Digital twin of functional gating system in 3D printed molds for sand casting using a neural network

Abstract

The filling stage is a critical phenomenon in sand casting for making reliable castings. Latest research has demonstrated that for most liquid engineering alloys, the critical meniscus velocity of the melt at the ingate is in the range of 0.4-0.6 m s −1 . The work described in this research paper is to use neural network (NN) technology to propose digital twin approach for gating system design that allow to understand and model its performances faster and more reliable than traditional methods. This approach was applied in the case of sand casting of liquid aluminum alloy (EN AC-44200). The approach is based first on a digital representation of filling process to perform the melt flow simulations using a combination of the gating system design parameters, selected as a training cases from Taguchi orthogonal array (OA). The second step of the approach is the data capture of functional gating design system to train up the feed-forward back-propagation NN model. The validation of the well-trained NN model is assessed by interrogating predicted ingate velocity to it and making reliable predictions with high accuracy. The claim is that such digital twin approach is an effective solution to recognize the functional design parameters from the entire filling systems used during casting process.

Impact of AI-Driven Digital Twins in Industry 4.0: an Exploratory Analysis

“…Additional instances can be discovered in industrial robot energy modeling, kinematics, communication, control, planning, and manufacturing use cases like welding, cleaning, pick-and-place, assembly, manufacturing, warehouse, maintenance, and construction. A few popular robotics simulation programs are CoppeliaSim (also known as V-REP) [23], MuJoCo [28], and Gazebo [29]. New ideas and examples of applying artificial intelligence to partially and completely autonomous robotic systems have recently been published.…”

“…A smart soft-robotic gripper system based on triboelectric nanogenerator sensors was described by Jin et al at the sensing stage in order to record continuous motion and tactile data for soft gripper control. To improve task performance and system understanding, data-or AIdriven methods are also included in various touch, haptic, and force sensing [28]. A humanoid robot was able to lift a weight of unknown mass through autonomous trial-and-error search thanks to Verner et al's implementation of online reinforcement learning via a fake digital twin at the controller stage, one level higher [27].…”

“…Additionally, the system may not be able to learn anything beneficial due to the constraint of dimensionality [27]. In [28], Matulis et al integrated digital twin and reinforcement learning for a robotic manipulator to plan pick-andplace motions; in [25], Liu et al proposed a multitasking oriented robot arm motion planning scheme based on deep reinforcement learning and twin synchro-control. These examples suggest that the combination of "DT + RL" is a promising approach (SG-factor) when a robotic system's digital twin matures and provides reliable data.…”

“…3.4. HRI and HRC One of the most significant advantages that DT could have for robot-involved scenarios in the context of Industry 4.0 is safer human-robot interaction and collaboration [28] (SG-factor). It seems sense that HRI and HRC situations would be more complicated and difficult than robot-only applications because of the variety and randomness of human behavior in addition to the uncertainties in the environment and sensors.…”

Intelligent approach based on FEM simulations and soft computing techniques for filling system design optimisation in sand casting processes

“…In sand casting process, the number of FSD parameters that can influence the MFB in the mould cavity is relatively large [43] which make the computation of all combinations of the studied FSD parameters, viz. S A , S B , S C , S D , G E and G F (see Fig.…”