Determination of transient flow pressure losses due to leakage from pipe wall using intelligent algorithms

Abstract

Intelligent algorithms, have greatly enhanced the ability of engineers to analyze and model complex hydraulic phenomena. Among them is the analysis of transient flow, which are always an important part of hydraulic pressurized pipelines .over time, the pipelines will , leak, and so on. one of the most important characteristics of transient flow is the rate of losses in pressure waves, which will be exacerbated by leakage from the pipe wall. In this research, using intelligent algorithms such as artificial neural network (ANN), genetic algorithm (GA) and gene expression programming (GP), the rate of pressure losses passes the leak hole in the pipe wall (HLPW) to be determined and its application to be compared with hydraulic analysis. Therefore, first, with the help of dimensional analysis, the effective parameters on (HLPW) were determined and then a total of 120 experiments with 6 discharges, 5 leak diameters and 4 leak locations on a polyethylene pipe with a nominal diameter of 63 mm and a length of 47 m was considered for research. The results showed that the ANN model has the best performance among intelligent algorithms to estimate and calculate (HLPW). Also, ANN, GEP, ITA and GA models with R2 equal to 0.987, 0.905, 0.891 and 0.721, respectively, have the best performance in estimation of (HLPW). In general, for estimation of (HLPW) some intelligent algorithms are more powerful than the existing hydraulic analyzer. Therefore, their use is recommended in both terms of reducing time and increasing the accuracy of calculations.

Keywords

• artificial neural network (ann)
• genetic algorithm (ga)
• gene expression programming (gep)
• pressure wave losses
• leakage