基于人工智能的工业用温度检测仪器自动化控制方法研究

Research on automatic control method of industrial temperature measuring instrument based on Artificial Intelligence

为了解决传统方法参数分析过程复杂,学习性能差,导致工业用温度检测仪器控制效果差,精测精度低的问题,提出一种新的基于人工智能的工业用温度检测仪器自动化控制方法。分析了工业用温度检测仪器工作原理,人工智能方法选择计算过程简单、学习性能强的单隐含层前馈神经网络,通过最小二乘法对隐含层节点数据进行拟合处理,获取隐含层节点数量,任意形成隐层节点参数,求出隐层输出矩阵与输出权值,完成学习。在实际应用中,把温度检测仪器输出信号当成前馈神经网络的输入,通过大规模工业用温度检测仪样本对前馈神经网络进行训练,在前馈神经网络输出误差趋近于0的情况下,通过输出权值模糊调节器获取输出融合权值,实现误差控制结果的融合,得到准确的温度检测结果,提高温度检测仪检测精度。实验结果表明,所提方法输出结果和目标结果间的差异小,泛化误差低,自动化控制效果好。

In order to solve the problems of complex parameter analysis process and poor learning performance of traditional methods,which lead to poor control effect and precision of industrial temperature measuring instruments,a new automatic control method of industrial temperature measuring instruments based on artificial intelligence is proposed.The working principle of industrial temperature detection instrument is analyzed,the artificial intelligence method selects the single hidden layer feedforward neural network with simple calculation process and strong learning performance.Through the least square method,the hidden layer node data is processed by fitting,the number of hidden layer nodes is obtained,the hidden layer output matrix and output weight are obtained by arbitrarily forming the hidden layer node parameters,and the learning is completed.In practical application,the output signal of temperature detector is regarded as the input of feed-forward neural network.The feed-forward neural network is trained by a large-scale industrial temperature detector sample.When the output error of feed-forward neural network approaches to zero,the output fusion weight is obtained by the output weight fuzzy regulator to achieve the fusion of error control results and accurate temperature detection The measurement results can improve the detection accuracy of the temperature detector.The experimental results show that the difference between the output results and the target results is small,the generalization error is low,and the automatic control effect is good.