Research On Intelligent Fault Diagnosis System Of Multi-cylinder Hydraulic Cone Crusher

At present,with environmental protection playing an increasingly important role and the rapid growth of demand for machine-made sand,the domestic crusher market has also ushered in another spring.Due to its crushing characteristics and superior performance,the multi-cylinder hydraulic cone crusher has gradually become the first choice in the second and third stages of the crushing process.However,due to a late start in China,lack of experience,poor electronic control,and lack of technical indicators,the failure rate of this type of crushing is very high,which greatly affects production.Therefore,we should learn from advanced foreign companies and technologies.First,establish a crusher monitoring system based on the establishment of a complete and accurate fault indicator system,so as to achieve preliminary fault diagnosis and early warning;then,introduce BP nerves Network and other intelligent algorithms to combine the current popular artificial intelligence with the excellent characteristics of BP neural network in pattern recognition,so as to research and explore a further intelligent fault diagnosis system based on the comprehensive analysis of multiple sensor parameters,to make up for the the limitation of the method on many failures based on the first fault diagnosis by monitoring threshold.Based on the above content,this article mainly completed the following research content:(1)By understanding the structure and composition of the crusher and the operating principle of each part and the crusher as a whole,the reasons for the high failure rate of this type of crusher are analyzed.from the view of structural factors of the multi-cylinder hydraulic cone crusher and the harsh working environment.Meanwhile,by analyzing some main and typical faults of the crusher,the specific causes of the faults and the effects the faults will cause are realized,so as to prepare for the next step of selecting sensors and designing the electronic control system.(2)Select the corresponding detection sensor,and set the standard value or threshold value of the relevant sensor according to the latest collected parameter indicators;by analyzing the start-up process of the multi-cylinder hydraulic cone crusher and the current domestic general crusher electrical control logic,re-optimized and designed a new electronic control logic and drew the corresponding electrical schematic diagram,using PLC and touch screen to achieve remote monitoring and control functions.(3)Collect and analyze the original data of the failure of the multi-cylinder hydraulic cone crusher.According to the input and output characteristics of the BP neural network and the actual working status of the multi-cylinder cone crusher,select the appropriate failure data corresponding to each type of failure.The feature parameter matrix provides a set of fault samples for the training and testing of the diagnostic model.(4)By studying the concept,principle and characteristics of BP neural network,and combining the faults in the actual work of the multi-cylinder hydraulic cone crusher,the method of applying the BP neural network to the field of fault diagnosis of the multi-cylinder crusher is analyzed.Input the fault characteristic sample data into the network,use the fault type as the expected network output,set the corresponding training samples,establish the fault diagnosis network,and compare the training results to determine the specific structure of the network,and establish the BP neural network multi-cylinder hydraulic cone crusher fault Diagnostic model.(5)Validation and testing of fault diagnosis model for multi-cylinder hydraulic cone crusher.Taking the real fault parameters of Shaorui Heavy Industry SCH8000 crusher as an example,using real fault samples collected in real engineering applications to train the fault diagnosis model of the multi-cylinder hydraulic cone crusher that has been preliminarily established using the BP neural network algorithm in the previous part.And the result test of the diagnosis model after the training is completed,finally verifies the fault diagnosis ability of the multi-cylinder hydraulic cone crusher fault diagnosis system.