Examining the effectiveness of anti-vibration gloves with a neural network.

Whether anti-vibration gloves are effective in protecting against vibrations depends not only on the materials they are made of, but also on the parameters of the source of vibration. Depending on those parameters, the effectiveness of the same means of protection may be radically different. This article presents a methodology of using a neural network to test anti-vibration gloves. A network can map gloves in various conditions, i.e., for vibrations of various amplitudes and spectra, and for various forces exerted by the worker on a tool. Real, measured vibration signals produced by different tools were used in training a neural network. The results presented in this article prove that real properties of gloves are accurately represented by their models developed as a result of training a neural network.