Tool wear monitoring is one of the critical issues in the automated industry. Though use of artificial neural networks for tool wear monitoring is widely reported in the literature, the models are built only for dry machining. In the present work, a neural network model for cutting fluid assisted machining is proposed. Experimentation has been carried out using different cutting fluids and the results were used to build up and test the model. Further, an improvement in the network is proposed using simulated annealing, which can automatically and effectively optimize the network architecture, as opposed to the conventional trial and error method.