传感器数据证实技术应用研究

传感器技术是现代测试和控制技术的灵魂 ,传感器输出信号的质量关系到整个系统性能的好坏程度。传感器数据证实技术是测控领域中的关键技术之一 ,文中提出了基于神经网络预测器的传感器数据证实技术。首先利用神经网络对传感器输出的时间序列建立预测模型 ,然后利用该模型预测出传感器输出值 ,并用该值与传感器实际输出值之差判断传感器实际输出的可靠性和有效性 ,进而减小不同环境噪声对传感器输出数据的影响。同时 ,为了实现在线应用 ,采用了遗传算法对神经网络的优化技术。实践表明 :文中提出的基于神经网络的传感器数据证实技术对缓变信号和快变信号都有很好的效果

Study and Applications of Sensor Validation Technique

Sensor technology is the soul of modern monitoring and control systems. And the system performance usually depends on the output signals of sensors. Sensor validation is the key technique of the monitoring and control fields. It proposes a new method to sensor validation based on predictor using neural network. The procedure of the method can be described as following:firstly,training a kind of neural network based on series predictor for a target sensor using a time data series of the sensor's output in certain working condition through G A. And then,the built predictor model is validated by a correlation analysis. Moreover, the discrepancy between the estimation of the predictor for the target sensor and its actual value in on-line condition may be used to determine whether sensor is working normally or not. At the end,it's proved that the method is effective and feasible by a simulation result and actual condition value for gear testing apparatus.