基于AI选型的汽车悬架系统参数化设计流程构造

Construct the Process of Parameterized Design on Automobile Suspension System Based on Artificial Intelligence Lectotype

建立了用于汽车悬架系统选型的多层前馈型人工神经网络模型,它采用自适应调整的BP算法。选型推理的依据有:轴距、前轮距、后轮距、整备质量、驱动形式、最高车速、制动形式、发动机最大功率和发动机排量。推理结果为前、后悬架的具体型式。经测试,其与实际情况的吻合率,对前悬架达到了100%,对后悬架达到了70%。进而开发出典型悬架型式的三维参数化设计模板,并集成相关技术环节,构造出基于AI选型的悬架系统参数化设计的一般流程。

Applying the adaptive adjustment BP algorithm,a multiple-layer-forward artificial neural network model used for the lectotype of automobile suspension system has been established. The reasoning of the lectotype based on wheelbase, front track, rear track, complete vehicle shipping mass, driving form, speed limit, braking form, maximum power and displacement of engine,with the reasoning result being the form of front and rear suspension system. Tested,the anastomosis between the result and the actual situation is 100% for front suspension system and 70% for rear suspension system. Further more,the 3 dimensional parametric design templates of some typical suspension systems were developed. Then, an integration of interrelated techniques was built to construct the general process of parameterized design of suspension system based on artificial intelligence lectotypo.