项目名称: 基于神经网络PID控制在人体头颈部有限元模型中实现肌肉主动控制的理论和方法
项目编号: No.11202077
项目类型: 青年科学基金项目
立项/批准年度: 2013
项目学科: 数理科学和化学
项目作者: 黄晶
作者单位: 湖南大学
项目金额: 26万元
中文摘要: 肌肉主动力在提高有限元人体模型生物逼真度方面的重要作用使其成为新的研究热点。神经系统对肌肉主动控制模型的建立及其在有限元研究平台与肌肉力学模型的有机耦合是关键。本项目从人体骨骼肌系统复杂的动力学特性以及对肌肉主动力影响因素的挖掘出发,工作重点围绕解决人体有限元模型肌肉主动控制模拟的关键科学问题,将中枢神经系统进行运动控制过程中感觉系统的动态特性用传递函数模拟,开发得到自主创新的基于神经网络PID闭环控制的肌肉主动控制模型,通过控制技术与力学模型的有机耦合,进一步提高人体有限元模型的生物逼真度。并以实现主动力模拟的模型为工具,研究汽车碰撞过程中人体头颈部损伤机理,明确肌肉的主动反应对人体响应的影响,为汽车安全设计,特别是主、被动相结合的安全系统奠定生物力学基础和提供有效的研究方法与工具。该研究将促进肌肉主动控制在人体有限元模型中的应用基础研究,并提供相关的科学依据和方法。
中文关键词: 肌肉主动力;神经网络;PID控制;LS-DYNA二次开发;有限元模型
英文摘要: Active muscle modeling in finite element human body model has attracted considerable attention during recent years in the field of academic research and the new display technology. Muscles active control model and its coupling with muscle mechanical model on finite element analysis platform are the key issues. This project begins with the dynamics research of human skeletal muscle system as well as influence factors of active muscle force, and the dynamics of sensory systems available for the CNS to use in motor control process are modeled with a transfer function, then develop an innovation active muscle control model based on neural network PID closed-loop controller. The coupling of control technology and muscle mechanical model further improves the bio-fidelity of finite element human body model. Finally The new developed finite element human body model with active muscle response used to research the human head-neck injury mechanism in vehicle impact, determine the influence of muscle active response on human kinematic response, the research results will provide biological mechanical basis and effective research tools for vehicle safety design, especially the integrated safety system. The study will promote research and application of muscle active control in finite element human body model, and provide r
英文关键词: active muscle force;neural network;PID control;ls-dyna customization;finite element model