By utilizing AI, it is possible to detect and predict machine anomalies that are difficult to assess with threshold judgment. This contributes to improving manufacturing quality and productivity through the detection of unusual sounds and vibrations.
Build an AI model to detect and predict machine anomalies from time-series data such as sensor data. Since it is possible to create simulated numerical data from normal to failure, the burden of data collection is reduced, allowing for the development of AI functions for failure prediction in a short period.
Detect "something is wrong (anomaly)" and predict "it will change (failure)." To perform anomaly detection and prediction of equipment using AI, a large amount of effective data for detecting and predicting those anomalies is required. AlliomWave can automatically generate a large amount of simulated waveform data necessary for learning from a small amount of actual waveform data. This enables the generation of high-precision AI models in a short time.
【Features】
■ Data Generation: Generate a large amount of simulated waveform data from a small amount of waveform data. The accuracy of the generated simulated waveform data can be determined by "similarity."
■ AI Creation: AI models can be completed in as little as 5 minutes.
■ AI Execution: Immediately available for purchase by integrating with YASKAWA Cockpit.
■ The generated simulated waveform data can be used in any AI development environment of your choice (optional feature).
■ AI Cube will select data suitable for your purpose from the data collected by the customer.
Sanmei is an agent (trading company) for AI Cube.
