We account for the problem of optimal control of ceramic mould manufacturing in lost wax cast processes with the aid of a mixed linear algebraic-statistical approach based on the employment of Singular Value Decomposition (SVD) and Neural Networks (NN).We consider the peculiar aspect of minimizing ceramic inclusions occurrence in equiaxed superalloy turbine components which are manufactured resorting to gravitational pouring. The optimization consists in finding optimal extrema of scalar and/or vectorial functions of the type Rk->Rmi.e. Key Process Variable domain (KPV) vs. Target Variable domain (TV) over a large set of experimental data affected by acquisition noise leading to a typical sparse multiblock array.The goal of the work consists in the assessment of possible significant statistical multivariate correlations amongst the KPV and TV when the dimension of domain space, k, has an order of magnitude of tens, in the presence of quasi-rank deficient input matrix.
This study analyzes the thermal behaviour of a ceramic shell mold during the a stage of the investment casting process. The investment casting process is a multi-step procedure which allows producing very reliable metal components such as turbine blades. Many parameters influence the final product of an investment casting process such as the initial mold temperature. This initial thermal status strongly affects the radiative heat transfer during the casting and, consequently, the product quality. This work describes and discusses the 3D temperature distribution over the ceramic shell mold of a turbine blade.
he reduction of the scraps is fundamental to achieve goals of competitiveness. Some key parameters have a direct influence on any process and they need to be predicted and taken under control. This paper present an approach ) is to develop a robust monitoring solution of the ceramic shell manufacture that will be able to determine a significant reduction of the inclusion scraps (due the ceramic shell) of the superalloy components. The control will be obtained by processing data coming both from sensors and laboratory measured values. The sensor data come from the new equipment of the Europea Microfusioni Aerospaziali SpA (EMA) and have been tested and used to develop the EMA demonstrator within the EC FP7 Project on "Intelligent Fault Correction and self-Optimizing Manufacturing systems-IFaCOM". The sensor data will merge the data measured in the EMA laboratories and both the values will concur to create the sensor fusion pattern vector, which will be used to feed an automatic system for the prediction of the process parameters. The automatic system will be implemented using cognitive paradigms, in particular Artificial Neural Networks, that will combine both data. The first testing phase will predict the number of blades with inclusions. It will provide a first idea of the correlation between the input, as a matrix composed by the sensor fusion pattern vectors per each worked blade, and the outputs, as a vector of rejected blades on the total. Moreover, this work will be the basis to implement a predictive system to estimate which is the reference range of each working parameter
The competitiveness of a casting system in modern lost wax production of superalloy turbine blades strongly depends on the reduction of scraps, which commonly affect superalloy cast parts. In order to achieve a focused goal of competitiveness, some key and vital parameters (Key Process Variables) have to be continuously taken under control to make very accurate predictions of Target Variables, which represent, as mapped KPVs domain, the ultimate performance of the entire production link. Such an approach is based on the development of robust control monitoring of the ceramic shell manufacture, which is specifically conceived to foster a possible reduction of scraps in the production if superalloy components. The concerned control will take into consideration data coming from both sensors and measured values in laboratory. The sensor data, which is originated from both new adopted inline and offline equipments at Europea Microfusioni Aerospaziali S.p.A. (EMA) and data measured in the EMA laboratories, will be merged into a sensor pattern vector which represents the basis to develop the EMA demonstrator within the Intelligent Fault Correction and self Optimizing manufacturing systems EU project funded in FP7. The sensor pattern vector will be used to feed an automatic system for the prediction of the process vital parameters. An automated system, based on artificial intelligence paradigms, in particular neural networks, will be fed with the data coming from the sensor pattern vector in order to produce an optimal multi-object output
scite is a Brooklyn-based organization that helps researchers better discover and understand research articles through Smart Citations–citations that display the context of the citation and describe whether the article provides supporting or contrasting evidence. scite is used by students and researchers from around the world and is funded in part by the National Science Foundation and the National Institute on Drug Abuse of the National Institutes of Health.
hi@scite.ai
10624 S. Eastern Ave., Ste. A-614
Henderson, NV 89052, USA
Copyright © 2024 scite LLC. All rights reserved.
Made with 💙 for researchers
Part of the Research Solutions Family.