Parameter Calibration Method Of VISSIM Model For Road Intersection Based On Self-adaptive Orthogonal Genetic Algorithm

As a powerful way to solve traffic problems, the traffic simulation technology is widely used in the Field of Traffic Engineering. However, most of the simulation softwares are based on the traffic conditions of those countries that developed the softwares, and may not be suitable for our country. So calibrating the model parameters is required to express the object of simulation more accurately. In response to this phenomenon, this article intends to study the implement of self-adaptive orthogonal genetic algorithm in the parameter calibration of VISSIM simulation model for road intersection.First of all, this article introduces the development tendency of parameter calibration for microscopic traffic simulation models, reviews and summaries the selection of parameter calibration method for microscopic traffic simulation models at home and abroad, and points out the realistic significance of the parameter calibration for microscopic traffic simulation model in domestic.Secondly, with the background of VISSIM simulation model for road intersection, this article introduces the simulation model parameters from the three aspects respectively:the selection of evaluation index; the selection of calibration parameter; the identification of sensitive parameter. And determines the final calibration parameter set using the sensitivity analysis, which lays a foundation for further research.Thirdly, aiming at the shortcomings of the traditional genetic algorithm, this article introduces the self-adaptive orthogonal genetic algorithm. The main feature of this method is applying the orthogonal experimental design to improve the method of generating the initial population, and make the initial population distribute uniformly. At the same time, applying the adaptive algorithm to make the crossover have self-adaptive characters.Fourthly, designs a VISSIM model parameter calibration method for road intersection based on self-adaptive orthogonal genetic algorithm, and describes the encoding and decoding of chromosome, the design of fitness function and the design of self-adaptive orthogonal crossover in this optimization method. The parameter calibration program of VISSIM model for road intersection based on the self-adaptive orthogonal genetic algorithm is completed with the aid of COM interface and VB programming language.Finally, the parameter calibration program based on the self-adaptive orthogonal genetic algorithm is applied into the intersection model of the Ronghuazhong road and the Rongjing west street in Beijing city. The automatic calibration of model parameters can be realized. The model can acquire the simulation parameters and results which can reflect the current state of the intersection better. Through the comparison and analysis of the experiment, it can reflect the advantages of adaptive orthogonal genetic algorithm.