| The hydraulic system of the excavator has the characteristics of strong concealment,strong nonlinear time-varying signal,and complicated energy transfer mechanism when the fault occurs.In particular,most fault characteristics are weak in the early stage of the fault and it is difficult to extract effective information.If failures cannot be discovered and resolved in the early stage,it will easily lead to major safety production accidents.Therefore,for the excavator hydraulic system,it has great research significance to achieve fast and accurate diagnosis.In order to solve the above problems,the author focused on the fault diagnosis technology of excavator hydraulic system by consulting a large amount of domestic and foreign literature according to the research direction of this topic.These studies were based on the school-enterprise cooperation project "FW080 full hydraulic crawler excavator development"(project no.FW/RD201717).Author has analyzed the advantages and disadvantages of several common fault diagnosis technology,then the main hydraulic components of the common faults in the excavator hydraulic system are summarized,the result shows: hydraulic components failure often leads to abnormal changes in its running parameters,so the change of operation parameters can reflect the fault information effectively,and the fault diagnosis research scheme of this paper is proposed,which is based on the theory of machine learning regression fitting prediction applied to the diagnosis of excavator hydraulic system and combined with the expert system.A diagnosis method based on extreme learning machine algorithm(ELM)regression fitting prediction model is proposed: first of all,according to the operation under the normal state of excavator hydraulic system parameter fitting prediction model is established,when the fault occurs,Input the operating parameters in the fault state into the prediction model,and then compare the predicted values of the parameters output by the prediction model in the normal state with the actual operating parameters,and determine whether the system has failed through the comparison results.Then the author combines the expert system diagnosis method with the above method in order to further infer and explain the prediction model output residual statistics: the author first built the knowledge of the diagnosis process,fault feature information and fault maintenance plan of the hydraulic excavator hydraulic system The knowledge base of the expert system for fault diagnosis of mechanical hydraulic systems is in the form of an ontology model,and then the reason and location of the fault are inferred by setting inference rules;the author further proposes a diagnosis method based on case matching in order to repeat the past fault diagnosis knowledge Use,first analyze the different parameter changes caused by different fault causes,and then store the feature information such as parameter changes and fault phenomena in the form of case feature information to build a case database for fault diagnosis.When a fault occurs,rapid diagnosis of the fault can be achieved through feature selection,case retrieval,and case matching.At last,the author designs a set of system to realize the above fault diagnosis method by using Visual Studio software,so that the user can know the fault location,fault reason and fault maintenance method of the excavator.In this paper,the simulation model of the experimental prototype hydraulic system is established in the AMEsim environment,and by changing the physical parameters of its hydraulic components,a variety of fault examples are simulated,and the corresponding fault feature information will be obtained to verify the diagnosis method of the excavator hydraulic system proposed in this paper.Effectiveness,the results show that the scheme described in this article is reasonable and effective,and the scheme described in this article also provides a certain reference for the diagnosis of other engineering vehicle faults. |
PDF Full Text Request