| Summary: | 碩士 === 國立臺灣科技大學 === 化學工程系 === 90 === Abstract
Last decade has seen significant progress in the design of plantwide control systems and most of the work addresses the issue of control structure design or the effects of material recycle on overall process dynamics. The timely publication of Luyben et al. (1999) provides an updated summary. However, much less work has been done on the practically important process: heat-integrated recycle plants where both material and energy recycles exist simultaneously. The heat-integration is a trend of no-return for energy intensive chemical process industries. This work analyzes the tradeoff between steady-state economics and dynamic controllability for heat-integrated recycle plants. The process consists of one reactor, two distillation columns, and two recycle streams first studied by Tyreus and Luyben (1993) and further explored by Cheng and Yu (2002) and, in this work, the two distillation columns are heat-integrated. At steady-state design, the concept of optimality regions for column sequencing (Glinos and Malone, 1988) is extended to the heat-integrated recycle plants. First, a boundary in the composition space can be established to identify the trajectory with most significant percent energy savings as well as correct direction for heat-integration (e.g., forward or backward integration). Because the design problem differs from the column sequencing problem in that we can design the reactor composition, optimal trajectories for recycle plants with direct and indirect sequences are derived as the conversion varies. Provided with the direction of heat-integration, at any given conversion, the correct flowsheet is established for both sequences. Moreover, the results can be derived analytically using simplified cost model of Malone et al. (1985). For dynamic controllability, the reachable composition space is identified as the recycle ratios (recycle flow rate/ production rate) vary. This provides the most severe test on the disturbance rejection capability for any given design, and, consequently, the undesirable regions for operation with different separation sequences are identified. The results clearly indicate that a little tradeoff between steady-state design and dynamic control may result at low conversion (e.g., less than 3%), however, at medium to high conversions, it is not likely to occur, if the plant is designed along the optimal trajectories. Moreover, at the true optimum, the trade-off is not likely to occur. Finally, rigorous nonlinear simulations are used to illustrate the operability of different designs. The results reveal that good control can be achieved for well designed heat-integrated recycle plants (e.g., compared to the plants without energy integration) with close to 20% saving in total annual cost.
|
|---|
