A new improved genetic algorithm (IGA) based on floating point encoding is proposed. Firstly, IGA uses information entropy to produce better initialized species population. Secondly, after synthetically studying the searching properties of crossover operator in GA, it designs a new crossover strategy that effectively increases searching efficiencies of IGA. Thirdly, to avoid searching being trapped in local minimum, it designs a chaos degenerate mutation operator that makes the searching fast converge to a global minimum. At last IGA is used to solve the problem of the optimal design to crane girder, which is a typical problem of mechanical optimal design. Compared with the traditional random direction method, neural network method, genetic-neural network method, hybrid genetic algorithm, chaos-GA, PSO algorithm, chaos-PSO algorithm and standard GA, IGA shows better performance at the aspect of solution precision and convergence speed than that of these algorithms.
Through analysis of water cleaning trajectory in a railway tanker, a mathematical model of cleaning process is set up. Then the actual cleaning trajectories are obtained, and cleaning rate of high pressure water cleaning to the railway tanker is analyzed. Simulations and experiments show that much water is saved and it also has a great significance to improve the cleaning efficiencies of cleaning device.
At present, the detection accuracy of monitoring method by human ears is not satisfying in the pure tone detection of on-vehicle loudspeakers. To solve this problem, a new method is proposed to convert vehicle loudspeakers response signals into a two-dimensional image signal via wavelet packet analysis, which can increase the time-frequency of malfunction information. Through image binaryzation and pretreatment image-edge detection, the resulting signal would be recognized with box-counting dimension acquired in the process of gaining time-frequency image through fractal dimension as malfunctioned indications. Experiments show a rate of fault recognition as high as 95% , which meet the requirements of online vehicle loudspeaker detection.
Automotive abnormal sound is an important manifestation of the automotive system components early malfunction. It can effectively improve safety and reliability of the vehicle system by characteristics extraction of abnormal vehicle sound and prediction of the development trend of fault. What’s more, it can solve the problem that ECU can’t identify tire failure and other problems. As a result, it will reduce the number of traffic accidents, protect human safety and property.
scite is a Brooklyn-based organization that helps researchers better discover and understand research articles through Smart Citations–citations that display the context of the citation and describe whether the article provides supporting or contrasting evidence. scite is used by students and researchers from around the world and is funded in part by the National Science Foundation and the National Institute on Drug Abuse of the National Institutes of Health.