This paper presents several neural network based modelling, reliable optimal control, and iterative learning control methods for batch processes. In order to overcome the lack of robustness of a single neural network,...This paper presents several neural network based modelling, reliable optimal control, and iterative learning control methods for batch processes. In order to overcome the lack of robustness of a single neural network, bootstrap aggregated neural networks are used to build reliable data based empirical models. Apart from improving the model generalisation capability, a bootstrap aggregated neural network can also provide model prediction confidence bounds. A reliable optimal control method by incorporating model prediction confidence bounds into the optimisation objective function is presented. A neural network based iterative learning control strategy is presented to overcome the problem due to unknown disturbances and model-plant mismatches. The proposed methods are demonstrated on a simulated batch polymerisation process.展开更多
The intense competition in the current marketplace ha s forced firms to reexamine their methods of doing business, using superior manu facturing practices in the form of just-in-time (JIT), production with JIT pra cti...The intense competition in the current marketplace ha s forced firms to reexamine their methods of doing business, using superior manu facturing practices in the form of just-in-time (JIT), production with JIT pra ctices pursue completion on time and zero inventory, which is often instruct ed according to the custom’s demand or the sale contract. Earliness and tardine ss are undesirable because both of them will bring the extra cost, cost will als o be increased by some factors such as operation condition, intermediate storage , clean method, etc, to minimize the total cost is often the main scheduling objective, but sometime it is most important for factories to eliminate the tar diness cost in order to maintain the commercial credit and to avoid penalty, the refore, minimum of tardiness cost becomes the first objective. It is more import ant to select a reasonable objective by the actual condition during scheduli ng. In this paper scheduling problem of chemical batch process with due date is studied, two different intermediate storage policies and two different productio n modes are also discussed, production scheduling with different intermediate st orage policy and different production mode is proposed and the result is compare d. In order to complete all products within the due date, not only earliness and tardiness but also holding problem is considered, the objective is to selec t a proper intermediate storage policy and production mode and to minimize the c ost resulted by the earliness and tardiness, even the cost result by the interme diate storage. Scheduling with multiple stage and multiple machine is known as a NP-hard problem, mathematical program (MP) method, such as branch-and-bound (BAB), mixed integer linear program (MILP), etc, is often used to solve the sche duling problem. But as is well known, MP method is not good for combination opti mization, especially for large scale and complex optimal problem, whereas geneti c algorithm (GA) can overcome the MP method’s shortcoming and is fit for solvin g such scheduling problem. In this paper a modified genetic algorithm with speci al crossover operator and mutation operator is presented to solve this schedulin g problem. The results show such problem can be solved effectively with the pres ented method.展开更多
采取专利CN 102863437A的工艺,设计间歇工艺过程,将Aspen Batch Process Developer软件应用于盐酸鲁拉西酮原料药车间设计的全流程模拟和优化。间歇操作具有显著的优势:生产灵活,同一设备可生产不同产品,可根据市场需要调节生产能力及...采取专利CN 102863437A的工艺,设计间歇工艺过程,将Aspen Batch Process Developer软件应用于盐酸鲁拉西酮原料药车间设计的全流程模拟和优化。间歇操作具有显著的优势:生产灵活,同一设备可生产不同产品,可根据市场需要调节生产能力及变更产品。适合于小批量,高收益的精细化学品。过去的四十年里,使用计算机对化工连续化生产进行模拟和设计已经十分普及。制药工业与传统化工最大的区别是生产过程多采用间歇法操作。目前世界上应用于化工间歇生产的计算机软件有BATCHES、gPROMS和Aspen Batch Process Developer。本文所用版本为Aspen Tech V8.6,以年产25t盐酸鲁拉西酮原料药车间为例,对车间进行全流程模拟及优化。整个设计贯彻质量源于设计理念,运用元葱模型,将盐酸鲁拉西酮的生产工艺分为磺化、氨解、氢化、缩合、成盐、精烘包等6个模块。展开更多
This paper introduces a dynamic facilitating mechan is m for the integration of process planning and scheduling in a batch-manufacturi ng environment. This integration is essential for the optimum use of production re...This paper introduces a dynamic facilitating mechan is m for the integration of process planning and scheduling in a batch-manufacturi ng environment. This integration is essential for the optimum use of production resources and generation of realistic process plans that can be readily executed with little or no modification. In this paper, integration is modeled in two le vels, viz., process planning and scheduling, which are linked by an intelligent facilitator. The process planning module employs an optimization approach in whi ch the entire plan solution space is first generated and a search algorithm is t hen used to find the optimal plan. Based on the result of scheduling module an u nsatisfactory performance parameter is fed back to the facilitator, which then i dentifies a particular job and issues a change to its process plan solution spac e to obtain a satisfactory schedule.展开更多
基金Supported by UK EPSRC (grants GR/N13319 and GR/R 10875)
文摘This paper presents several neural network based modelling, reliable optimal control, and iterative learning control methods for batch processes. In order to overcome the lack of robustness of a single neural network, bootstrap aggregated neural networks are used to build reliable data based empirical models. Apart from improving the model generalisation capability, a bootstrap aggregated neural network can also provide model prediction confidence bounds. A reliable optimal control method by incorporating model prediction confidence bounds into the optimisation objective function is presented. A neural network based iterative learning control strategy is presented to overcome the problem due to unknown disturbances and model-plant mismatches. The proposed methods are demonstrated on a simulated batch polymerisation process.
文摘The intense competition in the current marketplace ha s forced firms to reexamine their methods of doing business, using superior manu facturing practices in the form of just-in-time (JIT), production with JIT pra ctices pursue completion on time and zero inventory, which is often instruct ed according to the custom’s demand or the sale contract. Earliness and tardine ss are undesirable because both of them will bring the extra cost, cost will als o be increased by some factors such as operation condition, intermediate storage , clean method, etc, to minimize the total cost is often the main scheduling objective, but sometime it is most important for factories to eliminate the tar diness cost in order to maintain the commercial credit and to avoid penalty, the refore, minimum of tardiness cost becomes the first objective. It is more import ant to select a reasonable objective by the actual condition during scheduli ng. In this paper scheduling problem of chemical batch process with due date is studied, two different intermediate storage policies and two different productio n modes are also discussed, production scheduling with different intermediate st orage policy and different production mode is proposed and the result is compare d. In order to complete all products within the due date, not only earliness and tardiness but also holding problem is considered, the objective is to selec t a proper intermediate storage policy and production mode and to minimize the c ost resulted by the earliness and tardiness, even the cost result by the interme diate storage. Scheduling with multiple stage and multiple machine is known as a NP-hard problem, mathematical program (MP) method, such as branch-and-bound (BAB), mixed integer linear program (MILP), etc, is often used to solve the sche duling problem. But as is well known, MP method is not good for combination opti mization, especially for large scale and complex optimal problem, whereas geneti c algorithm (GA) can overcome the MP method’s shortcoming and is fit for solvin g such scheduling problem. In this paper a modified genetic algorithm with speci al crossover operator and mutation operator is presented to solve this schedulin g problem. The results show such problem can be solved effectively with the pres ented method.
文摘采取专利CN 102863437A的工艺,设计间歇工艺过程,将Aspen Batch Process Developer软件应用于盐酸鲁拉西酮原料药车间设计的全流程模拟和优化。间歇操作具有显著的优势:生产灵活,同一设备可生产不同产品,可根据市场需要调节生产能力及变更产品。适合于小批量,高收益的精细化学品。过去的四十年里,使用计算机对化工连续化生产进行模拟和设计已经十分普及。制药工业与传统化工最大的区别是生产过程多采用间歇法操作。目前世界上应用于化工间歇生产的计算机软件有BATCHES、gPROMS和Aspen Batch Process Developer。本文所用版本为Aspen Tech V8.6,以年产25t盐酸鲁拉西酮原料药车间为例,对车间进行全流程模拟及优化。整个设计贯彻质量源于设计理念,运用元葱模型,将盐酸鲁拉西酮的生产工艺分为磺化、氨解、氢化、缩合、成盐、精烘包等6个模块。
文摘This paper introduces a dynamic facilitating mechan is m for the integration of process planning and scheduling in a batch-manufacturi ng environment. This integration is essential for the optimum use of production resources and generation of realistic process plans that can be readily executed with little or no modification. In this paper, integration is modeled in two le vels, viz., process planning and scheduling, which are linked by an intelligent facilitator. The process planning module employs an optimization approach in whi ch the entire plan solution space is first generated and a search algorithm is t hen used to find the optimal plan. Based on the result of scheduling module an u nsatisfactory performance parameter is fed back to the facilitator, which then i dentifies a particular job and issues a change to its process plan solution spac e to obtain a satisfactory schedule.
