基于TS-SVR的飞行员服装耗能测试与评价方法研究  

Experimental Study and Prediction Method of Energy Consumption of Pilot Clothing Based on SVR

作　　者：隋修武[1] 刘启俊 戚晓玲 杨国甫 Sui Xiuwu;Liu Qijun;Qi Xiaoling;Yang Guofu(Tianjin Key Laboratory of Modern Electromechanical Equipment Technology,Tiangong University,Tianjin 300387,China;Aviation Key Laboratory of Science and Technology on Life-Support Technology,Xiangyang 441002,China)

机构地区：[1]天津工业大学天津市现代机电装备技术重点试验室,天津300387 [2]航空防护救生技术航空科技重点试验室,湖北襄阳441002

出　　处：《航空科学技术》2021年第8期73-78,共6页Aeronautical Science & Technology

基　　金：航空科学基金(201729Q2001)。

摘　　要：不同特性的服装给飞行员带来不同的体能消耗,进而会影响飞行员的工作状态和工作效率,为了解决飞行服装耗能的测量和评价问题,本文提出了基于禁忌搜索算法(tabu search,TS)优化支持向量机回归(support vector machine regression,SVR)的服装耗能测量与评价方法。首先以模特机器人为核心,再现飞行员在行走与驾机工作模式,搭建了飞行服装耗能测试平台,为减少试验次数,设计正交试验,探讨了不同服装特性参数对耗能的影响,提出了服装耗能率的评价指标,建立了基于禁忌搜索算法改进SVR算法的服装耗能模型。试验结果表明,该模型对服装耗能的预测准确度达95.5%以上。In order to solve the problem of energy consumption caused by different characteristics of pilot clothing, an identification method of human body energy consumption suitable for different types of pilot clothing was proposed.Firstly, the model robot was designed to reproduce the motion state by simulating the human body in the working mode, and the protective energy consumption test platform was built. In order to explore the influence of different characteristics of clothing on the energy consumption of equipment system, orthogonal experiments were carried out on the model robot experimental platform, and a prediction model of clothing energy consumption based on Support Vector machine Regression(SVR) was proposed. The experimental results show that the model based on SVR can well predict the relationship between different characteristics of clothing and physical consumption, and can accurately evaluate the clothing grade of different characteristics of clothing.

关 键 词：服装耗能 支持向量机回归 测试平台 等级评估 禁忌算法 

分 类 号：V19[航空宇航科学与技术—人机与环境工程]