天津大学学报
天津大學學報
천진대학학보
JOURNAL OF TIANJIN UNIVERSITY SCIENCE AND TECHNOLOGY
2014年
12期
1057-1064
,共8页
隔壁精馏塔%简捷设计%严格模拟%响应面法
隔壁精餾塔%簡捷設計%嚴格模擬%響應麵法
격벽정류탑%간첩설계%엄격모의%향응면법
divided wall column%short-cut design%rigorous simulation%response surface methodology(RSM)
针对隔壁精馏塔节能工艺,提供了一套完整的设计优化方法。首先基于 Fenske-Underwood-Gilliland-Kirkbride 方程建立了完整的简捷设计方法,得到了隔壁精馏塔塔实际理论板数、适宜的进料位置、侧线采出位置及回流比等参数。然后在简捷计算的基础之上,选用 Multifrac 模型对隔壁塔进行了严格计算模拟,同时利用 Aspen Plus 进行单因素优化分析得到最优设计参数。最后利用响应面优化法(RSM)中的箱线图设计(BBD)方法对隔壁精馏塔设计参数进行了实验设计,在验证模型有效的基础上运用Design-Expert软件进行数据处理,预测出了最优设计参数,并将预测值进行实验验证,将验证结果与单因素优化结果进行对比,结果表明响应面优化法得到的最优设计参数使隔壁塔的能耗较低、纯度较高。
針對隔壁精餾塔節能工藝,提供瞭一套完整的設計優化方法。首先基于 Fenske-Underwood-Gilliland-Kirkbride 方程建立瞭完整的簡捷設計方法,得到瞭隔壁精餾塔塔實際理論闆數、適宜的進料位置、側線採齣位置及迴流比等參數。然後在簡捷計算的基礎之上,選用 Multifrac 模型對隔壁塔進行瞭嚴格計算模擬,同時利用 Aspen Plus 進行單因素優化分析得到最優設計參數。最後利用響應麵優化法(RSM)中的箱線圖設計(BBD)方法對隔壁精餾塔設計參數進行瞭實驗設計,在驗證模型有效的基礎上運用Design-Expert軟件進行數據處理,預測齣瞭最優設計參數,併將預測值進行實驗驗證,將驗證結果與單因素優化結果進行對比,結果錶明響應麵優化法得到的最優設計參數使隔壁塔的能耗較低、純度較高。
침대격벽정류탑절능공예,제공료일투완정적설계우화방법。수선기우 Fenske-Underwood-Gilliland-Kirkbride 방정건립료완정적간첩설계방법,득도료격벽정류탑탑실제이론판수、괄의적진료위치、측선채출위치급회류비등삼수。연후재간첩계산적기출지상,선용 Multifrac 모형대격벽탑진행료엄격계산모의,동시이용 Aspen Plus 진행단인소우화분석득도최우설계삼수。최후이용향응면우화법(RSM)중적상선도설계(BBD)방법대격벽정류탑설계삼수진행료실험설계,재험증모형유효적기출상운용Design-Expert연건진행수거처리,예측출료최우설계삼수,병장예측치진행실험험증,장험증결과여단인소우화결과진행대비,결과표명향응면우화법득도적최우설계삼수사격벽탑적능모교저、순도교고。
A set of comprehensive methods was proposed for the design and optimization of divided wall col-umn(DWC). A short-cut design method based on Fenske-Underwood-Gilliland-Kirkbride equations for DWC was used to get initial values of design parameters of theoretical stages,feed stage,side-product stage,reflux ratio and so on. With the initial values of all parameters from short-cut design,rigorous simulation of DWC was carried out using Multifrac model. The optimization result was obtained through single-factor experiment using Aspen Plus. In the last stage,Box-Behnken design(BBD)under response surface methodology(RSM)was used for the optimization of DWC and to evaluate the effects of parameters and their interactions on energy efficiency and product purity. Design-Expert software was used to tackle experiment data and predict optimization result based on significant model. Com-paring the optimization result of single-factor experiment and RSM,we found that RSM could render more optimized result in respect of energy saving and high purity.