A Multi-Objective Perspective To Satellite Design and Reliability Optimization

Abstract

Development of a communication satellite project is highly complicated and expensive which costs a few hundred million dollars depending on the mission in space. Once a satellite is launched into orbit, it has to operate in harsh environmental conditions including radiation, solar activity, meteorites, and extreme weather patterns. Since there is no possibility of physical maintenance intervention in space, reliability is a critical attribute for all space and satellite projects. Therefore, the redundancy philosophy and reliability measures are taken into account in the design phase of a satellite to prevent the loss of functionality in case of a failure in orbit. This study aims to optimize the payload design of a communication satellite by considering the system's reliability, power consumption and cost simultaneously. Since these objectives are conflicting in their nature, a multi-objective optimization approach is proposed. We offer a systematic approach to the satellite design by determining the best redundancy strategy considering contradictory objectives and onboard constraints in the multibillion-dollar satellite industry. The proposed approach promotes trade-offs and sensitivity analyses between cost, power consumption and system reliability in the early design phase of satellites using Compromise Programming. By using different sets of weights for the objectives in our model, it is possible to address different types of satellites depending on their mission and priorities. Because of the NP-Hard characteristics of the reliability optimization problem and the nonlinear equation in the proposed model, the Simulated Annealing algorithm is utilized to solve the problem. As a case analysis, the implementation is carried out on the design of a communication satellite system with active hot-standby and warm-standby onboard redundancy schemes. Results reveal that huge savings in million dollars can be attained as a result of approximately 5% reduction in reliability. © 2024 Elsevier Ltd