Prescient's new Predictive Kit
24/06/2022
Prescient, a supplier of cloud-to-edge data automation software and solutions, has announced the release of its low-code predictive maintenance solution kit. Developed jointly with National Control Devices (NCD), a top provider of long-range industrial wireless sensors, and WAGO, a leader in interconnect, interface and automation solutions, the new Prescient Predictive Kit is a customisable and extensible predictive maintenance solution.
Predictive maintenance is one of the most popular Industry 4.0 applications. Using sensor data such as vibration, current and temperature to predict machine failures often weeks in advance, predictive maintenance systems deliver significant cost savings due to unplanned downtime. According to Statista, the market will grow from US$4.5 billion (approximately £3.57 billion) in 2020 to
US$63.3 billion (approximately £50.3 billion) in 2030, exhibiting a compound annual growth rate of 30% during the forecasted period.
Today’s predictive maintenance solutions are either turnkey solutions with limited flexibility or custom-built solutions that require significant technological development. Prescient, NCD and WAGO are releasing what they claim is the industry’s first low-code predictive maintenance solution kit. This kit enables customers to customise its predictive maintenance solution without requiring software development expertise.
The Prescient Predictive Kit includes an NCD wireless predictive maintenance sensor, a WAGO edge computer and a low-code predictive maintenance template dataflow inside Prescient’s flagship data automation software, Prescient Designer.
The NCD wireless predictive maintenance sensor is one of the most advanced on the market. It includes a vibration sensor with a frequency range between 1.56 Hz and 6.4 kHz and a sample rate of up to 25.6 kHz. It also includes an AC current sensor with a range up to 100 A RMS and a high-grade K-type thermocouple temperature sensor with a rating of 260°C. The NCD wireless predictive maintenance sensor has an encrypted communication range of up to two miles.
Anil Bhaskar, NCD’s CEO, said: “Our predictive maintenance sensor is specifically built to predict the failure of rotary machines based on vibration, current and temperature data. It is one of the most advanced predictive maintenance sensors on the market. Its industrial performance and long-range wireless capability mean that it can monitor machines operating in the most difficult environments.”
The Linux-based WAGO Edge Computer features an Intel Atom E3845 Quadcore 1.91 GHz processor, 8 GB of RAM and supports a variety of physical interfaces. In addition, the Prescient Predictive Kit is compatible with WAGO’s full line of edge controllers and computers, ranging from small, compact controllers to powerful Intel i7-based edge computers, giving customers full flexibility in choosing the edge-computing power they need.
“Modern control systems require highly capable computers and that is where the WAGO 752 Edge computer family fits,” commented Jesse Cox, WAGO’s Senior Application Engineer. “We can pair these with low-code technology innovations such as Prescient Designer and Prescient Edge runtime and hardware leaders such as NCD sensors to bring the most advanced functionality to your control system’s edge-of-network.”
The Prescient Predictive Kit dataflow is a low-code solution. It allows customers to collect NCD wireless sensor data, quickly create alerts and display sensor data and alerts on a local or cloud-based dashboard. Customisation capability is critical, as each customer may have different requirements.
“At Prescient, we have been working on predictive maintenance projects for many years,” said Ashish Yadav, Prescient’s Director of Software Development. “We have seen many different requirements, so we are happy to be able to bring the benefits of this experience to the marketplace.”
The Prescient low-code solution allows customers to quickly visualise sensor data and customise the detection algorithm any way they want. For example, vibration data could be analysed in the time domain or the frequency domain or the data could be fed into a machine learning model.
