Chemical Process Transition Control
Supervisor:luo xiong zuo
Chemical process systems are characterized by complicated flowsheet,varied production conditions,and huge energy consumption.Because of the varied production conditions for chemical process,the large-scale change of working conditions is frequent and familiar.In addition,the smooth transition between the working conditions is challenging with complicated flowsheet under conventional control strategy.Process transition is defined as the large-scale change of working conditions,whereas the process transition control is the control strategy of process transition.In this work,there are three aspects that have been discussed,namely dynamic modeling,steady state optimization,and dynamic control.First of all,a novel dynamic modeling approach is proposed for distillation column with the mixing of vapor and liquid phase,and the modeling approach is employed for the discussion of a commercial ethylene column in cracking ethylene complex.For the previous research on modeling approaches,the dynamic process on the trays is considered,whereas the dynamic process on the downcomers and vapor spaces is ignored.In this work,considering the holdup of downcomers and vapor spaces,the completed dynamic mechanism model is established.Furthermore,the correctness and reasonability of the novel dynamic modeling approach are verified by the simulation of the ethylene column.Next,for the process transition of the ethylene column,the steady state optimization is discussed.In detail,the optimal end-point of process transition is obtained by the steady state optimization based on the change of production conditions.Note that the steady state model is a special form of the dynamic model established in the early.In this work,both the economic benefits of subsequent production and the operability of transition process are discussed for the optimization of working conditions.By introducing the energy efficiency coefficient,the economic benefits are quantified as the total energy consumption,whereas the operability is quantified as the variation of vapor and liquid loads.Furthermore,the balanced optimization model is established,whose objectives are economic benefits and operability.The multiobjective optimization problem is solved by variable constraint method,and the noninferior set(Pareto set)is obtained.The optimal solution,namely the optimal working condition,is obtained by a trade-off in Pareto set.Based on the steady state optimization of working conditions,the dynamic transition between different working conditions is studied.In general,the target of discussing process transition is achieving the working conditions transition by a simple and reliable control strategy under the premise of satisfying production index requirements.This work discusses the dynamic control of process transition in the following aspects.(1)Two feasible control strategies,namely conventional control and dynamic optimization,are discussed for process transition.First of all,the influence of control structure on control performance is discussed,which indicates the necessity of the study on dynamic optimization.For dynamic optimization,considering the amount of differential variables is large,the dynamic optimization problem is transformed into a mixed integer nonlinear programming problem by control vector parameterization.The optimal solution,namely the optimal transition trajectory,is obtained by the alternation between integer variables and real variables optimization.(2)Considering the coordination between conventional control and dynamic optimization,the adjustment scheme of optimization structure is discussed.For adjusted optimization structure,the original control structure is retained.The optimization object is the integration of conventional controllers and actual processes,whereas the optimization variable is the input of conventional controllers.The change of the control system structure is avoided by the optimization structure adjustment.In addition,by introducing concept of equivalent filter,the positive influence of optimization structure adjustment on the control performance of process transition is illustrated.Furthermore,the control performance is improved by tuning the parameters of equivalent filter.(3)Considering the selection between conventional control and dynamic optimization,the selection criterion of control strategies is proposed.While the production condition is changed,it is necessary to judge whether the change of working conditions is beyond the adjustable range for conventional control.In this work,the intermediate variables are introduced,and the mapping relationship between the intermediate variables and production index is obtained under conventional control strategy.For an unfamiliar process transition,the intermediate variables are calculated by steady state optimization,and the production index is estimated by the mapping relationship.A more suitable control strategy is selected by the estimated value of the production index.Furthermore,the applicability of the selection method is verified by the simulation of the ethylene column process transition.Finally,the main contents are summarized and the future work is prospected.