Monitor critical components
Monitor critical components and systems with automated anomaly detection and get a warning when behavior outside the norm is detected. Recognize required maintenance in time and schedule it during planned downtimes avoiding fails and maintenance during production times.
Critical components can be found in large numbers in many processes. So, even if one component fails only rarely, the large number of components leads to unplanned process interruptions due to component failure over and over again. This has a negative impact on productivity. With a suitable monitoring solution, impending component failures can be detected and dealt with in a timely manner before they actually occur.
Automated Anomaly Detection
Usually, a critical component provides some tens to several hundreds of continuous data signals. Only seldom is it clear which of the signals need to be monitored for abnormal behavior. Even worse, anomalies rarely show up clearly in a single signal, but rather in changing relationships of critical signals to each other. Each signal on its own might continue showing normal behavior.
This is, why conventional anomaly detection requires a lot of domain knowledge and manual preparation making it expensive and hard to scale. aivis, on the other hand, provides automated anomaly detection: You simply define a target signal from which you think a component malfunction can best be read. aivis then automatically analyzes all available signals and their interrelationships to the target and with each other and uses them to build a model.
How to monitor critical components with aivis
Watching critical components with aivis is incredibly easy. You will be astonished about the quality and efficiency gain when using aivis on your critical components raw data.
Historical time-series data
Step 1: Setting the goal
aivis requires historical data of all available data streams of the component. This data is the foundation for aivis to learn, what normal behavior of this component looks like. A real or synthetic signal has to be defined as the target, for which anomalies should be detected based on all available data.
Step 2: Automated analysis
aivis then fully automated searches the data for all relevant relationships that help to determine the normality of the target signal. aivis creates a model, that recognizes abnormal behavior of the component.
Automated machine learning
Model warning about anomalies
Step 3: Deploying the model
The built model is an independent, lightweight software module. Once deployed and fed with the components live data stream, the operator is warned in time if the component acts out of usual behavior.
Check out this selection of application examples, where aivis has already successfully been applied in the past. Since aivis is industry-agnostic, it can be applied to countless other scenarios as well.
Goal: Predicitive maintenance of proportional valves
Proportional valves are important components for controlling volumetric flow rates and are used in numerous cases. If a valve fails during production operation, it requires unplanned maintenance, which is costly and can lead to bottlenecks. Therefore, it is important to be able to accurately monitor when a valve is no longer behaving normally so that it can be replaced in a timely manner during a scheduled maintenance interval.
Power & Energy
Goal: Predicitive maintenance for wind farms
The maintenance of wind turbines is a great challenge. Often the turbines are located in remote places that are difficult to access. This makes maintenance costly, so it should never be done unnecessarily. However, if a wind turbine fails, this often causes downtimes of a week or longer before the repair can be done. Therefore, it is crucial for the profitability of wind parks to know as precisely as possible when which turbine needs which kind of maintenance.
Power & Energy
Goal: Preventing vibration crisis of turbines
If the vibrations of a turbine axis exceed a certain maximum level, a vibration crisis occurs. If uncontained, such an event could lead to a major breakdown.