| 초록 |
This paper presents a systematic stochastic optimization method for a DMR (Dual Mixed Refrigerant) process modeled with the simulator Aspen HYSYS. At first, a base case design and an objective function are developed based on the simulator. Next, decision variables among many process variables are determined by the sensitive analysis. Among process variables, a sea-water temperature variation, which has a large impact on an operation cost, is considered as a random variable. Since it is not possible to use a deterministic optimization solver of the simulator for including a random variable, a PSO (Particle Swarm Optimization) technique, which is one of gradient free optimization tools, is employed to solve a stochastic optimization problem generated by sample average approximation. A case study shows the efficacy of the proposed algorithm. This method is general and can be applied to various processes modeled with the commercial simulators |
