Response Surface Methodology for Modelling and Optimizing Efficiency in Deep Well Pumping Systems

This study presents research on modelling the efficiency and flow rate of deep well pumping facilities using the response surface method, evaluating the models, and assessing optimization based on target flow rate. Regression and variance analysis techniques have been successfully employed to evaluate the relationship between input factors (input pressure and power drawn from the grid) and responses (system efficiency and flow rate). ANOVA analysis has been used to examine the effects of linear and quadratic terms, and the results have shown that pressure and power drawn from the grid have a significant effect on pump system efficiency. Additionally, the performance of the regression models has been evaluated using error metrics such as R2 value, RMSE, and MAPE. These values for the pumping facility system efficiency model were found to be 0.9993, 0.292%, and 0.71%, respectively, and for the flow rate model, they were 0.9997, 0.69 m3h-1, and 1.07%. The results obtained demonstrate that the model operates with high accuracy and explains a large part of the variance in the response variables. An optimization study was conducted to maximize pump system efficiency by maintaining the flow rate at a certain target value. According to the experimental results obtained, the target flow rate was predicted with an error rate of 1.49%, and the pump system efficiency was predicted with an error rate of 2.14%. This study highlights the effective use of various analytical and experimental methods to improve the efficiency of pump systems. Future researchers are encouraged to conduct similar analyses on a larger scale and under different operating conditions. Furthermore, evaluating different optimization strategies to improve the energy efficiency of pump systems, which can lead to significant energy savings in industrial applications, is recommended.