sensXPERT, intended for processors in the plastics industry, uses real-time material data from the mold and machine learning software to analyze material behavior and carry out direct in-process quality control of a single molded part, the company said.
According to Netzsch, the technology incorporates in-mold sensors that can provide real-time insights and transparency to react to material deviations and eliminate scrap, leading to up to 30% increase in production efficiency. “While allowing a dynamic and adaptive production, thus maximizing throughput, sensXPERT ensures direct in-process quality control of each single molded part,” a press release added.
In one example, the manufacturing of automotive composite wheels and airplane wing components showed significant increases in overall equipment efficiency (OEE), including solid return of investment (ROI), Netzsch said.
sensXPERT features an ‘Edge Device’ which integrates the hardware and software for machine learning models and can capture small deviations of material and process parameters. Based on measuring data collected from high-precision in-mold sensors, smart machine learning algorithms are applied to simulate, predict and analyze the actual material behavior on each individual machine. The learning models are trained with parameters, such as standard material and experimental data, glass transition temperature, pressure and required degree of curing, and are then continuously fine-tuned depending on the in-situ data measured over time.
Netzsch says that the technology can integrate with third-party sensors, manufacturing and control systems, production machines and molds.
sensXPERT can be used with common thermosets, thermoplastics and elastomers processing technique, from injection, compression and transfer molding to thermoforming, vacuum infusion and autoclave curing.
To register for the digital launch of the technology on 22 September 2022, go here.
