DiagSense Preventive maintenance solutions, using innovative statistical algorithms; we employ state of the art proprietary algorithms for monitoring and predicting malfunctions in various mechanical systems, such as pumps, pipelines, turbines.
Our solution will save you – idle time due to repairs, on any kind of machinery and pipelines. We reduce dramatically false alarms on the monitored machinery.
Last but not least, our unique solution improves misdetection of malfunctions in mechanical equipment up to 99%. On water systems we offer the opportunity to prevent leaks, and optimize water flow and usage.
DiagSense provides a unique solution for customer’s mechanical systems maintenance problems.
Year of establishment: DiagSense started R&D phase at 2012
Background of the company
DiagSense is working on the field of algorithm development for predicting failures or abnormal behaviour in mechanical systems. Our partners have the experience and expertise required for effective design, development and integration of algorithms for mechanical systems on-line monitoring.
Examples of projects
Develop of valve controller for a water pressure management project. Valve controller supplies values of the pressure that must be set on the valve that controls the pressure. We achieve minimal and still acceptable water pressure on all points of the network.
Monitoring oil piping network: Our system monitors the flow, analyses the data in real time and calculates “health grades” for the network. These health grades are based on variety of factors and their statistical derivatives in time, so the alerts of abnormalities can be given before the formation of real failure.
Technologies & products
Develop algorithms for monitoring of mechanical processes, prognostics and diagnostics of evolving failures.
Function of the product(s):
Our system diagnoses the usability of mechanical systems in real-time, and alerts system-operators about developing malfunctions in components that could lead to significant damage.
The system uses the existing set of sensors installed on mechanical system, reads the data from those sensors, analyses the data and calculates “health grades” in each point of time.
These grades are calculated by using statistical learning of normal system behavior and continuous testing if current behavior fits the normal one.
Statistical training considers the factors that have influence on the diagnosed signal and allows early detection of evolving failures much before its formation. Early detection of failures before they materialize is critical to avoiding a wide variety of harms and costs.
Objectives / Target companies
Monitoring, forecasting and early warning: real-time fault detection on mechanical systems, aimed to prevent damages that could cause system failures.
Providing our clients with customer service of the highest quality – maximum usability of the core mechanical systems. Our customers benefit from a significant decrease of unplanned machine interruptions, and enhance their ability to plan maintenance with maximum efficiency.
Reducing operational costs due to minimizing the idle time, which results from having to service failed systems and reducing the false alarms.