Politecnico di Milano

Consultancy : Research : Academic

Website

Milan, Italy

Politecnico di Milano is a public scientific-technological university which trains engineers, architects and industrial designers. The University has always focused on the quality and innovation of its teaching and research, developing a fruitful relationship with business and productive world by means of experimental research and technological transfer. Research has always been linked to didactics and it is a priority commitment which has allowed Politecnico Milano to achieve high quality results at an international level as to join the university to the business world. Research constitutes a parallel path to that formed by cooperation and alliances with the industrial system.

Assembly Line

A new intelligent fault diagnosis framework for rotating machinery based on deep transfer reinforcement learning

πŸ“… Date:

✍️ Authors: Daoguang Yang, Hamid Reza Karimi, Marek Pawelczyk

πŸ”– Topics: Bearing, Reinforcement Learning, Machine Health, Convolutional Neural Network

🏒 Organizations: Politecnico di Milano, Silesian University of Technology


The advancement of artificial intelligence algorithms has gained growing interest in identifying the fault types in rotary machines, which is a high-efficiency but not a human-like module. Hence, in order to build a human-like fault identification module that could learn knowledge from the environment, in this paper, a deep reinforcement learning framework is proposed to provide an end-to-end training mode and a human-like learning process based on an improved Double Deep Q Network. In addition, to improve the convergence properties of the Deep Reinforcement Learning algorithm, the parameters of the former layers of the convolutional neural networks are transferred from a convolutional auto-encoder under an unsupervised learning process. The experiment results show that the proposed framework could efficiently extract the fault features from raw time-domain data and have higher accuracy than other deep learning models with balanced samples and better performance with imbalanced samples.

Read more at Control Engineering Practice

πŸ–¨οΈ Design of additively manufactured moulds for expanded polymers

πŸ“… Date:

✍️ Authors: Franco Alessio, Mattia Alessio, Pietro Savoldelli, Maurizio Vedani, Roberto Viganò

πŸ”– Topics: Additive Manufacturing

🏒 Organizations: AlessioHiTech, Politecnico di Milano


The traditional tools used in steam-chest moulding technologies for the shaping of expanded polymers can be replaced today by lighter moulds, accurately designed and produced exploiting the additive manufacturing technology. New paradigms have to be considered for mould design, assuming that additive manufacturing enables the definition of different architectures that are able to improve the performance of the moulding process. This work describes the strategies adopted for the design and manufacturing by Laser powder bed fusion of the moulds, taking into specific consideration their functional surfaces, which rule the heat transfer to the moulded material, hence the quality of the products and the overall performance of the steam-chest process. The description of a case study and the comparison between the performance of the traditional solution and the new moulds are also presented to demonstrate the effectiveness of the new approach. This study demonstrates that the re-design and optimization of the mould shape can lead to a significant reduction of the energy demand of the process, thanks to a homogeneous delivery of the heating steam throughout the part volume, which also results in a remarkable cutting of the cycle time.

Read more at The International Journal of Advanced Manufacturing Technology