Research On Fault Diagnosis Algorithm Of Cement Rotary Kiln Based On Bayesian Network

Cement rotary kiln is one of the core equipment of the NSP cement production line and its running status directly affects the output and quality of cement clinker, energy wastage and environmental pollution. Currently, the fault diagnosis for cement kiln mostly relies on the experience of engineers, artificial diagnosis has been difficult to complete the fault diagnosis efficiently and accurately. Therefore, on the basis of analysis calcination process and main process parameters of the rotary kiln, this paper focuses on studying cement rotary kiln fault diagnosis algorithm, establishes Bayesian fault diagnosis network and Bayesian fault prediction network of the cement rotary kiln. Specific research work is shown as follows:First of all, aiming at the problem that the parameter learning of Bayesian network from incomplete data set is inaccuracy, this paper presents an improved genetic algorithm for learning Bayesian network parameters-GSA algorithm.Secondly, the variables of Bayesian fault diagnosis network of the cement rotary kiln is determined by analyzing the expert knowledge and calcination process of rotary kiln. Because the data of rotary kiln is difficult to collect and often lost, so SEM algorithm is used to learn the structure of the network under the condition of incomplete data and GSA algorithm is used to learn the parameters of the network under the condition of incomplete data, then apply the trained fault diagnosis network to diagnose and analyze the fault of rotary kiln.Finally, a research and discuss has been conducted towards fault prediction of cement rotary kiln based on Bayesian network, the basic process of the fault prediction of cement rotary kiln based on Bayesian network has been analyzed, an method has been proposed to determine the priori probability distribution of root node. The prediction model of the kiln ring has been established and used to predict the test data.