Quantitative prediction method for loosening defects of large complex thin-wall high-temperature alloy casting

Abstract

The invention provides a quantitative prediction method for loosening defects of large complex thin-wall high-temperature alloy casting, which comprises the following steps: analyzing the structure ofa large complex thin-wall casting, and calculating the thermal modulus of each part of the casting; calculating the crystallization temperature interval, density, specific heat capacity and shrinkagecoefficient parameters of the high-temperature alloy, importing the calculated parameters into ProCAST to carry out molding and solidification simulation on the casting, extracting the temperature gradient and cooling speed values, calculating the secondary dendritic crystal interval through a formula, and calculating the secondary dendritic crystal density of the high-temperature alloy; selecting the obtained thermal modulus, temperature gradient and secondary dendrite spacing as input parameters of the BP neural network and the porosity index as output parameters, constructing a BP neural network prediction model, training the BP neural network prediction model by adopting casting body anatomical data, and constructing a casting porosity quantitative prediction network after training, thereby realizing casting porosity defect quantitative prediction. The BP neural network is constructed to realize casting loose defect quantitative prediction, and loose defect sensitive process parameters can be obtained.

Description

technical field [0001] The invention relates to the field of formation and control of superalloy precision casting defects, in particular to a quantitative prediction method for loose defects of large complex thin-walled superalloy castings, electronic equipment and media. Background technique [0002] The aero-engine is the heart of the aircraft, and the design and manufacturing level of the aero-engine plays a decisive role in improving the overall performance of the aircraft. The third-generation commercial engine has very high reliability requirements, with less than 3 air shutdowns per million flight hours, and the life of hot-end components needs to reach 15,000 hours, which puts forward higher requirements for the overall reliability of large complex thin-walled superalloy components requirements, prompting the overall investment casting technology to become one of the mainstream technical routes for the manufacture of advanced aero-engine hot-end components. However...

AI Technical Summary

Problems solved by technology

[0011] Aiming at the defects in the prior art, the object of the present invention is to provide a method for quantitatively predicting loose defects in large complex thin-walled superalloy castings, electronic equipment and media

Images