Evaluation of the M5P Decision tree in Determination of the Maximum Negative Pressure Induced in the Sudden Expansion of the Pressurized Pipe Flow

Sudden expansion can be used as energy dissipater in the pressurized pipe flows. As the induced cavitation bubbles will cause large structural damage، estimating the amount of maximum developed under-atmospheric pressure (MDUAP) in sudden expansions is of particular interest. Pressure fluctuations، domain and intensity of cavitation depend on the amount of MDUAP. Therefore، estimating the amount of MDUAP is a vital task in the design stage. To carry out this study، various sudden expansions were developed by changing the geometry and flow properties. The developed sudden expansions were modeled using the CFD modeling technique. The obtained CFD based results were successful in analyzing the experimental observations. Therefore، the simulation model is capable of predicting the MDUAP under different conditions. In this paper، the M5P data mining model is used for the first time for establishing a fast، accurate، and cost-effective linkage between the geometry and flow properties of a sudden expansion and the amount of MDUAP. The results show that the developed M5P based simple rules can accurately predict the amount of MDUAP.