Additive manufacturing (AM) has opened up new possibilities for manufacturing functional parts and objects with reduced costs and time. In future AM can be used by astronauts to produce required tools and parts in outer space. In medical sciences, AM has contributed to making cost-effective medical tools, prosthetic limbs, teeth and even a functional artificial heart. Integrating machine learning into AM technology has resulted in further reduced material wastage and improved accuracy.
The current problem is accuracy when it comes to parts that need to fit together with extreme precision. Machine learning is being used to solve 3D printing problems by using generative design and testing in the pre-fabrication stage itself.
The errors in the AM process can be caused due to three main error sources:
• The mathematical geometry error due to data conversion from computer-aided design (CAD) model to the standard file input.
• The process error due to machine errors and process characteristics.
• The type of material may cause error such as thermal shrinkage and material distortion arising from the rapid heating and cooling process.
Due to the layer-wise manufacturing in AM processes, the effect of errors is reflected both inside of each layer and between layers, thus resulting out of plane deviation of product.
ML improves assists computer vision technology to find microscopic cracks in machine parts and other microscopic irregularities. By deploying high-resolution cameras to film the printing process for each layer to record streaks, pits, divots and different patterns in the printing powder which are invisible to the naked eye. The machine learning platform then matches recorded powder patterns to defects revealed by CT scanners. The ML platform is programmed to use high-resolution camera footage and CT scan data to understand and predict defects in the printing process. Artificial intelligence and machine learning enable the 3D printer to perform an inspection of parts simultaneously when they are in progress to improve cost and time savings in the additive manufacturing industry.
ML further decides when the next layer should be placed by calculating the time and temperature of the material in the previous layer. This reduces error in the object due to expansion or shrinkage of material due to temperature and results in perfection in making the desired object.