The Evolution of Energy Analytics

Robert Skiebe, Project Director, Siemens And Dr. Alec Gruss, Head of Digital Solutions for the Dresser-Rand business, part of Siemens Power and Gas

Robert Skiebe, Project Director, Siemens

The oil and gas industry has been collecting data and automating its systems for decades now; however, much of this data is gathered from different types of equipment on an “as-needed basis.” Because of this, only small “slices” of information are available for a particular piece of equipment. As a result, the data collected is highly uncorrelated and non-actionable, which makes for a time-consuming, manual data aggregation analysis that can deter time-sensitive decisions.

By combining all the institutional knowledge of the energy domain and the e quipment information, however, analytics can help companies architect solutions in a structured format. Using advanced analytics, enterprises can become more efficient in their operations, thereby improving productivity and reducing unplanned outages.

"The foremost challenge that most companies face is to gather actionable data"

Unlike other industries, data analytics in the energy sector was adopted at a much later stage. The first wave of energy analytics was in the form of small-scale data analysis, which could only compare individual or a small numbe r of equipment in a production line or refinery at a time. The opportunity and the challenge now lies in finding ways to scale the process globally, enabling companies to analyse turbines across all their units at one go and normalize operations based on different factors like weather, operating conditions, or production medium (oil, water, and gas).

Siemens has been doing vibration analysis to detect equipment faults and downtime as a component of their condition monitoring and predictive maintenance offerings for close to two decades now. Thus the company has the pre-requisite to understand what an enterprise needs to do in terms of its data collection and processing to make the energy equipment data available at a large scale, much more efficiently to conduct in-depth analytics.

Current Challenges in Most Companies

The foremost challenge that most companies face is to gather data and bring it into a single platform where it can be analysed and value generated out of it in a short amount of time. Currently, if a company needs to solve a certain operational problem, they typically end up spending weeks or even months (depending upon the scope of the problem) to collect and aggregate the necessary data.

The other concern is addressing data silos and data security. As a market leader, spending years interacting with the industry, Siemens has continuously confronted these challenges. Over the years, we have articulated solutions on how to solve cybersecurity problems and the methods to collect the right data and make it available to be analysed in a timely manner.

Dr. Alec Gruss, Head of Digital Solutions for the Dresser-Rand business, part of Siemens Power and Gas

Another significant challenge that Siemens has successfully resolved is allocating institutional skills and expertise throughout the organization in a cohesive, cost-effective manner. It’s about bringing the right information together to enable heretofore unknown interactions and influences (say a valve in danger of failing due to a pressure gradient further upstream in a process) to be discovered. Traditionally, companies would hire subject matter experts (SMEs) across all their facilities around the globe and deploy them to maintain a full coverage of skilled support. This process, however, is an expensive proposition.

Advice to the Technology Leaders

Our recommendation is to understand the end goal. It is easy to get wrapped up in technology, but technology should always serve the business requirements. Integrating a solution without a w ell-thought-through strategy will have little impact on the productivity or growth of the organization and potentially set progress back by months or more. Simply gathering, analysing, and visualizing data without empowering the enterprise to effectively change its operations on the basis of tha t data marginalizes the digitalization value proposition. A willingness to change an enterprise’s culture and processes is critical: it is the business model that should lead and the technology should follow. Leaders should, therefore, employ a strategic approach and not an opportunistic approach. Digitalization, when implemented correctly, is a powerful tool that can truly transform a business and make it more competitive.

Predictive Maintenance

Data Security