Fault Diagnosis For Loader Based On Rough Set Theory

Construction machinery is the main equipment in the national economy. Therefore it is very important to develop the fault diagnosis technology of construction machinery, which can quicken the rate of project's progress, and improve its efficiency. In the research of fault diagnosis the problem related to knowledge acquisition has become a bottleneck of the development of fault diagnosis technology. In this dissertation, the rough set theory has been applied to the fault diagnosis of the loader—a typical construction machinery, in order to solve the problems such as knowledge acquisition, fault diagnosis when the datum about working condition is imprecise, inconsistent, and incomplete, fault-tolerance performance of fault diagnosis system. The research statue of fault diagnosis theory and technique are analyzed comprehensively. Then the rough set theory and its current research are discussed generally and the necessity of the application of rough set theory in fault diagnosis field is also stated in this dissertation. On the basis of introducing the rough set theory, the method and technique of fault diagnostic knowledge acquisition based on rough set is discussed. This dissertation put forward the model of fault diagnosis using decision table. By taking the rough set and duality Boolean calculation as the basis, the algorithm to conduct the fault diagnosis knowledge acquisition is researched with emphasis. The practical aspect of this knowledge acquisition's method is illustrated by a concrete fault diagnosis example. An approach of knowledge acquisition based on hierarchical acquisition strategy is put forward and applied to the loader's fault diagnostic knowledge acquisition. The fault diagnostic knowledge of the loader is analyzed and represented by production rules. Then the inferential strategy and process of loader's fault diagnosis based on the rule got by knowledge acquisition are described With the usage of the knowledge acquisition and method, a Loader Teleservice and Fault Diagnostic System has been developed, validate the proposed knowledge acquisition methods. Finally, conclusion is drawn and the future research focus is pointed out.