Industrial sensor and actuator busses

Abstract

In recent years, industrial automation and control systems have been preferred in the implementation of Distributed Control Systems (DCS) instead of centralised systems, because of their advantage of great flexibility over the whole operating range. Other benefits include low implementation cost, easy maintenance, configurability, scalability, modularity, and extendibility. Conventional centralised control is characterised by a central processing unit that communicates with the field devices (sensors, actuators, switches, valves, drives etc) with separate (parallel link) individual point-to-point links. On the other hand, DCS interconnects devices with a single serial link, as illustrated in Fig. 18.1. Since I/O (Input/Output) points within the industrial control systems are distributed, and the number of auxiliary components and equipments are progressively increasing, DCS architecture is seen to be appropriate. Each I/O point can be defined in terms of smart device. Potential operational benefits of adopting DCS schemes can be summarised. These include: Sharing of the processing load to avoid the bottleneck of a single centralised controller Replacement of complex point-to-point wiring harnesses with control networks to reduce weight and assembly costs Freedom to vary the number and type of control nodes on a particular application in order to easily modify its functionality Ability to individually configure and test segments (components) of the system before they are combined Build and test each intelligent sub-unit separately Provisions for interfacing for data exchange between the run-time control system and other factory/business systems (e.g., management information, remote monitoring and control, etc.) DCS can be leveled into four layers of automation services. The bottom layer called the component level, including the physical components such as intelligent devices (e.g., PC, industrial PC, PLC, microprocessors, micro-controllers, etc.), and non-intelligent devices such as sensors, actuators, switches, A/D, D/A, ports, transceivers, communication media, etc. The interface layer is similar to a MAC sub-layer of the link layer protocol. The process layer includes application layer features. Since control systems do not transfer raw data through a physical media an application layer has to exist. The application layer defines the variables, which are responsible for transferring data from one place to other when they are logically connected. Additionally, the management layer generates object code from the source code by the use of resources and methods such as compilers and OLE/DDE (Object Linking and Embedding/Dynamic Data Exchange), respectively. The management layer manages the control design. There is a set of generic functions within the management layers, which are accountable for providing services for all aspects of control management. Typical management functions are installation, configuration, setting, resetting, monitoring, operator interfacing, testing, and so on. © Springer-Verlag Berlin Heidelberg 2007.