Crank Slider Mechanism Of Error Analysis And Compensation Control

With the rapid development of science and technology, traditional machinery,apparatus is working toward the development of high-speed, heavy, light,miniaturization and precision of the direction. With high-speed, precision machineryand instruments are proposed higher and higher demands for precision。 Themachinery and equipment design, manufacture, installation, wear and measurement ofthe process can produce an error, and the kinematic pair in gap is caused by the mainreason of error. And too large clearance not only greatly reduced mechanismkinematic accuracy, but also causes the motor side impact, appeared a large elasticdeformation, acceleration component wear, so that the clearance volume becomeslarger, producing excessive noise. If artificially reduce the gap, it will not onlyproduce the possibility of jam and affect the life of the machine, reducing thereliability of the machine, but also increases the difficulty of control. In order to studythe existence of motion pair clearance impact of the stability of mechanism, in thispaper, by using dynamic analysis and BP neural network method as an Slider-Crankmechanism with clearance example to set up a mathematical model and inverse，fromthe influence of different reasons of mechanism, analysis machinery and origin of theerror of characteristic, Construct an inverse model feed-forward control, PID plusinverse model feed-forward closed-loop control, neural network internal control andfuzzy control to control the gap caused by nonlinear error. The simulation resultsshow that: The mechanism with clearance due to the existence of the gap the stabilitycompared with of the ideal mechanism is instability; Owing to the mechanism withclearance is a nonlinear system and it is very complex to build accurate model,compared with using the Traditional kinetic analysis method, neural network modelis more close to the actual condition of crank slider mechanism, the simulation speed faster and results more meaningful; Compared to the traditional inverse modelfeedforward open-loop control, closed-loop control of the PID plus inversefeedforward model, thanks to the neural networks’ strong robustness and fuzzycontrol’s humanization and intelligent, using neural network internal model controland fuzzy control to control clearance due to the existence of the gap are moreefficient and stable.