Research On Intelligent Fault Diagnosis Technique Of Telemetering Equipment

With the development of telemetry equipment to high performance, high integration,high automation, the failure is more complex and variety. In addition, it existed many to manynonlinear mapping relation between the cause of a fault and fault phenomenon, so thetelemetry equipment fault is complex and more difficult to diagnose. The previous faultdiagnosis method can not guarantee the requirement of diagnosis. Therefore, it’s verynecessary to study and build a set of intelligent fault diagnosis system.A PXI bus-based telemetry equipment intelligent fault diagnosis platform is constructedin this paper, with the telemetry equipment integrated support for background, practical andconvenient as the goal, the fault tree, expert system, neural network and virtual instrument testdiagnosis technology as the foundation.Firstly, the method of intelligent fault diagnosis system is introduced, the failure modeand the principle for establishing test point of telemetry equipment is studied, the system’soverall structure and function principle of telemetry equipment intelligent fault diagnosissystem based on virtual instrument is given, on the basis of analysis of telemetry equipmentfault characteristics.Secondly, the diagnosis expert system based on the fault dictionary and the fault tree isstudied. The fault dictionary of the system is designed and constructed. The fault tree model isbuilt. The basic theory of rule-based expert system and the various parts of composition andfunctions of expert system integration advantages was introduced detailedly in the paper. Anew idea which combines fault tree analysis with rule–based expert system was proposed. Byusing a fault tree technology, the expert system can acquire the necessary knowledge. Thefault tree model of telemetry equipment down converter is constructed, the functionalmodules of telemetry equipment expert system of intelligent fault diagnosis were designedand the fault is reasoning with telemetry down converter as an example.Thirdly, method and technique research on fault intelligent diagnosis based on BP neuralnetwork. Aiming at complicated nonlinear mapping relationship of fault symptoms-causesand self-study capability, the BP neural network method was introduced into fault diagnosis inthis paper. The principle was explained. The structure of the BP neural network model wasbuilt. And, the process of learning and training is studied and applied into the telemetryequipment fault diagnosis. Telemetry equipment fault diagnosis model of neural network isbuilt. Then the simulation research has been carried out with the telemetry downconverter, The sample data of telemetry downconverter fault phenomenon and the cause is built, thetraining simulation data and error curve is got, with training these sample data by neuralnetwork program which is written by MATLAB, and the correctness of neural network faultdiagnosis is verified by testing sample data.Finally, hardware and software design for the fault diagnostic system was studied basedon virtual instrument. On the base of PXI system, system hardware architecture was presentedby selecting the virtual instrument which is corresponding with telemetry equipment testing.The information database of intelligent fault diagnosis system is designed by using LabVIEWand SQL server, the software modules of the system is studied and designed. Thecorresponding fault diagnosis system software program is written, the reliability and validityis verified through experiment.The traditional experience troubleshooting turn on intelligent troubleshooting for thebuilding of intelligent fault diagnosis system, and the troubleshooting process is morescientific and reasonable. It shortens the fault equipment troubleshooting time, effectivelyimproves the detection and fault automatic diagnosis ability, accelerates the support andmaintenance efficiency of telemetry equipment.