Manufacturing Data Acquisition with SKF

Currently, a significant amount of collected machine data does not meet the requirements to be suitable for machine learning applications, thus hindering accuracy and efficiency gains. The key to overcoming this challenge lies in developing a solution for data acquisition that can efficiently achieve high sampling rates from, e.g., grinding applications, potentially leveraging edge computing. While magnetic bearings have already been optimized for this purpose, there is a need for a lighter version of the technology to cater to non-magnetic bearings.

The envisioned solution should encompass various components that address the high sampling rate requirement. The goal is to create a product that can effectively gather machine tool data with the necessary granularity for training machine learning algorithms. A more comprehensive and detailed dataset will improve the accuracy and efficiency of the machine learning models applied to these systems.

The sought-after solution should be a comprehensive package that tackles the challenge of achieving high sampling rates for machine tool data, e.g., vibration. Attractive solutions include edge computing systems capable of handling data at high frequencies and integration processes that can efficiently pack and manage data from multiple sources, allowing machine learning algorithms to process and analyze the data effectively. The solution should be scalable and adaptable to different machine tool setups and industrial environments, providing a versatile product catering to various applications.