The Precision Improvement Strategy Of 3D Printing Based On Fuzzy Neural Network PID By The S Curve Gradual Approaching Value

3D printing technology is a new technology that has changed the world.It not only reduces the consumption of materials but also reduces the pollution to the environment.People just need a 3D printer to make the way personal and customized.But now the product precision is low,the printing model has the ladder pattern to affect the product quality and the printing efficiency is also low.In this paper,the 3D printer with fused deposition modeling(FDM)is the main research object,analyzing the reasons for the errors of the rough surface of the 3D printing model and the step-shaped printing.In order to improve the accuracy of the 3D printer,the main research contents of this paper include the following aspects:(1)Firstly,the paper analyzes the printing principle of 3D printer,finds out the reason of printer error in the printing process,and establishes the formula of the surface smoothness of 3D model.Secondly,it is verified that model slice,layer thickness and printing angle are the main reasons that affect the printing quality.Finally,the basic structure and operating principle of the main components of the printer(stepper motor)are analyzed from the control level.At present,the main problems of stepping motor are low frequency vibration and high frequency step loss.The traditional S-shaped acceleration and deceleration curve is segmated by Logistics function,so that the acceleration and deceleration curve is smooth and continuous,which can reduce step loss and vibration of stepping motor and improve the efficiency of printing.(2)In closed loop control,larger step response curves may cause the whole system vibrating and the ordinary PID algorithm could not cover the need of precise stepper motor.Therefore,the PID by the approaching value has been adopted.With comparing results of inputting signals between conventional step modulation and the logistics curve step modulation,the use of Logistics curve is more in line with the acceleration process of stepping motor.However,there is still a problem that the initial acceleration is too slow to affect the printing efficiency.An improved S speed curve has been used to upgrade the working efficiency of 3D printer.Finally,a simulation software has been used for the simulating.The results verify that the improved s-curve PID by the approaching value decreases the frequency of vibration of the stepper motor,which makes the printing more accurate as well as improving printing efficiency.(3)The parameter adjustment is difficult and tedious in the process of s-curve PID by the approaching value control.So the fuzzy algorithm has been used to adjust PID parameters offline and the fuzzy neural network is used to modify PID parameters online.The two algorithms have been verified by the simulation software,and it has been.found that the fuzzy PID controls more accurately and the anti-interference ability is also improved.The adaptive PID of fuzzy neural network makes the stepping motor accelerating increasingly in ‘S curve'.The adaptive PID of fuzzy neural network conforms to the acceleration law of stepper motor and improves the self-learning ability of fuzzy PID.Moreover,the adaptive PID of fuzzy neural network has more precision and shorter stabilization duration than that of fuzzy adaptive PID.