Edge-Based RNN Anomaly Detection Platform in Machine Tools

ABSTRACT

With the rapid advances in machine learning algorithms and sensing technologies, machine prognostics and health management (PHM) via data-driven approaches has become a trend in sophisticated machine tool industry. The run-to-failure data are necessary for data-driven approaches. However, the average life of the machine is two to three years, the time of collecting data is extended. It is a big challenge to collect run-to-failure data and build a PHM model. Therefore, we propose an Edge-based RNN Anomaly Detection Platform (ERADP). ERADP builds the model based on healthy data and notify anomalies two hours in advance. The true alarm rate is up to 100%. Besides, ERADP can accelerate the training time almost 120 times faster than the traditional model.

Graphical abstract

We propose an Edge-based RNN Anomaly Detection Platform (ERADP) to solve the data-imbalance issue and demonstrate detect anomalies in real time for machinery industry. ERADP can make the true alarm rate up to 100% and speed up model training almost 120 times faster. Besides, we cooperate with TongTai, which is the biggest machine tool company in Taiwan. Equipped ERADP with machine tools, the cost of repairing and failure products can be intensively decreased. The price of machine tools can increase by 6%. The revenue of the machinery industry can increase by about 0.27 billion US dollars. ERADP can really make a significant impact on the machinery industry.

KEYWORDS:

• Prognostics and health management (PHM)
• anomaly detection (AD)
• recurrent neural network (RNN)

Disclosure statement

No potential conflict of interest was reported by the authors.

Additional information

Funding

This work is supported by Minstry of Science and Technology, Taiwan, under the project 106-2634-F-009-002-CC2.