A quantum-based diagnostics approach for additive manufacturing machine

Abstract

Current Additive Manufacturing machines have limited techniques to observe process conditions and to decrease process errors. In order to overcome these limitations and increase the level and accuracy of machine intelligence, machine conditions need to be monitored more meticulously. A novel method for the condition monitoring of a 3D-printer is proposed in this paper. Quantum support vector machine (QSVM) is compiled for recognizing the health condition of the 3D-printer. The proposed quantum machine learning approach helps in monitoring the health state of the machine and classifies the same as healthy or aberrant. Classical machine learning approaches are inefficient to process the large amount of experimental data in real time. For better decision-making on such big data, quantum machine learning approaches are deployed which are much more efficient due to their exponential speed and parallel operation on complex sensor data, they show speedups in both the dimensionality and number of experimental data deployed to train the algorithm. The simulation results show that the proposed method has higher accuracy in fault diagnosis than the traditional Support Vector Machine. All the numerical simulations and experiments have been carried out on a real quantum hardware provided by the IBM Quantum computing over the cloud.