Predictive Maintenance

Cost effective and reliable plant operation is the key for success in any competitive industry.

With our knowledge and dedicated hardware, we help you tackle the sometimes elusive area of Predictive Maintenance and make sure you don’t fight windmills. We identify issues before they become problems and help you to fix things before they are broken.

Sensors

clockworkX firmly believes in a system that gives you the right data in the right place and at the right time, built for the tight margins in O&M. Our unique plug&play wireless sensor solution provides synchronized data acquisition from numerous sensor units installed on all your main components. And where others tackle the question of enough data with too much data, we tackle it with flexibility.

We focus on our plug&play ability. The central gateway is the only part that needs to be supplied with power, but it can be positioned wherever convenient. The operating range of the sensors is in the range of kilometers with a direct line-of-sight. They can be positioned freely with different mounts, on static as well as rotating components, our wireless system never needs to unwind. A typical installation takes only 30 minutes.

Each sensor provides high resolution, raw acceleration signals of up to three axes, as well as temperature data. Detectable frequencies range from 0 to 1500 Hz with a sampling from 4S/s up to 4kS/s. The dynamic range can be configured between ±2g to ±8g.

One of the key elements of the technology is the synchronized data acquisition ensured by the radio protocol. The raw acceleration signals coming from the sensor are synced to less than 0.1ms. We do not have to rely on an external time signal like GPS and transmit only what is relevant.

Our sensors are configurable through a web-based interface. We can set sampling frequency, sample time, sample resolution, sensitivity, and which axes to record, just to name a few.

With a wireless solution we cannot gather data 24/7. However, nothing in industry is built to last only a couple of hours. Recording accelerations of two axes @ 1000Hz/16s and 16bit resolution four times a day, we can detect problems well in advance and for five years without ever touching the battery. Our data enables key insights into many aspects of structural health monitoring and condition monitoring: (main) bearing health, gearbox health, operating modes, imbalance and more.

Have a look at our first Case Study.

SCADA Analysis

As with almost all machines, wind turbines are subject to ageing and degradation in performance over their lifetime. Our specialized algorithms provide insights into the past and present of your assets and offer recommendations for the future. With a detailed SCADA Data Analysis, we can assess whether your assets perform within the expected bounds and where the reasons lie, if they do not. We assess reliability, availability, and downtime in the context of the available resource. In a consulting approach and in close cooperation with you, we work out how to increase reliability and availability through an optimized Asset Management Strategy.

We can perform a standard power curve analysis. Based on statistical methods we can identify underproduction and classify actual availability of the turbine. Combining our statistical expertise with our domain knowledge, we tackle this problem from angles you might not expect.

“No wind turbine is an island.” We analyze wind parks in their entirety to spot issues or potentials stemming from the interaction between the turbines in a park.

CAPE

With our ClockworkX Advanced Processing Engine we aim to achieve what we call Holistic Predictive Maintenance. By combining not only CSM and SCADA systems, but also service and maintenance records, we can improve our prediction bounds further and offer you true insight into your assets.

We use probabilistic models to combine measured data series with expert knowledge and data about discrete events. Based on historic data, the models output predicted boundaries for operation and can detect anomalies in real time.

The full potential of this method can be utilized once all your non-sensor data is digitalized and available for the algorithms. STARDUST is our Semantic Life Cycle Database that can hold all your assets data and supports full integration with CAPE

Our Clients

Latest News

Robin Weber

Team Lead Data Science & Predictive Maintenance

Contact