Understanding Batch Chemical Processes

Modelling and Case Studies

Thokozani Majozi, Esmael R. Seid, Jui-Yuan Lee

• Publisher: CRC Press
• Year: 2017
• Print ISBN: 9780367878634
• eBook ISBN: 9781351679060
• Topic: Biological Sciences
• Subtopic: Biotechnology
• Index: Biological Sciences

Batch chemical processes, so often employed in the pharmaceutical and agrochemical fields, differ significantly from standard continuous operations in the emphasis upon time as a critical factor in their synthesis and design.

With this inclusive guide to batch chemical processes, the author introduces the reader to key aspects in mathematical modeling of batch processes and presents techniques to overcome the computational complexity in order to yield models that are solvable in near real-time. This book demonstrates how batch processes can be analyzed, synthesized, and designed optimally using proven mathematical formulations. The text effectively demonstrates how water and energy aspects can be incorporated within the scheduling framework that seeks to capture the essence of time. It presents real-life case studies where mathematical modeling of batch plants has been successfully applied.

1
Introduction to Batch Processes

1.1INTRODUCTION

Batch processes differ from continuous processes in several ways. The main difference is that time is inherent in batch processes. In batch processes, every task has a definite duration with starting and finishing times, whereas in continuous processes, time is important during non-steady-state operation. As a result, scheduling of batch processes is vital to the operation of any batch facility. Furthermore, in batch plants, detailed requirements for the various products may be specified on a day-to-day basis. A production schedule must indicate the sequence and manner in which the products are to be produced and specify the times at which the process operations are to be carried out. It is clear that the overall productivity and economic effectiveness of batch plants depend critically on the production schedule as it harmonizes the entire plant operation to attain production goals. While flexibility of batch plants improves productivity, it also makes plant scheduling a challenging task. Much research has focused on developing optimization techniques for scheduling batch plants with the aim of reducing the CPU time required to attain the optimal objective value.

This chapter provides a detailed literature review on various scheduling techniques. The review covers work which has been done in the field of mathematical models used for scheduling of batch plants. Papers presented in the last two decades are considered in the review, with focus on models based on unit-specific event-point continuous-time representation. The chapter is systematically divided into six major sections. Section 1.1 discusses recipe, State Task Network (STN) and State Sequence Network (SSN) representations for batch plants. One of the major characteristics that differentiates batch plants from continuous plants is the existence of intermediate storage in batch plants in order to separate operations in multipurpose equipment and to free the equipment for subsequent processes. The different intermediate storage operational philosophies that exist in batch plants are discussed in Section 1.2. Section 1.3 discusses the different types of batch plants under the main division of multiproduct and multipurpose batch plants.

Section 1.4 discusses the different types of models developed for scheduling of multiproduct batch plants. In this section, the models are categorized under the main division of graphical technique and mathematical technique. In the mathematical section, the different models are grouped into models based on sequence precedence and slot time representations. A brief of scheduling techniques for continuous and semi-continuous plants is given in Section 1.5. Section 1.6 details scheduling techniques developed for multipurpose batch plants. Finally, conclusions and limitations of the current scheduling techniques are briefly discussed in Section 1.7.

1.2BATCH PROCESS REPRESENTATION

Batch plant processes were first represented in terms of ‘recipe networks’, as used by Reklaitis (1991). This is analogous to the flow sheet representation of continuous plants, but proposed to describe the process itself rather than a specific plant. Each node on a recipe network corresponds to a task, with directed arcs between nodes representing task precedence. Although recipe networks are certainly adequate for several processing structures, they often involve ambiguities when applied to more complex ones. Kondili et al. (1993) showed that a recipe network was not enough to represent batch processes without vagueness, as illustrated in Figure 1.1. It is not clear from this representation whether task 1 produces two different products later used as inputs of tasks 2 and 3, respectively, or whether it produces one type of product which is then shared between unit operations 2 and 3. Similarly, it is also impossible to determine from Figure 1.2 whether task 4 requires two different types of feedstock, respectively produced by tasks 2 and 5, or whether it only needs one type of feedstock which can be produced by either 2 or 5. Both interpretations are equally plausible. The former could be the case if, say, task 4 is a catalytic reaction requiring a main feedstock produced by task 2 and a catalyst which is then recovered from the reaction products by the separation task 5. The latter case could arise if task 4 were an ordinary reaction task with a single feedstock produced by 2, with task 5 separating the product from the unreacted material which is then recycled to 4.

FIGURE 1.1 Recipe representation of chemical processes.

FIGURE 1.2 STN representations for recipe representation.

In order to remove these ambiguities in a systematic fashion, Kondili et al. (1993) proposed a new representation for chemical processes called State Task Network representation. The distinctive characteristic of the STN was that it had two types of nodes: namely, the state nodes, representing the feeds, intermediate and final products; and the task nodes, representing the processing operations which transform materials from one or more input states to one or more output states. State and task nodes were denoted by circles and rectangles, respectively. State Task Networks are free from the ambiguities associated with recipe networks. Figure 1.2 shows two different STN representations, both of which correspond to the recipe network of Figure 1.1. The STN representation is equally suitable for networks of all types of processing tasks, such as continuous, semi-continuous and batch.

Majozi and Zhu (2001) introduced a new concept called the State Sequence Network representation for scheduling of batch plants. The SSN was a graphical network representation of all the states that exist in batch plants, and the network was formulated based on the production recipe. A state changes from one state to another state when it undergoes a unit operation such as mixing, separation or reaction. The building blocks of the SSN are shown in Figure 1.3, Figure 1.4 and Figure 1.5. From these building blocks, it is easy to construct a SSN for any process recipe. The SSN representation was developed by realizing that (1) the capacity of a unit in which a particular state is used sets an upper limit on the amount of state used or produced by the corresponding task, (2) the presence of a particular state in an operation corresponds to the existence of a corresponding task and (3) the usage of state s corresponds to the production of another state s′.

FIGURE 1.3 Simple unit operation.

FIGURE 1.4 Unit operation with mixing/reaction.

FIGURE 1.5 Unit operation with splitting.

1.3DIFFERENT STORAGE OPERATIONAL POLICIES IN BATCH PLANT

According to Reklaitis (1982), the multistage nature of a batch processing network allows several different storage and waiting options:

• Unlimited intermediate storage (UIS)
• Finite intermediate storage (FIS)
• No intermediate storage (NIS)
• Central intermediate storage (CIS)
• Unlimited wait (UW)
• Finite wait (FW)
• Zero wait (ZW)

The ZW or FW policy is used where unstable intermediate material must be processed immediately or within a short time after the previous step has been completed. In the NIS policy, there is no intermediate storage between process units, but a product can be stored in a process unit before moving to the next available processing unit. A more practical operational storage policy is that of FIS, where there is limited storage capacity between process units. The UIS policy is more of a theoretical one. If the intermediate products are compatible, the CIS policy is recommended. After materials are processed in a unit...

Table of contents

• Cover
• HalfTitle Page
• Title Page
• Copyright Page
• Foreword
• Acknowledgements
• Authors
• Chapter 1 Introduction to Batch Processes
• Chapter 2 Modelling for Effective Solutions: Reduction of Binary Variables
• Chapter 3 Methods to Reduce Computational Time: Prediction of Time Points
• Chapter 4 Integration of Scheduling and Heat Integration: Minimization of Energy Requirements
• Chapter 5 Heat Integration in Multipurpose Batch Plants
• Chapter 6 Design and Synthesis of Heat-Integrated Batch Plants Using an Effective Technique
• Chapter 7 Simultaneous Scheduling and Water Optimization: Reduction of Effluent in Batch Facilities
• Chapter 8 Optimization of Energy and Water Use in Multipurpose Batch Plants Using an Improved Mathematical Formulation
• Chapter 9 Targeting for Long-Term Time Horizons: Water Optimization
• Chapter 10 Long-Term Heat Integration in Multipurpose Batch Plants Using Heat Storage
• Index