Profit Maximization Techniques for Operating Chemical Plants. Sandip K. Lahiri

Чтение книги онлайн.

СКАЧАТЬ are initially discussed to enlighten the reader about the scope and huge potential of optimization in the process industry. This chapter introduces new advanced Meta heuristic optimization techniques that can be applied where application of a conventional technique is limited due to the complexity of the industrial context. Different Meta heuristic optimization techniques, like the genetic algorithm (GA), differential evolution (DE), simulated annealing (SA), etc., are described in detail in this chapter. A basic algorithm, step‐by‐step procedure to develop an optimization technique and different uses of GA, DE, and SA in various fields of process optimization are explained here in order to develop an understanding of this new area. A case study in reactor optimization is illustrated to explain the advantage and ease of implementation of Meta heuristic methods over conventional methods.

      Process Monitoring

      Today's complex chemical plants need advanced monitoring and control systems to quickly identify the suboptimal operation of process equipment and implement a quick optimization strategy. Running the plant at the highest possible capacity for profit maximization necessitates the development of an intelligent real‐time monitoring system. However, due to the large amount of process data, it is a herculean task to monitor each and every piece of process data. Chapter 9 enlightens the readers about an online intelligent monitoring system, KPI‐based process monitoring, a cause and effect‐based monitoring system, etc. It also gives an idea regarding the development of a potential opportunity‐based dashboard, loss and waste monitoring systems, a cost‐based monitoring system, a constraints‐based monitoring system, and how all these can be integrated into business intelligent dashboards.

      In this chapter, a new advanced computational technique, namely principal component analysis (PCA), is discussed to visualize data. The advantage of such an online monitoring system is to visualize the plant condition from a higher level but with a lower dimension space. A step by step procedure to build a PCA‐based advance monitoring system is explained in detail, with examples and industrial case studies.

      Fault Diagnosis

      Chemical industries recently discovered that a large amount of profit becomes eroded due to unplanned shutdowns of the plant. Due to spurious trips of equipment much potential profit is lost. One major ingredients of profit maximization is to increase plant reliability and running hours. Plant shutdown can be avoided by building a robust fault diagnosis system that will detect and alert the operator about any potential event that can lead to plant disturbance and eventually plant shutdown before it starts happening. How a robust fault diagnosis system can be made by PCA and ANN that can be implemented in industry is discussed in detail in this chapter with industrial case studies. Different aspects of enhancement of plant reliability by an advance monitoring and fault diagnosis system is the main focus of the chapter.

      Optimization of the Existing Distillation Column

Often distillation columns cause a bottleneck to increase plant capacity. It is very important to understand the operation and capacity limits of distillation columns in commercial plants. Chapter 11 enlightens the reader about how to evaluate a feasible operating window by using a capacity diagram. Calculations based on the capacity diagram and the effect of different design and operating variables on the capacity diagram are explained in detail with example calculations. This chapter enlightens the reader about operating profile assessment, tower rating assessment, tower efficiency assessment, and hydraulic performance evaluations of running distillation columns. It also provides practical guidelines regarding what to look for in distillation column optimization in an industrial context and explains the whole concept with real‐life case studies.

      New Design Methodology

      Due to intense competition among chemical industries across the globe, it is now absolutely necessary to minimize the cost of equipment during the design phase. Equipment costs consist of the initial capital cost of the equipment and the operating costs of the equipment. Due to the availability of a faster computer, it is now feasible to design one million different design configurations for any equipment. It is important to choose the lowest cost equipment among those one million options, but one that also obeys all of the constraints of operation, safety, maintainability, etc. Hence, to survive in today's cut‐throat competition, it is necessary to put the minimization of equipment cost as the main design target and an optimization algorithm is required to search all feasible design configurations to arrive at a minimum cost design quickly. This gives rise to a new design methodology of process equipment. Earlier traditional design methodology, where cost is not considered as a design target during the design phase, no longer produces a competitive design. In this chapter, a new design methodology of a plate‐type distillation column is considered as a case study to show the essence of the new design methodology. This chapter evolves a strategy to optimize various tray geometric parameters, like tray diameter, hole diameter, fractional whole area, downcomer width, etc., and also decides on the optimum feed tray location based on the overall cost minimization concept by particle swarm optimization techniques.

      Genetic Programing for Modeling of Industrial Reactors

      Industrial reactors are the most potential candidates used to increase profit, yet they are the most neglected in the optimization project in industry. This is due to fear of process engineers to change reaction parameters beyond their usual boundaries because of poor knowledge of reaction kinetics. Conventional methods for evaluating complex industrial reaction kinetics have their own limitations. Chapter 13 introduces a completely new advanced computational technique, namely geometric programing (GP), to model industrial reaction kinetics. Being a new computational technique, the main advantage of GP is that process engineers do not have to assume any form of kinetic equation beforehand; it will be generated on its own from available industrial reactor data. The theoretical basis of GP with its various features, an algorithm of GP, and different case studies are discussed in detail to enlighten the reader about this new technique. How a generated kinetic model can be used online and offline to increase profit from an industrial reactor is described in detail through case studies.

      Maximum Capacity Test Run and Debottlenecking Study

All over the world, chemical plants are running 100–140% of their installed nameplate capacity. Profit maximization by capacity enhancement is a very common route in process industries. This high capacity running more than their design capacity is possible due to inherent safety margins available in process equipment. This chapter helps readers to understanding different safety margins available in process equipment and explains the strategy to exploit those margins. Chapter 14 explains in a systematic way how to increase plant capacity without affecting safety and reliability. Plant real bottlenecks and the potential opportunity to increase capacity can be evaluated by actually performing a demo test run of the plant with the highest possible capacity for one week, which is commonly known as a maximum capacity test run. Fourteen key steps to carry out a maximum capacity test run in commercial running plants is the key focus area of this chapter. All the steps of the maximum capacity test run are explained with real‐life examples and industrial case studies. The next step of the maximum capacity test run is to find real bottlenecks of the plant by a debottlenecking (DBN) study. How to carry out a DBN study from the existing data and its different aspects are discussed in this chapter.

      Loss Assessment

      One of the strategies of profit maximization is to reduce the wastage of resources. In other words, this essentially СКАЧАТЬ