The main goal of our D2Lab for Oil and Gas solution is to enable predictive maintenance in oil and gas domain and to prevent asset failure, detect quality issues and improve operational processes.
We collect all relevant data from multiple sources in real time, process and analyze this information using advanced predictive analytics, that can detect even minor anomalies and failure patterns. We use pattern recognition and anomaly detection algorithms to send alerts at the earliest signs of machine degradation.
These early warnings will help our customers to deploy limited resources wisely and effectively. Predictive maintenance costs much less time and money than unexpected breakdowns.
D2Lab allows us to deploy the monitoring solution on huge datasets, i.e. we can combine information and real time data streams from many machines and data sources. Our algorithms never sleep which means they are continuously improving and adapting when your operations change.
We have secure connectors and api’s to all major data historians vendors such as Osisoft and Honeywell. We can output our results into our user interface or integrate with existing solution. Due to our modular software architecture we are flexible on how we implement D2Lab in an organisation.
Our team is comprised of 15 multidisciplinary data scientists and engineers. Based in Den Haag (Netherlands), Karlsruhe (Germany) and Nis (Serbia) we focus on enabling data driven analytics solutions for manufacturing, transportation, additive manufacturing and oil & gas firms.
Compressor monitoring is commonly done on premises and based on thermodynamic and mass balance models. These models are unable to handle changes in conditions without human interference. Adding machine learning improves condition monitoring of these assets over time and helps reduce costly downtime and optimize maintenance planning and cost.
Power supply in any operational environment is one of the most critical aspects of your operations. Whether you are working with state of the art technology or ageing equipment we can add value by reducing downtime and overall maintenance cost.
Electrical Submersible pumps (ESP)
Maintaining your ESP’s in the ultimate operating envelope is often a challenge. Including historical data as a monitoring parameter gives you additional predictive analytical capabilities. We use machine learning to keep your ESP’s at perfect maintenance and operating conditions.
We offer to test our platform and algorithms on historical data. By applying the process described earlier we transform your historian data in to actionable insights, which from that point onwards can also be applied in real-time. If our initial screening is successful we would like to deploy our platform in real-time for a minimum commitment of 6 months.