Centrifugal pump performance curve sample obtaining method and application thereof in machine learning

Abstract

The invention discloses a centrifugal pump performance curve sample obtaining method and application of the centrifugal pump performance curve sample obtaining method in machine learning. The method comprises the steps that firstly, data of one performance parameter of multiple centrifugal pumps changing along with flow are collected; performing mathematical modeling on the other two unknown performance parameters to obtain a simplified formula of the unknown performance parameters; then, based on the relation between the performance parameters and a simplified formula of unknown performance parameters, calculation values of known performance parameters are obtained through calculation; and finally, taking a mean square error of a true value and a calculated value of the known performance parameter as an objective function of a differential evolution optimization algorithm, and optimizing to obtain coefficients of two unknown performance parameter calculation formulas. Finally, unknown performance parameter values under different flows are obtained through calculation. According to the method, a large number of performance parameter samples can be quickly and conveniently obtained. The method is further combined with a machine learning model, performance prediction results of different centrifugal pumps can be obtained based on geometric parameters of the centrifugal pumps, calculation time is short, implementation is convenient, and therefore the research and development progress is accelerated.

Description

technical field [0001] The invention relates to the field of centrifugal pump performance curve prediction, in particular to a centrifugal pump performance curve sample acquisition method and its application in machine learning. Background technique [0002] A centrifugal pump is a fluid machine that converts the mechanical energy of a prime mover into liquid energy, and is widely used in petrochemical, nuclear power, irrigation, urban water supply and heating systems. However, the internal flow of a centrifugal pump is very complicated. The development of a pump product needs to go through tedious processes such as design, trial production, testing, and improvement, which will consume a lot of time, manpower, and material resources. Therefore, the performance prediction research of centrifugal pump becomes the key to product development. The impeller and volute are important flow-passing parts of the centrifugal pump. If the performance curve of the centrifugal pump can be...

AI Technical Summary

Problems solved by technology

[0005] In view of the lack of samples of the performance prediction model of the existing centrifugal pump and the large amount of calculation of the performance prediction model, the present invention proposes a method for obtaining the performance curve sample of the centrifugal pump, which can be calculated based on a single known performance parameter Obtaining the other two performance parameters can greatly reduce the difficulty of obtaining a complete centrifugal pump performance curve training sample. On this basis, this method is combined with a machine learning model to propose a performance prediction method for centrifugal pumps. Based on the geometric parameters of the centrifugal pump, the corresponding performance parameters can be quickly obtained, the calculation time is short, and the implementation is convenient

Images