For the fault detection and diagnosis problem in largescale industrial systems,there are two important issues: the missing data samples and the non-Gaussian property of the data.However,most of the existing data-driven methods cannot be able to handle both of them.Thus,a new Bayesian network classifier based fault detection and diagnosis method is proposed.At first,a non-imputation method is presented to handle the data incomplete samples,with the property of the proposed Bayesian network classifier,and the missing values can be marginalized in an elegant manner.Furthermore,the Gaussian mixture model is used to approximate the non-Gaussian data with a linear combination of finite Gaussian mixtures,so that the Bayesian network can process the non-Gaussian data in an effective way.Therefore,the entire fault detection and diagnosis method can deal with the high-dimensional incomplete process samples in an efficient and robust way.The diagnosis results are expressed in the manner of probability with the reliability scores.The proposed approach is evaluated with a benchmark problem called the Tennessee Eastman process.The simulation results show the effectiveness and robustness of the proposed method in fault detection and diagnosis for large-scale systems with missing measurements.
Radionuclide dating techniques characterized by137Cs and210Pbex have recently been applied in the study of lake sediments around the world.In this study,a chronological series of sediment cores was established based on137Cs and210Pbex analyses along with the evaluation of sediment properties,such as particle size distribution,total organic carbon(TOC),carbonate content,and acid-insoluble residue,to study sediment accumulation rates,sediment sources,and responses to human activities in the Jiuzhaigou National Nature Reserve in southwestern China.In terms of the particle size distribution of sediments,silt content was the highest,and clay and sand contents were relatively low.The sediments displayed high TOC contents because of the significant amounts of vegetation grown in the lakes.The carbonate content was also high due to the overall geological background of carbonates in Jiuzhaigou.Carbonate content tended to decline from top to bottom in the sediment cores,whereas the acid-insoluble residue tended to increase.These results suggested that the depth variation of the environmental parameters of the sediments in two lakes in Jiuzhaigou would correspond to each other.The results indicated that the sediment rate of Jiuzhaigou was generally high with strong siltation,indicating that serious soil loss was induced by intensive human activities in the basin over the past decades.The increases in the mass accumulation rate,contents of acid-insoluble residue,and mean particle size during the periods of 1840–1900s,late 1930s–early 1950s,1966–1978,and2003–2006 revealed the occurrence of severe soil and water loss as a result of extensive agricultural expansion,large-scale deforestation,and road construction in Jiuzhaigou.The deposition rate and the properties of lacustrine sediments could reflect the significant impact of human activities on lake sedimentation during Jiuzhaigou′s history.
LIANG KangkangHU XinxinLI ShiguangHUANG ChengminTANG Ya
