Optimal design and control of dynamic systems under uncertainty: A probabilistic approach

This paper presents a new methodology for the simultaneous design and control of systems under random realizations in the disturbances. The key idea in this work is to perform a distribution analysis on the worst-case variability. Normal distribution functions, which approximate the actual distribution of the worst-case variability, are used to estimate the largest variability expected for the process variables at a user-defined probability limit. The resulting estimates in the worst-case variability are used to evaluate the process constraints, the system's dynamic performance and the process economics. The methodology was applied to simultaneously design and control a Continuous Stirred Tank Reactor (CSTR) process. A study on the computational demands required by the present method is presented and compared with a dynamic optimization-based methodology. The results show that the present methodology is a computationally efficient and practical tool that can be used to propose attractive (economical) process designs under uncertainty.

► A probabilistic-based simultaneous design and control methodology is presented.
► A worst-case variability analysis is used to assess the optimal process design.
► The method was applied to simultaneously design and control a CSTR process.
► The method is efficient when compared to dynamic optimization-based methodologies.