Study On Crack Cause Analysis And Rule Extraction Algorithm Of Concrete Arch Dam

The most common of which is the cracks in concrete dams.There are more or less types of cracks in concrete dams that have been built or built at home and abroad.In order to prevent the occurrence and development of dam cracks,in the dam body buried a large number of monitoring equipment,and each type of monitoring instruments have dozens or hundreds of monitoring points,often accumulated a large amount of monitoring data.If the traditional analysis method point by point modeling analysis,the workload is large,the calculation time is long,and can not fully analyze and understand the relationship between a large number of physical quantities,it is difficult to achieve massive data processing needs.The data mining method can extract useful information efficiently from massive data,and automatically excavates useful knowledge hidden in massive,random,fuzzy,noisy,incomplete data,and is the main analysis of the era of large data One of the means.With the development of information technology,data mining technology in the dam monitoring data analysis continue to be applied.Based on the data mining method and the monitoring data of a concrete dam,this paper uses the advantages of rough set analysis,fuzzy clustering,decision tree technology and fusion method to form a new algorithm,which is related to the causes and rule extraction of concrete dam during operation Analysis methods were carried out in-depth study,the main research results and conclusions are as follows:(1)A discretization algorithm with supervised(based on attribute importance)and unsupervised(based on fuzzy C-means clustering)is studied,and the corresponding discretization program is compiled.By discretizing the monitoring data sequence and comparing the two algorithms The advantages and disadvantages and the scope of adaptation.The example shows that the fuzzy C-means clustering algorithm has a fast convergence speed and can take into account the correlation between conditional attributes and condition attributes and decision attributes,which is more efficient and effective in mass data processing.(2)Based on rough set-fuzzy C-means clustering method,the excavation information table of dam damage was constructed,and the excavation model of concrete dam structure was established.The calculation and extraction of crack influencing factors were carried out.Through the example verification,the influencing factors of the cracks of a concrete dam are analyzed,and the main influencing factors of the structural fracture are obtained.The calculation results are consistent with the conclusion of the data analysis.(3)For the rule extraction,the concept of trustworthiness and support of the rough set is introduced on the basis of the attribute reduction set to generate the decision tree,and the rules are extracted and the concept of credibility,support and coverage are introduced into the decision rule evaluation.The new rule is established based on the rough set and the decision tree to extract the change rule of the crack opening degree.The low temperature and low water level is the most unfavorable combination of the crack opening degree,the low temperature crack opening degree is big,the low water level crack opening degree is big,the influence factor temperature has the greatest influence on the crack,It is consistent with the conclusion of the data analysis report,which verifies the validity of the new algorithm of rule extraction and has the advantages of high generalization ability,high reasoning efficiency,less subjective factor control and simplification of algorithm.Can be used for dam disease genetic excavation and rule extraction.(4)The fractal cause analysis and rule extraction system of concrete arch dams are created to realize the visualization of the algorithm,and the main frame of the system,the main interface and the design of some module calculation programs are completed.