The basis for the BRU21 program area on operation, maintenance, safety & security is the opportunities we see in light of the digitalization & automation taking place these days. Important aspects are predictive maintenance/real-time monitoring of asset condition, Big data analytics and artificial intelligence (AI)/machine learning (ML), as well as “physical” achievements like the use of drones and other autonomous tools. There are also concerns related to intro-duced vulnerability and cyber security, which become even more important than before.

Challenges and opportunities

Predictive maintenance means that maintenance decisions are based on information regarding the current state of components and assets and future state development. There is a large variety of condition monitoring technolo-gies available today to assess the current state. This is often referred to as diagnostics. For rotating equipment, signal processing of vibration measurements has been available for decades, similarly to ultrasonic measurement and magnetic particle inspection utilized for structure inspec-tion. A challenge is still how to use these models and tools for prognostics, i.e., prediction of remaining useful lifetime required for maintenance planning.

For larger systems involved in the production and processing of hydrocarbons, there is no single sensor that can capture the state of the system. The idea is to use information for a set of sensors to monitor relevant process parameters like pressure, temperature and flow. Then surveillance systems are trained to recognize the normal situation, and then establish algorithms for efficient anomaly detection. The literature reports on successful implementation of both machine learning algorithms and algorithms based on the so-called first principles. However, to scale up such approaches for an entire installation seems challenging and we see the need for “self learning” approaches in order to implement anomaly detection on a large scale.

A digital twin is a digital representation of a real-world entity or system. The implementation of a digital twin is an encapsulated software object or model that mirrors for example a physical system, historical and future maintenance activities, or an operational plan. Data from multiple digital twins can be aggregated for a composite view. The notion of a digital representation of real-world entities or systems is not new. Its heritage goes back to computer-aided design representations of physical assets or profiles of individual customers. The difference in the latest iteration of digital twins (adopted from Gartner Top 10 Strategic Technology Trends for 2019) is:

(i) Robustness of the models with a focus on high reliability and efficient maintenance, (ii) digital twins’ link to the real world, potentially in real-time for monitoring and control, (iii) application of advanced Big data analytics and AI/ML, and (iv) ability to interact with them and evaluate “what-if” scenarios.In the O&G industry we see a huge effort in implementing digital twins. Systems drawings, system documentation, available data and sensor information are connected to represent a digital twin of the system. However, these digital twins often suffer from lack of comprehensive mathematical system models which enable what-if scenario analyses required to optimize maintenance and operations decisions. This is an important topic for our research

Cyber safety and cyber security: There is no free lunch! This also applies in relation to the digitalization taking place these days. Cyber security and safety have to be dealt with in a systematic manner. Several malicious cyber-at-tacks have been reported in the literature. Examples are the Maroochy Water Breach event, the Stuxnet malicious computer worm causing substantial damage to Iran’s nuclear program, and finally the Hydro aluminium security attack where costs after just one week were estimated to 30 million euro. In addition to malicious acts we also have cyber safety challenges like the Boeing MAX 737 catastrophe exemplifying how automated solutions might fail, and also the challenges related to integrating tech-nology, people and organizations.

Research strategy

The research on operation, maintenance, safety & security will focus on developing digital tools, solutions and new ways of working for optimized and safe operation:

• Predictive maintenance

• Anomaly detection, fault diagnosis and remaining useful lifetime prediction

• Decision support models for planning and optimizing maintenance

• Safety and risk analysis

• Control and safety systems technologies

• ICT/Cyber-security analysis

• Requirements for, development of and demonstration on the use of digital twins