Community Research and Development Information Service - CORDIS


CHEM Report Summary

Project ID: G1RD-CT-2001-00466
Funded under: FP5-GROWTH
Country: United Kingdom

Chem application to a blast furnace

The aim of the CHEM project was to develop and implement advanced Decision Support Systems (DSS) for process monitoring, data and event analysis, and operation support in industrial processes. The systems are synergistic integration of innovative software tools, which improve the safety, product quality and operation reliability as well as reduce the economic losses due to faulty states, mainly in refining, chemical and petrochemical processes.

The CHEM applications consist of integrated sets of software toolboxes that provide robust detection and diagnosis of process problems in real-time. The systems assist operators in assessing process status and responding to abnormal events. The project provides a flexible architecture and a methodology in order to facilitate the development of such applications on many processes.

An application has been developed to predict process instability on a blast furnace based on wall pressure trends and gas analysis. The objective is to give sufficient warning to enable remedial action to be taken to reduce the extent of the instability. Several combinations of toolboxes have been used.

Combinations of Toolboxes:
- Toolbox "Performance and quality management based on MSPC" (iMPSC) is used alone to combine the signals described above. The bi-variate trend is assessed using contribution analysis so that if the action limit is exceeded, an alarm message is generated when a certain number of parameters have been identified as experiencing the greatest change.

- Toolbox "Qualitative representation of process trends" (Qualtrend) with "Qualitative situation assessment" (Salsa) to generate episodes using 4 individual signals from the process. The 4 sets of episodes are classified to either normal, pre-fault or fault.

- Toolbox "Performance and quality management based on MSPC" (iMSPC)is used with "Qualitative representation of process trends" (Qualtras). Episodes are generated from the first two principal components. Heuristic rules have been developed to analyse the sequence of episodes generated from both principal components. An alarm is generated when certain sequences of episodes are detected. These rules are written in G2.

- Toolbox "Performance and quality management based on MSPC" (iMPSC) is used with "Qualitative situation assessment" (Salsa). Salsa continually classifies the combination of outputs from "Performance and quality management based on MSPC".

It was not found to be worthwhile to classify episodes generated from the outputs of "Performance and quality management based on MSPC".

It was found that from the 22 days test period analysed almost all of the major unstable events could be predicted using either "Performance and quality management based on MSPC" alone (a) or "Performance and quality management based on MSPC" with "Qualitative representation of process trends" plus G2 rules (c).

Application of a simple enabler, based on heat flux trend(heat loss through the furnace wall to the cooling water), was required to remove false alarms generated by "Performance and quality management based on MSPC", "Qualitative representation of process trends" plus rules. This did not remove any of the valid alarms.

On line evaluation of the packages which include Salsa is continuing. The off line version suggests that the classification of episodes produced by "Qualitative representation of process trends" from raw signals is more robust than the classification of outputs from "Performance and quality management based on MSPC" in that fewer false alarms are generated. However, not every event was predicted.

The anticipated benefits, once the advice from these techniques has been released to operators, is to give better advice on when the production rate needs to be slightly reduced. This will reduce the extent of instability and so cause less stress to the furnace.

The package as tuned is applicable only to this particular plant, but it could be adapted to other blast furnaces which have appropriate stack pressure measurement available.

Reported by

Redcar Blast Furnace
Redcar Blast Furnace, Corus Teesside Works
TS10 5QW Redkar
United Kingdom
Follow us on: RSS Facebook Twitter YouTube Managed by the EU Publications Office Top