Predictive Process Control

Technical Applications


Predictive process control involves the ability to monitor and control a continuous materials process in real time. This allows the conditions of the process to be adjusted quickly and responsively, and avoids the delay associated with only monitoring the final product. The potential of this technology sub-area is great, as it can improve the yields and productivity of a wide range of industrial processes. It can also contribute to the reduction in unwanted or polluting side processes.


Advancing the state of the art in predictive process control requires advances in sensor capability, in data communications and data processing, and in modeling. Improved interfaces with operators, usually via graphic displays, will also provide improved control system performance. The most important class of sensors for this sub-area is non-imaging sensors which can be used to measure a vast range of phenomenology such as temperature, pressure, humidity, radiation, voltage, current, or presence of a particular chemical or biological material. In addition to passive sensors, there are active sensors based usually on lasers. Specialized microsensors can be used to detect particular chemical or biological agents. The information generated by the sensors must be combined and processed using data processing and models specific to the process being monitored.


The United States is a major player in all of the technologies which make up predictive process control. For example, historically Honeywell has had a major presence, having introduced the first distributed control system (the Honeywell TDC 2000) in 1975. Honeywell continues to be a leader, advancing the state of the art with the introduction of the TDC 3000, which incorporates protocols to address modeling errors. Honeywell smart transmitters are available to provide data to the control system. Other U.S. players include Rosemount, Foxboro, Digital, Setpoint, DMCC, and Gensym. Many other countries are also players in this area, however. In the UK, BNFL has developed the Promass advanced control system, and Predictive Control has developed an advanced controls package called Connoisseur. In Germany, Siemens Industrial Automation, AEG, and Lockner Moeller have been leaders in designing control systems with open architecture. The Japanese company, Yokogawa, is active in the International Fieldbus Consortium. Little information about Japanese work in this field is available in open English- language literature.