| As an important component of air conditioning equipment,such as air conditioning,structural integrity of cross-flow fan has a d irect impact on level of noise control and stability of airflow.Cross-flow fan is produced by injection molding process.Due to die wear or changes of injection molding process control parameter,there will be material shortage fault.If these substandard products can not be detected and eliminated,failure rate will increase,and quality of products will be seriously affected.At present,material shortage fault of cross-flow fan is mainly by artificial visual detection.Due to mass production and high detection speed production line requires,visual fatigue will produced in long time visual detection.In addition,artificial visual inspection also has the disadvantage of being easily influenced by human subjective factors,which further leads to increase of false positives.To solve the problems above,the first generation of on-line monitoring system of cross-flow fans’ material shortage fault was designed.Due to the using of a single type of infrared sensor to detect faults,there are some problems in the first generation monitoring system,such as instability of detection accuracy and lack of effective detection of small fault.Therefore,based on the first generation of monitoring system,improving detection scheme,using laser and infrared composite array,extracting new signal features,using feature parameter fusion algorithm and fault recognition algorithm based on support vector machine.Product automatically placed identification function is increased,and human-computer interaction is improved.At the same time,in order to improve stability and anti-interference ability of monitoring system,design of the circuit is improved.Prototype of second generation monitoring system is completed and debugged,and performance index is better than the first generation equipment.The detection accuracy,stability and small fault(fault area ≤ 1mm~2)detection rate are improved.Under laboratory conditions,the second generation prototype of material shortage fault detection rate of more than 98.4%,of which the small fault detection rate is about 94.3%,when the first generation prototype of fault detection rate is about 97%,and small fault can not be detect effectively.The main contents of this thesis are as follows:(1)Based on the analysis of the performance characteristics of the first generation of monitoring system,for its shortcomings,the principle of the new detection scheme is put forward,the combination of laser and infrared detection method is used,and the overall plan of the second generation monitoring system is designed;(2)Laser and infrared composite detection circuit is designed,including signal excitation circuit,receiving circuit,analog signal processing circuit,communication circuit,laser and infrared composite array layout and mechanical structure are designed,circuit design of the first generation monitoring system is improved,anti-interference ability and stability of long time detection is improved;and product automatically placed identification function is increased;(3)Main control circuit which based on C8051F020 MCU is design;(4)characteristics of optical signal through material shortage fault are analysed,fault feature parameters are extracted;fusion method faults characteristic parameters are proposed,and the fault recognition algorithm based on support vector machine are also studied;(5)Software of monitoring system is written,including the initialization procedures,product placement recognition program,parameters acquisition program,data extraction and fusion procedure,fault recognition program based on support vector machine,results output program and human-computer interaction program;(6)Prototype of second generation monitoring system is made,debugged and tested to verify the performance of the monitoring system. |
PDF Full Text Request