Development of methods for studying and assessing the reliability of roads at the stages of their design

The article outlines the problems of research and evaluation of the reliability of highways at the stages of their design. As a result of the study, a three-stage method of obtaining generalized statistically homogeneous information on the same type of roads was developed based on sufficiently rigorous methods and models of pattern recognition theory, theories of similarity and dimensions, and a new predictive method for analyzing the condition of roads for their further use of the information obtained for the study and evaluation of the probability of failure-free operation of new road structures. The technique is based on taking into account the maximum use of experience in the design of highways-analogues of the world’s leading countries with high levels of reliability. The use of the probabilistic-statistical approach proposed in this work to assessing the reliability of roads, taking into account the risk of their road construction’s destruction over time during road maintenance, can be used as a priori and predictive calculation information in a quantitative study of the reliability of new road modifications and their elements on stages of road data design. The results of the study carried out in this paper allow us to conclude that the provisions of the developed methodology make a significant contribution to the process of studying and evaluating the reliability of roads, which during the design stages can contribute to a significant adjustment.

The Damage Detection of Pressure Pipe Based on the Statistical Pattern Recognition Theory

Based on the statistical pattern recognition theory, the AMRA timing analysis methods are used in the article, through the combination of long autoregressive model residuals method and the least squares method the model parameters are estimated, and a system model is established. By using mean control chart method the vibration information and feature of the pressure pipe are extracted and selected, so whether the pressure pipes is damaged can be judged effectively. The simulation results show that structural abnormalities test method of the mean value ,which is Based on the recognition theory of statistical pattern, can accurately diagnose structural damage detection state ,the injury degree and damage location, it has a very strong sensitivity