From Months to Minutes – Artificial Intelligence Achieves It for Structural Design of Fixed Offshore Platforms

Abstract Traditionally, structural design of offshore platforms requires a considerable portion of total engineering time, ranging from several weeks to months, depending on its complexity and design specifications’ requirements. In most real-life cases, this long duration is mainly due to waiting for non-structural inputs and resolving interdisciplinary issues. The present work proposes an AI-aided and knowledge-based method to configure and size of offshore platforms considering only a few basic input parameters when topside details are unavailable. The input includes the type of platform, estimated gross weight of topside facilities, water depth, design wave height, seismic coefficient and number of well conductors, as applicable. The new technique bypasses the commonly adopted sequence of structural sizing. The computer program developed for the purpose generates a detailed structural analytical model and summarized material take-off (MTO) for structural steel. The built-in knowledge of the program determines the size of the deck structure considering the available experience on the equipment layout. The program then develops the size of the jacket and deck members based on existing engineering knowledge on structural design. The design is considered preliminary, as all mandatory analyses will be performed later when equipment details are available. However, preliminary structural analyses can be performed immediately with the equivalent and conservative uniformly distributed applied loads on the deck. The program takes only a few seconds to generate an entire structure including the conductors and its connections, boat landing, mudmats, crane pedestal, flare boom, helideck and number of anodes, as applicable. The analytical model can be updated with a single click when the equipment details are available and provided in a tabular format. During detailed design at various stages of a project, the design may require minor update without considerable impact to the generated MTO. Numerical studies show a number of examples of auto-designed structures at up to 50 m water depth that can withstand omnidirectional environmental forces due to 15.0 m high waves. Structural design is performed using API RP2A working stress design method. The results show adequacy of the structures with the required minimum reserve strength and acceptable values of natural time periods. The auto-design program can greatly help in carrying out feasibility studies by creating different optional structural configurations. The generated MTO for each configuration can be used for estimating the fabrication and installation costs.