Research On Transmission Line Path Optimization Problem Based On Intelligent Optimization Algorithm

We consume electric power all the time in the daily life and production activities. The power system can be taken as the blood of social life. Due to the increase in demand, new transmission lines need to be constructed. The preliminary design of transmission line is so important because it can affect the construction, maintenance. So to design a reasonable, economic and safe line is a tough task. Traditional design methods are time-consuming and often can't generate a good line due to the skills of people. So this paper will use Google Earth and CAD to assist the design to improve the optimization results and increase efficiency.When routing transmission line, there are many factors need to be considered, terrain information, building density and social factors, technical conditions and so on. It is a complex problem of multi-objective programming. It is hard for the traditional mathematical methods to solve problems with multi-objective and multi-constraint. In recent years, with the rapid development of intelligent optimization algorithms, they have grate advantage in dealing with this kind of complex problems. Intelligent optimization algorithms can generate the exact best solution, but the approximate solution of the exact best solution, improving the efficiency of the computing at the cost of loss of accuracy.As there are too many constraints when routing the line, we intend to design a satisfactory line by simplifying the model and withdrawing some constraints of less influence. Taking the satisfactory line as a reference, the designers can modify the line later. The simplified model is to find the shortest line between two points on a 3D surface with obstacles. On the raster map, we use Google Earth to extract geographic information data, and then process the data in our system. Then the primary design of the route and the simulated terrain can be shown. Finally through the inverse transformation of the coordination, the line can be displayed on Google Earth.Two kind of algorithms, ant colony optimization(ACO) and the hybrid of genetic algorithm(GA) and ACO, are used to optimize the line. Both of them are approximate algorithms that are derived from biology, having a strong computational efficiency and better convergence effect. In this paper, ACO is used to optimize the line at first. In accordance with the practical needs, some improvement and innovation is made. Leaving the pheromone on the grid cells that ants have traversed, both the amount of calculation and computational space are reduced. In addition, taking the limit of slope into account, weight is added to avoid steep slopes when calculating the length of the line.In the raster map, it is hard to generate a feasible solution by using GA. And the crossover operation and mutation operation can cause the line to break. As a result, GA is not suitable to be used alone in this model. But GA has better global search ability; therefore, the hybrid of ACO and GA will be used in this paper. First, ACO is used to generate primary solution, and then crossover and mutation operations are carried out by GA. In the mutation operation, ACO is used to speed up the convergence of the hybrid algorithm.