. They even threaten the normal run of power system and other equipments. The harmonic analysis has already been an important research task in the field of electric power system. At present, this field includes some contents as following: harmonic tide flow calculation in power system, the research of the measure method and equipment of non-sinusoidal current and voltage, the equipment's design and research of reactive power compensation.One of the traditional methods of harmonic suppression is to set the reactive power compensative capacitors and LC filters. But in fact, the condition of harmonic sources and load levels and compensative capacitors in the distribution system are variable. The traditional methods cannot satisfy the request. The value of harmonic voltage and the correct site can not been found with the methods. We must find out a more well way to search the nodes that should be set filters to improve the distortion of the system's harmonic under the variable condition. Then, we can calculate the capability and design the configuration of filters depending on the conclusion. This paper presents briefly the base concept of harmonic, the harm of harmonic, and the significance of research the harmonic, the calculation methods of harmonic tide flow. Mean time the paper presents that we can resolve the problems using a genetic arithmetic.The genetic algorithm is a global optimizing robust technique, based on the natural selection and genetic production mechanism. One of the characteristics of the genetic algorithm is implicit parallel processing so that the speed of calculation is faster than other optimal algorithms. For the fact that all kinds of variable values will induce the change of net parameters, the genetic algorithm is a good way; moreover the GA begins the optimization process only according to the object function value, not according to themathematic model or the differential coefficient of the mathematic model of the object to be optimized. So it creates the condition for the application of harmonic analysis in the distribution system. And it's widely implemented in many kinds of optimization problems.However, although the GA has some advantages, the other hand it also has some disadvantages: low computation efficiency, limited precision and weak local search ability. It's very easy to get in the phenomena of "premature ". The paper presents an improving algorithm that named improving self-adapting GA aimed at the above shortcomings. Two operational genes that are the fixed crossover probability and the mutation probability in the simple GA have been modified. The new crossover operation and mutation operation all use the self-adapting operations. So the improving algorithm can enhance the efficiency of the simple GA. An example of numerical simulation is done using the improving GA in the paper.At the end of the paper, an example of distribution system is done and the harmonic mathematic model of the system is built. Then an optimistic computation is run using the improving self-adapting genetic algorithm. As a result, it can search out the note of the worst-case of harmonic distortion in distribution system. The conclusion provides a guidance way for the analysis of harmonic and placement of filters. The instance shows that the GA is an efficiency method to analyze the harmonic distortion in distribution system.