Development Of An Intelligent Diagnosis System For Crushing Machinery Faults

The crushing machine is widely used in bulk material processing systems in various industrial fields and is an important equipment in bulk material handling systems.The crushing machine breaks through the material,and there are faults such as impact and wear.Due to the complexity of the fault and the variety of movements of the parts,the crushing machine is a heavy-duty impact type mechanical equipment.The actual production has a high failure rate and long maintenance period.With the rapid development of mechanical fault diagnosis technology,the crushing machinery urgently needs an intelligent diagnosis system with specialization,reliable performance,high diagnosis rate and intelligentity to reduce the error of manual subjective judgment and reduce the demand for professional technicians.Therefore,the establishment of a reliable dedicated intelligent diagnostic system to ensure the smooth operation of the crushing machinery has important practical significance for the bulk material processing system in various industrial fields.The mechanical fault intelligent diagnosis process mainly includes three steps: the acquisition and analysis of the vibration signal,the extraction of the vibration characteristic information,the analysis of the fault mechanism,and the pattern recognition of the fault by the vibration characteristic value of the fault mechanism.The vibration signal acquisition in the whole process is the basis,feature extraction is the key,which will directly affect the accuracy and reliability of pattern recognition,and fault pattern recognition is the result.In this paper,the crushing machinery is taken as the research object,the structure of the crushing machine is analyzed,and the failure mechanism and fault characteristics of the crushing machine are studied.By collecting fault samples of several common types of crushing machinery,the sample is analyzed by amplitude domain signal,time domain signal and frequency domain signal.After signal processing,feature extraction is performed to determine the neural network learning method,and the improved BP neural network is utilized.An intelligent diagnostic system is established by training the neural network through the collected samples.The practical significance of this paper is to develop a special intelligent diagnosis system for the crushing machinery through the characteristic analysis and feature extraction of vibration,and use the neural network to reduce the manual subjective judgment and the demand for professional and technical personnel.The intelligence can automatically complete the feature extraction and Fault identification improves the recognition rate and stability of fault intelligent diagnosis.