METHOD FOR GENERATING A NEURAL NETWORK FOR A FIELD DEVICE FOR PREDICTING FIELD DEVICE FAULTS AND A CORRESPONDING SYSTEM

摘要

A process for generating a neural network for a field device (2) for predicting field device errors is presented and described, with the field device (2) having a field device computer (7) for executing neural networks up to a maximum complexity.;The purpose of the invention is to specify a method for generating a neural network for predicting field device errors, which is executable from the field device computer (7) of a field device.;The task is solved by a procedure where a neuron-displaying neural network for predicting field device errors with an initial complexity is generated greater than the maximum complexity, generating training data from data on field device errors, training the neural network with the training data, so that it is trained to predict field device errors, reducing the initial complexity at least to the maximum complexity by first removing at least one neuron from the neural network in at least one reduction step and then re-training the reduced neural network,so that it is trained to predict field device errors.