Tools and methods to manage risk in autonomous subsea inspection, maintenance and repair operations

Abstract

Autonomous subsea interventions are anticipated to decrease operational costs, reduce response time and optimally maintain the subsea infrastructure. However, the introduction of autonomy may lead to emerging risk factors in subsea intervention operations. Research on managing risk in future autonomous subsea interventions is scarce. At the same time, the industry and research communities are spearheading this technological change in subsea inspection, maintenance, and repair (IMR) operations by developing and demonstrating new concepts to realize autonomous subsea interventions. Techniques to identify, assess, and manage risk factors affecting autonomous subsea IMR operations are therefore required. The purpose of this thesis is to develop novel tools and methods to manage risk in autonomous subsea IMR operations. Gaps in the industry standards, which lay requirements for current subsea interventions, have been mapped. The results show that technology and knowledge gaps exist in realizing autonomous subsea interventions and that the current standards are only partly applicable to future IMR systems. Risk influencing factors inherent in autonomous subsea interventions have been identified and analyzed. A Bayesian belief network is proposed to derive the probability of aborting an autonomous subsea IMR operation. Monitoring risk-influencing factors in terms of risk indicators can contribute to improved situational awareness and path planning. The proposed risk based indicators can highlight risk trends for the autonomous remotely operated vehicle. Vehicle behavior under faults, failures and exposure to surrounding subsea obstacles has been explored in this thesis. A fuzzy inference system is proposed to derive a decision support basis for the autonomous remotely operated vehicle to make decisions to either continue or discontinue an IMR operation. It is observed that fuzzy logic can be used to suggest appropriate safe actions when component faults or failures occur. Concerning avoiding collision with surrounding obstacles, a novel underwater collision avoidance system is proposed consisting of safety envelopes around the autonomous remotely operated vehicle and subsea traffic rules. The subsea traffic rules are proposed for known static and dynamic obstacles in the vicinity of the autonomous remotely operated vehicle. Overall, researchers, original equipment manufacturers, subsea system developers, and safety regulating bodies may benefit from the results of this thesis. The proposed tools and methods contribute to efficient identification, assessment and management of risks during autonomous subsea IMR operations.