Font Size: a A A

Study On Sanding Force And Optimal Sanding Parameters In The Belt Sanding Process Of Wood Materials

Posted on:2016-06-11Degree:DoctorType:Dissertation
Country:ChinaCandidate:B LuoFull Text:PDF
GTID:1221330461959609Subject:Wood science and technology
Abstract/Summary:PDF Full Text Request
Sanding is a very important procedure in wood processing. It directly affects the accuracy and surface quality of the product. Currently, the research of belt sanding for wood materials is very few. Most research is referring to the theory of wheel-steel grinding for homogeneous materials and calculative model of sanding force (sF).In this research, orthogonal testing method was adopted to test sF when different wood materials were sanded by sanding belt. The impact of sanding parameters on sF, surface roughness (R) and active power (P) was analyzed. Also the relationship of normal force (nF) with R and P was studied. Then the model of sF generating in the sanding process of wood materials was established by neural network system. Finally, fuzzy comprehensive evaluation (FCE) was adopted to optimize the sanding parameters. The main conclusions are as follows:(1) Wood grain has impact on sF. The sF is maximum when angle between feed direction and grain (λ) is 0°, and is minimum when λ is 90(?). Sanding thickness (Ts) and granularity (G) have the greatest impact on sF and nF, and sanding speed (V) has the least impact.(2) sF can be estimated by the empirical formulas:sF=Ax2+Bx+C+C1, A, B, C-constant, x-G or Ts, C1-feed speed (U) corrected value.(3) Predicting sF by the neural network model, the actual maximum error is 24.14%, the minimum error is 0.17%, and the average error is 10.46%. The neural network model of sF can be used as a universal model to forecast sF.(4) G has the greatest impact on R. The general trend of R is decreasing when G increases. There is not an obvious relationship between nF and R. The distribution of R has vacuum area when sanding middle density fiber board (MDF) and Pinus koraiensis Sieb (λ=0° and λ=45°).(5) V has the greatest impact on P. The general trend of P is increasing when V increases. There is not an obvious relationship between nF and P.(6) The optimal sanding parameters determined by FCE considering P and R are as follow:In most situations, when weight vector (P, Ra)= (0.7,0.3) or (0.3,0.7), G is equal or greater than 100, U, V and Ts are to minimum values;When weight vector (P, Ra)=(0.8,0.2), G is equal to or less than 80, U, V and Ts are to minimum values;When weight vector (P, Ra)=(0.2,0.8), G is equal or greater than 80, U and Ts are to minimum values, V is at a high speed.
Keywords/Search Tags:Wood material, sanding parameter, sanding force, neural network modelling, fuzzy comprehensive evaluation
PDF Full Text Request
Related items