Optimized Robust Compressor Station Design Methodology, 18-R8218
Principal Investigators
Benjamin A. White
Barron J. Bichon
David L. Ransom
Eugene L. Broerman
Inclusive Dates: 04/01/11 – 09/28/12
Background — Machinery reliability assessment has long been a significant portion of the work performed at SwRI. Since 1952, SwRI has provided gas compressor installation design analysis service to the natural gas industry. In the process of designing a gas compression facility, typically for pipeline applications, it is important to ensure the mechanical reliability of the system through proper analysis of the fluid and structural systems. Pulsation and structural models of reciprocating compressor systems are developed to study the dynamic pressure, flow characteristics and structural responses for the planned operating conditions. The predictions are then used to identify potential problems, and design modifications are studied to eliminate the problems. Machinery reliability is also addressed, and SwRI has worked with NASA and other organizations over the past two decades to develop a world-class approach to probabilistic design analysis. This work is performed using SwRI-developed commercial analysis tools known as NESSUS® and CENTAUR™. This project was focused on the developing methodologies for designing an "Optimized Robust Compressor Station." The goal was to develop a design practice that better incorporates the characteristics of the fluid and structural systems and efficiently determines the preferred robust, least-sensitive compressor pulsation filter bottle and piping system design.
Approach — The primary objectives of this project were:
- Develop an optimization methodology for the design of gas compression stations to maximize station reliability and efficiency.
- Develop a robust design methodology to minimize design sensitivity to typical operational variability and degradation of mechanical systems, such as piping restraints, over time.
Accomplishments — This project has been completed. The following key accomplishments were made during this project.
- Developed a new pulsation code post-processor for increased automation and cross platform compatibility necessary to perform the optimization and robustness tasks and added significant new capabilities to the structural model to allow for automation and importation of excitation loads from the fluid domain.
- Developed a simple one-cylinder compressor and piping model for testing and optimization of both fluid domain and structural characteristics.
- Developed computer coding that will automatically identify the worst load cases from the fluid system, optimize the full system to reduce vibration and bottle weight and recheck key parameters such as pressure drop and dynamic stress after the optimal solution is found. The robustness of the design can be checked after an optimal design is found or the robustness can be included as part of the design optimization.
- Demonstrated the feasibility of the optimization and robustness processes as applied to compressor design. The optimization process was able to provide improvements to the traditional design methods in several simultaneous areas (such as reducing bottle weight and vibration). Additionally, the reliability-based design optimization is able to successfully produce a design that is both optimized and robust to typical installation tolerances or identify a design that is not robust.
