In this study, we developed an artificial neural network-based real-time predictive control and optimization model to compare and analyze the difference in total energy consumption when the condenser water outlet temperature coming out of the cooling tower is fixed and when real-time control of the condenser water outlet temperature through the optimal ANN model is applied. An ANN model was developed through MATLAB’s built-in neural network toolbox functionality to predict total energy consumption. The model accuracy of the ANN was examined by applying Cv(RMSE), a statistical concept that shows the overall accuracy of the predicted values, and as a result, it was found to have a Cv(RMSE) value of approximately 25%. In addition, the predictive control algorithm was able to reduce cooling energy consumption by about 5.6% compared to the conventional control strategy that fix condenser water temperature set-point to constantly 30°C.

This content is only available via PDF.
You do not currently have access to this content.