W ith the development of information technology and economic globalization, process industries tend to be more networked and digitized. It is inadequate for the researchers to pay attention only to the energy fl ow or material fl ow in production processes. More efforts are concentrated to transform various production factors into information and integrate the information to optimize production process and the whole supply-chain (Aoyama and Naka 2003). As an important part of the supply-chain, it is signifi cant for process industries to incorporate the different process operations, ensure process industries rapidly response to market demands, while at the same time, continually striving for higher productivity, environmental regulation and low cost. Process operations include real-time simulation, on-line optimization, fault diagnosis, process monitoring, etc. In recent years, there are several software and tools to aid the process operation, which are called as the kind of computer aided process operation (CAPO). Unfortunately, these software and tools are designed for individual process operation separately, such as fault diagnosis, having not considered the process operation systems as a whole. At the same time, these software are developed with different program language and run on different operation systems or hardware platforms. These factors make the existing CAPO systems may not collaborate effi ciently.For the past years, several software and computer tools have been developed to aid the chemical process operations including real-time simulation, on-line optimization, fault diagnosis, process monitoring, and many other functions. These tools were designed separately and did not collaborate effi ciently, making it diffi cult to integrate different engineering tasks for the optimal process operation. In this paper, an agentoriented modelling approach is presented to address this problem. Elements in the process operation systems are divided into two classes. One class consists of equipment, units and processes, while the other class consists of production operation tasks. The two classes of elements are modelled as objects and agents, respectively. Then, three strategies are presented to implement the integration of the whole process operation system, which are integration of object models, integration of agent models and supervision of operator. Also presented is a case study of integration of process operation decision optimization and abnormal situation management using the proposed agent oriented approach for TE challenge problem.Ces dernières années, plusieurs outils informatiques et logiciels ont été mis au point pour faciliter les opérations des procédés chimiques, notamment la simulation en temps réel, l'optimisation en ligne, le diagnostic de défauts, la surveillance des procédés et de nombreuses autres fonctions. Ces outils étaient conçus séparément et ne coopéraient pas effi cacement, rendant diffi cile l'intégration des différentes tâches d'ingénierie en vue du fonctionnement optimal des proc...

1 C hemical processes are highly non-linear systems exhibiting complex time-dependent behavior. Modeling of these dynamics using first principle models is not always possible. As a result, alternative techniques using either data-driven or knowledge-based methods have been explored, such as statistical regression and fuzzy logic approaches. Conventional statistical regressions, however, have proven inadequate in many applications for modeling an underlying mechanism with significant non-linear characteristics, while fuzzy logic models depend heavily on expert knowledge which can be subjective. Consequently, hybrid models have recently received much attention. An interesting hybrid model proposed by Qi et al. (1999) combines first principle equations with artificial neural networks (ANN), the later being used to determine the parameters of the first principle models. The integration of fuzzy logic with principle model-based simulation was also studied by authors (Qian et al., 1999(Qian et al., , 2001) who used fuzzy logic to represent the imprecise relationships among process variables. In another effort, Baffi et al. (1999) studied the integration of polynomial spline functions with ANN.Given the large volume of data collected from computers and laboratories in modern process plants, data-driven techniques, and particularly neural networking, have become an attractive data-driven method. For modelling dynamic systems, recurrent neural networks (RNNs) need to be used (Himmelblau, 2000). While they are not suitable for high-order systems, RNNs including the Hopfield, Elman and Jordan neural networks, are effective in dealing with first order systems. To address this discrepancy, Zhu (1998) and Wu (2000) modified the Elman neural network and proposed two new approaches: a RNN with multiple subfeedback-layers, and a state-integrated RNN, respectively. The modified RNNs reserve, as much as possible, past time series information so as to facilitate the simulation of high-order dynamic non-linear behaviors.A major challenge to the use of these RNN neural networks is that for complex multiple input multiple output (MIMO) systems, the number of nodes in the input layer as well as in the hidden layer tends to be very large, resulting in training difficulties. In this contribution, a new RNN structure is proposed to address this issue, in that it integrates the past values for certain steps of the output variables with the input variables, and the original input variables are pre-processed using principal component analysis for the purpose of dimension reduction. A new methodology for modelling of dynamic process systems, the output integrated recurrent neural network (OIRNN), is presented in this paper. OIRNN can be regarded as a modified Jordan recurrent neural network, in which the past values for certain steps of the output variables are integrated with the input variables, and the original input variables are pre-processed using principal component analysis (PCA) for the purpose of dimension reduction. The main advanta...

A modular approach to the formulation and a solution of mixed‐integer non‐linear programming (MINLP) problems are presented, which reduce the size of MINLP and the computational expenses effectively. The method decomposes the synthesis task into three hierarchical levels—the superstructure, the structure, and the modules, with the layer of modules being the most critical to the problem solution. The strategy has been implemented in a simulation environment in which the variables of interest are defined as implicit functions of the optimization variables. The implicit relationships are handled using a data‐oriented process simulation technique (DOPS) that significantly simplify the simulation. The method has been effectively applied to two case studies, one from literature for the synthesis of hydrodesalkylation, and another from industrial process manufacturing methylene diphenylene diisocyanates.

No abstract

No abstract