IHPC Tech Hub showcases IHPC's suite of in-house developed applications, tools or technology to help you unlock the possibilities to overcome business challenges. Through providing valuable insights, you can now predict and shape the commercial outcomes, automate processes, and free up resources for repetitive and labour-intensive tasks.
Discover the power of computational modelling, simulation and AI that brings about positive impact to your business.
System for Prediction Of Remaining useful life of Tool (SPORT)
01 Jan 2021
In advanced manufacturing industries, there are increased requirements for predictive maintenance technologies. Predictive maintenance refers to maintenance performed after the occurrence of fault but before failure. Using such technologies, we can potentially
improve the reliability, availability and productivity of systems. In adopting predictive maintenance, the key is to accurately estimate the trends of the health degradation and predict the Remaining Useful Life (RUL) through prognostics modelling.
This will ultimately facilitate effective operations planning and timely maintenance.
In essence, SPORT is an end-to-end system capable of predicting and estimating the RUL of inserts, using raw vibration signals during the cutting process. SPORT has been developed with pre-trained models for predicting the state and RUL of the insert
whereby these models are independent of the cutting speeds.
The novelty of SPORT as follows:
- Relative degradation for each cut in determining the tool wear
- Unified model capable of estimating the tool wear / RUL under different cutting speeds
- Improve accuracy in RUL prediction
- Able to deal with uncertainties from sensor data
- Model-based RUL prediction under varying operating settings
- Real-time dashboard for tool wear conditions
The Science Behind
SPORT is a unified data-driven model capable of accurately predicting the tool wear using only off-line tool wear data and on-line sensor observations. In generating the unified model applicable across different cutting speeds, individual machine learning
models for each cutting speed served as inputs and had their parameters aggregated in the final model. Using data from vibration signals, results from the analyses showed that the model would accurately predict and estimate the RUL of the tool wear.
Furthermore, since the unified model is built on being invariant across cutting conditions, it is suitable for different scenarios while using minimal data.
SPORT could benefit industries in their push to adopt IIoT technologies within their shop floors and integrate devices with cutting tools such as grinding machines, drilling machines, etc.
For more info or collaboration opportunities, please write to email@example.com.