Research And Application Of Key Technology Of Centrifugal Pump Selection Based On Expert System

This paper is based on expert system to research and solve the selection optimization problem of centrifugal pumps.Pump selection is the first important step of pump design and usage.After investigate and survey the internal companies regard the centrifugal pumps as its main products,we found that centrifugal pumps have a wide range of products,and when facing so many complicated production samples,users might have problem to start.When the professionals are facing these samples,the selection results are usually less satisfied and needs to the selection work many times.This kind of selection work is not only low efficiency,but also can not guarantee that the efficacy of the results of the selection process.Based on the situation above enterprises are facing,on the basis of summarizing the theory of centrifugal pump selection,under the condition of considering both internal and external factors and process parameters,this paper puts forward to research and solve the selection optimization problem of centrifugal pumps based on expert system.The main works of the subject are as follows:(1)First elaborated the background of the research and significance of the subject;systematically summarized the knowledge based present research status of the centrifugal pump selection methods,explained the subject's main research content and the key issue that need to be solved.Summarized the relevant basic theory of centrifugal pump selection,summed up relevant theoretical calculation method,laid the theoretical foundation for the establishment of subsequent selection model.(2)Starting from the knowledge structures of selection expert system,centrifugal pump as the research object,established a common model to solve the uncertainty problem of selection.The establishment of the model is divided into two parts,one is establishing a model to solve the uncertainty problem in the process of acquirement of selection parameters,and the other is establishing an uncertainty model of solving selection operating.The former is based on the weight of certainty evidence and changes the acquisition of the parameters from dealing with the uncertainty of the problem into the solution of the problem of deterministic evidence.The latter gives the mathematical model of the best number of polynomial fitting curve and the determination of the speed and flow adjustment range model.Separately solves the problem of the fitting of the selection operating curve and the problem of the speed regulation of the selection operation.(3)The parameter mapping transformation algorithm based on evidence factor is proposed to solve the optimization problem of parameter acquisitionIn order to prove the certainty state factor,the attribute of the priori evidence is transformed into a deterministic state factor.And according to the placements of evidence conflict relationships status of state factors calculate the credible evidence weigh of the different state factors.The optimal selection of the pump type is selected by the selection parameters of the maximum weigh value.The instance results proved the effectiveness and high efficiency of pump type results that obtained by the flow range that meet the condition which according the lift parameters corresponding to the largest weight value.(4)Raise a method of using the improved decision tree pruning algorithm to optimize and amend the diagnosis rule of operating pointThe improved algorithm makes correction and pruning optimization of the rough classified decision Tree generated by C4.5 algorithm.And prevented the decision tree itself appeared over fitting problem.After correcting,each optimization is a branch reasoning optimized diagnostic rules.The instances proved that using the diagnostic rules conclusion that obtained from optimized pruning algorithm is the best optimal diagnostic adjusting method of that operating point deviates from efficient district.(5)This paper introduces the structure,the process,the management of the rule knowledge and the main function modules of the expert selection system.Through the analysis and comparison of the instances,we know that the selection results proved the lift efficiency of the selection scheme is higher than the standard average lift efficiency.The energy consumption of the selection final program was saved about 15%.The results of the instances proved the technical theory of the selection scheme and the correctness of the method.Ensured the efficiency,energy saving and reliability of the pump operation customs selected.Research shows that the research results of this paper can effectively improve the efficiency of centrifugal pump selection,improve the selection process,and reduce energy loss of selection result.The selected pump type can run smoothly in the best condition,and achieve high efficiency and energy saving purpose.