Research Of Universal Characteristics Curve Modeling Based On Modern Optimization Algorithm

With the rapid development of automobile industry and the improvement of people’s life quality, the number of vehicles in China has been increased fast in recent years. Meanwhile, the energy and environmental issues are increasingly severe because of the large quantity of vehicles. Most of the vehicles which people ordinarily prefer enough power, fuel economy and environmental protection. To achieve the goal, designers need to analyze engine performance in advance during the design and manufacturing stages of a vehicle. They must fetch the parameters of power and economy, such as torque, power and fuel consumption, which should cover all the speed range of the engine. To solve the contradiction between economy and power, these parameters should be analyzed then the appropriate powertrain system could be determined. Because the universal characteristics of the engine can fully evaluate its performance, it’s necessary to build a universal characteristics model and draw the universal characteristics curve during the development stage of vehicle.After reading a large number of documents, the traditional technique of drawing universal characteristics curve is considered strenuous and inefficient. Thus, a new method based on traditional cubic surface regression method and improved ant colony algorithm was proposed. The dimensionality reduction and discretization technique were used to improve the calculating speed. Also, a corresponding simulation platform was developed, with which some experiments had been made to determine the parameters of the algorithm, so that the efficiency and accuracy of the algorithm have been improved greatly. Finally, the feasibility of the algorithm in universal characteristics curve modeling was verified.The main work of this paper is as follows.(1) The traditional technique of the universal characteristics modeling and a special technique based on BP neural network and surface interpolation were analyzed.Ant colony algorithm and genetic algorithm were introduced, then the improved ant colony algorithm was designed.(2) A new method based on improved ant colony algorithm was proposed, and a universal characteristics curve had been created.(3) A simulation platform for algorithm verification based on VB was developed. Then the feasibility of the algorithm was verified and the relevant parameters were determined. All in all, this algorithm proposed could achieve the goal of universal characteristics curve modeling.