THE UNIVERSITY of BIRMINGHAM On-line Condition and Safety Monitoring of Pulverised Coal Mills Using a Model Based Pattern Recognition Technique

Around 40% of electricity in the UK is currently generated by coal-fired power stations. Coal-fired power stations nowadays are required to operate more flexibly with more varied coal specifications and regularly use coal with higher volatile contents and biomass materials; this can increase the risks of pressurisation or fires in milling plants. The power stations are also obliged to vary their outputs in response to the changes of electricity market prices, which results in more frequent mill start-ups and shutdowns. Frequent start-ups and shutdowns of mills will also have an impact on power plant operation safety. Although the increased risks are currently being mitigated by R&D work and the implementation of increased operational controls, in many cases, coal mills are shutdown and then restarted before they have cooled adequately, which creates a potential fire hazard within the mill. Mill fires could occur if the coal stops flowing in the mill and the static deposit is heated for a period of time. The UK PF Safety Forum had recently reported an increase in the frequency of mill incidents in the UK. However, it is difficult to identify if there will be a potential fire in the mill before it happens. Mill gas outlet temperature and CO measurement are established methods of detecting fires in mills, but at present they are not very effective for detecting small fires. The CO detection system becomes ineffective when the mill is in service due to dilution effects caused by primary air flow and associated oxygen content in the mill. Advanced distributed control and monitoring systems have been installed at almost all coal-fired power plants, which enable power plant operators to collect data from the major plant components, e.g. mills, boilers, turbines and generators. Therefore, the project is to develop a software package for condition monitoring and fire prediction through on-line implementation of a coal mill model using evolutionary computation and pattern recognition techniques. The aim of the project is to develop a software package for on-line mill condition and safety monitoring with particular emphasis on the following objectives: • to improve the multi-segment mill mathematical model using evolutionary computation techniques based on the on-site measurement data and to extend the current model from E-Type vertical spindle mills to other types of coal mills such as tube-ball mills and etc; • to identify coal quality variations through recognising the variation patterns of mill model parameters and …