The Engineering of Distributed Control Systems

by René Simon

The classical measurement, control and actuator devices were based on simple physical principles (mechanical, hydraulic, pneumatic, electrical). Often they were used as stand-alone devices for relatively closed automation solutions. With the introduction of microprocessor technology and its fast spreading, the focus shifted from stand-alone devices to much more complex device systems. These systems of automation devices including their necessary communication systems are called Process Control Systems (PCS). Process Control Systems provide control and supervision of production processes. They connect people (e.g. the operator) and machines. They consists of input / output devices, data processing units, human machine interfaces and communication systems. 1st generation PCSs are characterised by a centralised structure. A central device scans all relevant process data and computes the actuator values. There are two basic types of PCSs, one for process control and one for manufacturing. Their internal structures are similar, the market (user) will decide about a possible fusion. This fusion and the transition from centralised to de-centralised systems based on serial communication systems (e.g. fieldbus) are milestones of the development towards 2nd generation PCSs. This development hasn't been finished yet, but lots of solutions are emerging. A non-interrupted engineering process on the basis of common information models and data exchange technologies is the primary requirement for the design of 3rd generation PCSs. These 3rd generation PCSs are called Distributed Control Systems (DCS) here. Why becomes the engineering process so important? Traditionally, PCS are seen from a run-time point of view (function, device, system). The engineering aspect becomes more important due to its increasing complexity and costs involved. The complexity results from several factors influencing the engineering process such as different device components (input / output, data processing, HMI, communication), process physics (mechanical, electrical, ...), live cycle phases and device vendors. Today, the integration of these different worlds usually is done by building hardware and software interfaces and expensive commissioning processes based on trial-and-

error approaches. This will become impossible under a true cost of ownership principle. The whole engineering process must be supported by integrated tools in order to avoid data losses and inconsistencies. The use of paper and even of electronic means don't provide a solution as long as common transfer syntax and standardised data models are not used. The standard IEC 1131 defines hardware and software models as well as programming languages for Programmable Controllers. Unfortunately, no mechanism for data exchange between engineering (programming) tools was defined. To fill this gap, a neutral file exchange format (FXF) was specified by PLCopen. The FXF is based on a STEP (ISO 10303) model and allows to transport PLC programs from the programming environment of one vendor to another. FXF is one requirement for the Portability Level Certification by PLCopen and certainly a major step forward to open systems. IEC 1131 languages are suitable not only for PLC programming, but also as basis for the functional description of a Distributed Control System (cf. IEC 1499). The fieldbus community developed some other approaches to exchange data by electronic means. This includes Device Descriptive Languages and Device Data Base (HART, FF, PROFIBUS). As part of the Function Block activities (IEC TC65 (IEC 1499), IEC SC65C WG7) the topic is also treated. These different solutions begin to overlap. The international standard for the exchange of product data ISO 10303 (STEP) offers the only comprehensive solution (data description languages (EXPRESS), mapping to implementation technologies, tools, Application Protocols). Although developed for product data it is quite suitable to solve engineering problems of DCS as several projects show. The integration of all relevant data (functional, material, power supply, ...) becomes possible. However, some problems are still open (e.g. how to integrate modern implementation technologies such as OLE, CORBA, Internet?). Fieldbus wars prevented true interoperability between devices of different vendors, even today. Who wins, who loses? In PLC programming, only one standard emerged from many propriety solutions: IEC 1131. This is a real success of PLCopen and generates profits for all: technology providers, device vendors and end users. Why not continue this successful story? PLCopen has a good chance to become a leader of this new progressive development: know-how, technology and market penetration. It is necessary to act now, or other are doing it. Users, ask for it!

Sign up to vote on this title
UsefulNot useful

Master Your Semester with Scribd & The New York Times

Special offer: Get 4 months of Scribd and The New York Times for just $1.87 per week!

Master Your Semester with a Special Offer from Scribd & The New York Times