Traffic simulation software parameter calibration method and system based on learning algorithm and medium

Abstract

The invention provides a traffic simulation software parameter calibration method and system based on a learning algorithm, and a medium, and the method comprises the steps: A, carrying out the preliminary screening of parameters, carrying out the sensitivity analysis of the screened parameters, obtaining a first to-be-calibrated parameter meeting a preset requirement, and collecting a training data set; b, according to the obtained first to-be-calibrated parameters and the training data set, carrying out neural network model training, and obtaining a trained neural network model; and C, performing particle swarm optimization according to the obtained trained neural network model. The method provided by the invention can efficiently find the optimal parameter suitable for the current roadnetwork, and is suitable for practical application.

Description

technical field [0001] The invention relates to the field of traffic simulation, in particular to a method, system and medium for calibrating traffic simulation software parameters based on a learning algorithm. In particular, it relates to a method for calibrating parameters of micro-traffic simulation software based on machine learning algorithms. Background technique [0002] Micro-traffic simulation technology is the main technology for traffic engineers to evaluate traffic impacts. The effectiveness of the traffic planning scheme is verified by analyzing the output results of the simulation technology. To ensure the accuracy of the results, the micro-simulation software needs to conform to the actual traffic conditions as much as possible. However, factors such as driving behavior and road alignment in different locations are different. A software parameter calibration method is needed to quickly and accurately adjust relevant software parameters to provide support for ...

AI Technical Summary

Problems solved by technology

[0005] The existing micro-traffic simulation software parameter calibration method is not efficient and the algorithm rules are complicated, so it cannot be well applied in actual traffic evaluation projects

In addition, the existing parameter calibration methods are often used in combination with micro-traffic simulation software, which is too time-consuming and inefficient. A neural network training method is proposed to replace micro-traffic simulation software and particle swarm algorithm. Parameter Calibration Method of Network and Particle Swarm Algorithm

Images