Physics-driven analytics for connected mechanical systems
1. Collect: We start with two items you likely already have:
Piping and Instrumentation Diagrams (P&ID)
Historical Sensor Data. A copy of your historical file is typically where we start. If you are using spreadsheets (such as Excel) to analyze or review sensor data from your mechanical/industrial system, we can help you get to digital.
2. Digital Twin: Tignis provides a detailed view of the current state of your systems, with remarkable speed and accuracy. We create a digital replica of the piping and instrumentation systems, in just hours, and then handle ongoing changes with ease, by using our:
Patent-pending adaptive modeling capabilities that automatically handle changing conditions
Proprietary components library and drag and drop capabilities
Visualization capabilities, optimized for mechanical systems
3. Physics-Driven Analytics: Our analytics approach is unique. We use the principles of physics to drive our analysis in order to deliver better outcomes, while enabling you to leapfrog many of the obstacles limiting results today. Here is why:
It’s about the outcome. Prevailing machine learning approaches are based on a theory that if you collect enough data and use enough technology, you can accurately detect, predict and prevent failures. However, in reality, this kind of data collection and analysis is both costly and time consuming, as faults and failures are the exception rather than the rule.
The Tignis physics-driven approach analyzes physical properties, such as changes in temperature or water volumes and its significance within the operational systems, to detect anomalies. This allows Tignis to deliver immediate and highly accurate results. It not only makes us unique, it is what enables you to skip over today’s costly and time-consuming processes to realize results.
4. Ongoing Advantages: We build in advanced technology, machine learning and artificial intelligence, so that our algorithms and analytics are capable of delivering both initial results and ongoing advantages by:
Monitoring and learning – automatically
Detecting threats to reliability, even on diverse and complex systems – continuously
Identifying and predicting operational impacts – precisely