Research And Modeling Of Solid Wood Sawing Performance Based On Back Propagation Neural Network

Oak and pine are widely used in the field of solid wood product manufacturing due to their unique natural patterns,good touch texture,short growth cycle,and favorable prices.An important method for cutting and manufacturing solid wood products such as furniture,flooring,doors and windows.Sawing is a basic processing method for fine machining of solid wood products such as milling and carving.However,problems such as poor surface quality often arise during sawing,which is accompanied by high energy consumption and noise,and has a significant impact on environmental pollution.In view of this,this article adopts experimental methods to study the sawing processing performance of oak and pine.Experiments were conducted on a circular saw machine using a hard alloy circular saw blade to cut oak and pine.The relationship between sawing parameters(tooth rake angle,saw blade speed,feed speed)and sawing power,noise,and surface roughness was studied.The significance of the sawing parameters and their interaction on sawing power,noise,and surface roughness was determined through variance analysis.Subsequently,the surface damage mechanism of oak and pine processing was analyzed using an optical microscope.Finally,a prediction model for shear power,noise,and surface roughness was established using the BP neural network.The main conclusions obtained from the study are as follows:1、When the rake angle and saw blade speed increase,the sawing power of oak and pine can be reduced,while an increase in feed speed will increase the sawing power.The feed speed has the greatest impact on the sawing power of oak and pine.2、The rake angle has no significant impact on the sawing noise of oak.When the feed speed is high,the pine sawing noise increases with the increase of the front angle of the sawtooth.An increase in saw blade speed and feed speed can cause an increase in sawing noise.The biggest impact on the sawing noise of oak and pine is the speed of the saw blade.3、When the rake angle and saw blade speed increase,the surface roughness of oak and pine can be reduced,while an increase in feed speed will increase the surface roughness.The feed rate has the greatest impact on the surface roughness R_a of oak sawing,while the serrated rake angle has the greatest impact on the surface roughness R_a of pine sawing.4、The worst sawing power and surface roughness were obtained by sawing oak and pine at 10°rake angle,4000r/min sawing speed,0.15m/min feed rate,817W,15.3μm and 586W,26.4μm respectively;while the best sawing power and surface roughness were obtained by sawing oak and pine at 30°rake angle,6000r/min sawing speed,0.05m/min feed rate,.The best sawing power and surface roughness were obtained by sawing oak and pine at 30°front angle,6000r/min sawing speed,0.05m/min,with 261W,5.2μm and 203W,12.5μm respectively,a reduction of 68.0%,66.1%and 65.4%,52.7%.The correct choice of combined saw tooth rake angle,saw blade speed and feed rate can obtain better economic benefits.In the actual production processing,you can choose a larger feed speed for roughing the workpiece to quickly remove excess material,and then use a low feed speed for finishing the workpiece to obtain better surface quality.Reasonable matching to improve production capacity.5、The main reasons for the poor surface quality of oak sawing are the crushing of wood fibers at the pipe hole,surface scratches,and surface knife marks;Burrs are the main reason for the deterioration of surface quality during pine sawing.Taking into account both sawing power and noise,the surface quality of oak processing should be improved by increasing the rake angle of the sawtooth and reducing the feed rate,while the surface quality of pine processing should be improved by reducing the feed rate.6、The established neural network models for oak and pine have high prediction accuracy for their respective sawing power,sawing noise,and surface roughness.,The fits were 0.978,0.845;0.978,0.835;0.982,0.992,respectively.BP neural networks can be used to predict the sawing power,sawing noise,and surface roughness of oak and pine,without the need for complex,expensive,and time-consuming experimental research.