基于BP神经网络的AGV功能配置情景匹配方法

AGV FUNCTION CONFIGURATION SCENARIO MATCHING METHOD BASED ON BP NEURAL NETWORK

  • 摘要: 针对AGV产品功能配置方案利用率低、应用情景适配度不佳和功能-情景关系映射不明确的问题,提出结合情景仿真与BP神经网络的AGV功能配置情景匹配方法。在实现配置功能编码后,通过情景仿真实验获得AGV功能配置方案在不同应用情景中的表现数据,并结合层次分析法和模糊综合评价法获得各功能配置方案在不同场景中的匹配评价值,基于所得数据构建和训练BP神经网络得到功能配置方案与应用情景的匹配方法模型。实验分析结果表明,该方法建立的匹配模型具有较优的表现且对于AGV产品功能配置设计具有一定的参考性。

     

    Abstract: To solve the problems of low utilization rate of AGV product function configuration scheme, poor suitability of application scenarios and unclear mapping of function-scenario relationship, a scenario matching method of AGV function configuration combining scenario simulation and BP neural network is proposed. According to the configuration function coding set, the performance data of AGV function configuration schemes in different application scenarios were obtained through scenario simulation experiment, and the matching evaluation values of each function configuration scheme in different scenarios were obtained by combining analytic hierarchy process (AHP) and fuzzy comprehensive evaluation method. The matching method model of function configuration scheme and application scenario was obtained by constructing and training BP neural network. The experimental results show that the matching model established by this method has better performance and has certain reference for the functional configuration design of AGV products.

     

/

返回文章
返回