人工神经网络在仪表信息处理中的应用

Application of artificial neural network in information processing of measuring instrument

介绍了一种用于测量二次仪表中充满管道的过热蒸气流量和载热计算的人工神经网络。它节省了储存空间和运算时间,具有本征性滤波功能,在高准确度,防止发散、偏离和非唯一性方面较之传统算法具有很大优越性。同时,还进一步介绍了由此而产生的具有学习能力的仪表。

In the paper, a kind of artificial neural network models applied in measuring instrument for calculating measurand is suggested. The kind of models save memory space, reduce counting time and have advantages to guarantee against non uniqueness, deviation and divergence, specially have the essential capability of restraining random disturbance. Furthermore, a kind of ANN based self learnable instrument are suggested.