Research Of Working-status Recognition For The Grate Cooler In The Cement Production Process

The cooling process of grate cooler for cement clinker is a very important part in thecement production. The grate cooler can reduce the temperature of clinker rapidly, improveclinker quality and grinding ability, and recycle heat. Working-status of the grate cooler notonly directly affect the quality, production of cement clinker, and the kiln materialcharacteristics,but also play an important role in guiding the real-time control of the gratecooler. The cooling process of grate cooler is a complex dynamic process, it has thecharacteristics of nonlinear, multivariable, strong coupling, big time lag, time-variable. Underthe influence of the raw material composition, calcining conditions and other factors, thefluctuation of working-status will be great, which causes the working-status of the grate coolersystem to be extremely complex. What’s the worse, the operation of the entire cooling systembecomes more difficult when the abnormal conditions occur.In the process of cement clinker production, the control of grate cooler is still rested onsemi-automatic state compared to the automatic control for calciner and preheater linkspresently. In the operation of the grate cooler, operators know the current condition based onthe change of technological parameters. Some parameters will be changed to ensure thestability of equipment. However, this operation can’t ensure the optimal working condition ofthe grate cooler when the clinker with different states and accounts enter owning to thesubjective factors. Therefore, how to adopt an effective method for grate cooler conditionsrecognition and realize grate cooler automatic control under the guidance of workingcondition is becoming a big concern by many experts and scholars.After analysis of grate cooler cooling mechanism and field data, combining with theoperator’s experience, the characteristics of typical working conditions are summarized.Therefore, we select working identification parameters and a solution of working-statusintellective recognition based on expert system proposed and put into practice.Based on the analysis of different working conditions’ characteristics andinterrelationship, combined with correlation analysis, eleven key parameters are chosen andfuzzified. In this dissertation, the adjustable fuzzy reasoning mechanism is used to realizeautomatic modification for fuzzy inference rules on-line and improve the accuracy of the fuzzy inference. The grey comprehensive correlation analysis method is applied to distinguishthe abnormal conditions which seriously influence the grate cooler control. Using the realtime values and the trend of the values makes the results reliable and effective.The realization of expert system for working-status recognization uses a series of toolsand techniques, such as VB, SQL Server, OPC. A relational database is established for expertsystem knowledge base, and it is easy to realize addition, modification and deletion function.The tables in the database can be connected through numbers or terms, which is convenientfor maintenance. Operating system based on expert system is convenient to operate andvarious working conditions can be known directly. The verification results in the fielddemonstrate the effectiveness and practicality of this solution.