Using a neural network in assessing the efficiency of personal protective equipment protecting against local vibration

The aim of the project was to develop a method of determining personal exposure to hand-arm vibration of workers after using protective equipment. Neural networks have been used to develop an algorithm for calculating personal exposure after protective equipment.Software was developed, which makes it possible to model parameters of equipment protecting against hand-arm vibration with neural networks. Preliminary tests involved using the newly developed software. Parameters of an artificial neural network that imitated protective equipment protecting against hand-arm vibration were changed in simulated calculations. The next stage of the project involved developing a methodology of transmitting vibration by protective equipment protecting against mechanical vibration from real hand tools. Three operators took part in the tests. During typical operations with selected hand tools that represented the source of hand-arm vibration of various kinds. The test results were used to teach neural networks, which were used to model the behaviour of the tested anti-vibration gloves. The process of teaching the neural network was preceded with the new software being used to prepare a set of files with sample test signals.The relative error between the squared sums of the samples in the buffer zone of the signal measured on the adapter inside the glove and produced at the output of the neural network that was the model of a glove was used as a measure of the quality of the process of teaching