基于有限测点信息的大型空间网架结构智能工作状态评估技术研究

A study on estimation of the intelligent work status of large space truss structures based on limited measurement information

针对大型空间网架结构的特点,建立基于有限测点信息的智能工作状态评估技术。该技术是首先通过测量少数网架结构杆件的响应信息,获得结构响应的峰因子;再通过安置在网架结构上的风压力传感器获得结构的荷载信息,采用BP神经网络法和加权荷载本征模态分解法分别精确识别作用在网架结构上的平均风荷载和脉动风荷载;并通过安置在网架结构上的加速度传感器获得结构在环境激励下的加速度响应信息,对结构进行有限元模型修正;然后采用正向演绎的方法,得到整个网架结构的工作状态,根据预先设定的安全级别就能对其进行工作状态的评估。最后,通过该技术成功地应用于深圳市民中心屋顶网架结构的智能健康监测系统中,说明该技术的可行性和有效性。

The technique of estimation for the intelligent working status of large space truss structures is established based on limited measurement information. The intelligent technique are as follows： first, the structural response peak factor is obtained by measuring the response of a few elements; then, the structural load information is obtained by installed wind pressure sensors, and the BP neural network method is adopted to identify the mean wind pressures and the weighted proper orthogonal decomposition method is applied to detect the characteristics of impulsive wind pressures at all internal nodes on the truss structure; next, the structural finite element model is updated by using information from finite acceleration sensors installed on the structure; finally, the entire truss structural working performance is obtained by a positive deductive method, and the structural working status is estimated according to the predetermined safety level. The feasibility and effectiveness of the proposed technique is verified by an application to the intelligent health monitoring system of the Shenzhen Citizen Center roof truss structure.