The evolution of Industrial Internet of Things (IIoT) boosts the availability of real-time insights and transformational business outcomes through the massive streams of data and analytics. The advancement in wireless connectivity enables mobility and flexibility in the transferences of such data and insights at a fraction of the cost compared to wired system. However a few technical challenges must be addressed to ensure the high performance of robustness, availability, reliability, and latency of the wireless system to handle these digital transformation.
The major challenge in developing an intelligent IIoT system for key applications, such as preventive and predictive maintenance, adaptive process control, etc. is two-folds. Infrastructure-wise, the sensor network-based monitoring system needs to be highly efficient and reliable for accurate real-time sensory data collection and transmission. Algorithm-wise, the prediction model needs to accurately evaluate the machines and/or process conditions and predict the outcomes; it also needs to identify proper adjustment and action to optimize productivity at all times.
In the bid to enhance the delivery of such system, IHI and A*STAR I2R collaborated to develop an intelligent and low power IIoT edge platform based on IHI IoT boards. Developed with highly optimized cognitive industrial wireless and advanced edge analytics that support multiple wireless protocols for real-time model-based analysis such as predictive maintenance, targeting high reliability and high availability performance with minimum of 99%. Together with IHI, the technologies will be integrated as a system and provided as a total solution.
The team will be working on two key technologies on the IIoT edge computing platform: (i) high reliability and cognitive industrial wireless, supporting industrial-grade data acquisition and transmission; and (ii) advanced edge analytics.
The wireless system is built with three main features; namely self-configuration, self-optimization, and self-healing to achieve high reliability. This is achieved through the built-in cognitive learning capability at the edge platform.
Self-configuration realizes the ‘plug-and-play’ paradigm that enables multi-protocol wireless gateways and devices to be automatically configured and securely integrated into the industry network.
Self-optimization function continuously monitors network performance and triggers appropriate actions to ensure seamless connectivity, mitigate interference and improve mobility robustness.
Self-healing is the ability to intelligently detect, diagnose and rapidly recover the wireless system from faults, regardless of the source of failures. This avoids the occurrence of connectivity outage that adversely impacts business operations, given that IIoT system operates in dynamic and harsh environment.
Building artificial intelligence into a fully connected edge computing platform unleashes the full potential of IIoT. One of our core technologies enables the deployment of model-based learning algorithms into the small edge gateways and devices. This is made possible via highly optimized machine learning and deep learning algorithms that are further accelerated with Field-Programmable Gate Array (FPGA) implementation to deliver low latency and more accurate predictions.
*IHI is a comprehensive heavy-industry company in Japan, creating value for customers in four main areas: Industrial systems and general-purpose machinery; Resource, energy and environment; Social infrastructure and offshore facilities; and Aero engine, space and defense (http://www.ihi.co.jp/en/).