以年总操作费用(TAC)、CO_(2)排放量(GEC)和精馏塔热力学效率(η)为目标,提出了基于粒子群算法(PSO)的优化方法,并将该方法应用于三氯氢硅(SiHCl_(3),TCS)歧化制取硅烷(SiH_(4),MS)反应精馏塔的优化设计。在Aspen Plus V7中建立流程进...以年总操作费用(TAC)、CO_(2)排放量(GEC)和精馏塔热力学效率(η)为目标,提出了基于粒子群算法(PSO)的优化方法,并将该方法应用于三氯氢硅(SiHCl_(3),TCS)歧化制取硅烷(SiH_(4),MS)反应精馏塔的优化设计。在Aspen Plus V7中建立流程进行模拟,使用平衡级模型,对RD-2IC(带有2个中间冷凝器的反应精馏塔)和高压分离塔的双塔构型建立稳态模型,考察了塔压、塔板数、回流比、进料位置、反应段持液量和中间冷凝器气相分率等影响因素,初步确定了各参数的最优值,为进一步深度优化提供了初值和可行域。结果表明,与单因素灵敏度分析结果相比,PSO算法优化后的TAC节省了54.50%、GEC减少了38.13%、η提高了22.55%。展开更多
The shift scheduling system of the transmission has an important effect on the dynamic and economic performance of hybrid vehicles. In this work, shift scheduling strategies are developed for parallel hybrid construct...The shift scheduling system of the transmission has an important effect on the dynamic and economic performance of hybrid vehicles. In this work, shift scheduling strategies are developed for parallel hybrid construction vehicles. The effect of power distribution and direction on shift characteristics of the parallel hybrid vehicle with operating loads is evaluated, which must be considered for optimal shift control. A power distribution factor is defined to accurately describe the power distribution and direction in various parallel hybrid systems. This paper proposes a Levenberg-Marquardt algorithm optimized neural network shift scheduling strategy. The methodology contains two objective functions, it is a dynamic combination of a dynamic shift schedule for optimal vehicle acceleration, and an energy-efficient shift schedule for optimal powertrain efficiency. The study is performed on a test bench under typical operating conditions of a wheel loader. The experimental results show that the proposed strategies offer effective and competitive shift performance.展开更多
文摘以年总操作费用(TAC)、CO_(2)排放量(GEC)和精馏塔热力学效率(η)为目标,提出了基于粒子群算法(PSO)的优化方法,并将该方法应用于三氯氢硅(SiHCl_(3),TCS)歧化制取硅烷(SiH_(4),MS)反应精馏塔的优化设计。在Aspen Plus V7中建立流程进行模拟,使用平衡级模型,对RD-2IC(带有2个中间冷凝器的反应精馏塔)和高压分离塔的双塔构型建立稳态模型,考察了塔压、塔板数、回流比、进料位置、反应段持液量和中间冷凝器气相分率等影响因素,初步确定了各参数的最优值,为进一步深度优化提供了初值和可行域。结果表明,与单因素灵敏度分析结果相比,PSO算法优化后的TAC节省了54.50%、GEC减少了38.13%、η提高了22.55%。
基金Project(51805200)supported by the National Natural Science Foundation of ChinaProject(20170520096JH)supported by the Science and Technology Development Plan of Jilin Province,ChinaProject(2016YFC0802900)supported by the National Key R&D Program of China
文摘The shift scheduling system of the transmission has an important effect on the dynamic and economic performance of hybrid vehicles. In this work, shift scheduling strategies are developed for parallel hybrid construction vehicles. The effect of power distribution and direction on shift characteristics of the parallel hybrid vehicle with operating loads is evaluated, which must be considered for optimal shift control. A power distribution factor is defined to accurately describe the power distribution and direction in various parallel hybrid systems. This paper proposes a Levenberg-Marquardt algorithm optimized neural network shift scheduling strategy. The methodology contains two objective functions, it is a dynamic combination of a dynamic shift schedule for optimal vehicle acceleration, and an energy-efficient shift schedule for optimal powertrain efficiency. The study is performed on a test bench under typical operating conditions of a wheel loader. The experimental results show that the proposed strategies offer effective and competitive shift performance.
