The pH process is challenging to control using conventional techniques because of its nonlinear and time varying process characteristics. This necessitates the design of model based control strategies for non-linear pH process. In this paper, Deep Neural Network is used to develop the forward and inverse models of pH process using its input-output data. The developed forward and inverse models are plugged into the Internal Model Control structure. To get the optimum performance, the number of hidden neurons in the hidden layers of DNN is determined using Grey Wolf Optimization. The effectiveness of the Grey Wolf Optimization based Deep Neural Network Internal Model Controller is contrasted with those of the Deep Neural Network Internal Model Controller and the traditional PI Controller.