Model-based prediction of part characteristics
How does the system work? The quality predictions are based on models that make it possible to calculate the characteristics of injection-molded parts. This cutting-edge technology makes use of supervised machine learning algorithms based on neural networks. The first step is to generate a statistical test plan (Design of Experiments, or DoE) which is worked through with ComoNeo on the machine. ComoNeoPREDICT then uses the parameters obtained (such as pressure and temperature curves) and the measured quality characteristics of the part to create a prediction model that serves as a reference for subsequent production runs. If a part is outside of the defined tolerance band limits, it can be separated out automatically. Special PC software is provided to generate the DoE and the prediction model.
Users of the new ComoNeoPREDICT benefit from a convincing range of advantages:
- Accurate scrap separation method thanks to complete quality records
- Users acquire know-how from direct monitoring of predicted quality characteristics
- Pseudoscrap is minimized
- Tolerances and scrap criteria are easily defined
- Relevant curve points for calculation are selected automatically
- No special knowledge of math or statistics is needed to operate the software
- Ongoing training of models results in continuous improvement
This standardized and documented procedure means that tests and model calculations can be performed regardless of which individuals are available to operate the system. The transparency of the method gives users a far better understanding of the process. Manufacturers of sensitive precision parts for the medical technology sector and producers of other critical high-caliber components will benefit from this new function, which gives users extensive control of all quality features in the injection molding process.
ComoNeo: enabling smart production in keeping with Industry 4.0
ComoNeo is the core module in the digital value chain, both at horizontal process level on the shop floor and at vertical level for overarching systems.
At process level, ComoNeo offers features for a variety of functions: for example, ComoNeoMERGE provides cavity pressure-based monitoring of conventional and multi-component injection molding processes. ComoNeoSWITCH allows switchover of the injection molding machine in response to cavity pressure, while ComoNeoMULTIFLOW handles balancing of the hot runner control system.
All quality-related process data generated with ComoNeo can be transmitted via OPC UA to higher-level systems such as MES or ERP. All of these advantages make ComoNeo the launch pad for your smart production.