1968 年 14 巻 6 号 p. 330-333
The present investigation was undertaken to apply theoretical methods to the ventilation of polluted air in room where carbondioxide was not homogeneously diffused. In the case of one dimensional diffusion, C is the concentration of carbondioxide at time nδT, and subscripts m, m+1 and m-1 denote space mδX, (m+1) δX and (m-1) δX, respectively, and superscript+denotes time (n+1) δT. Assuming that the concentration of carbondioxide depends on time and space as described in the previous paper, a following equation can be obtained : [numerical formula] where X is a distance from a source of evolution of carbondioxide, and D is a constant (not equal to 0). According to a Implicit difference analogue for the above equation, Taylor's expansion theorem, and Crank-Nicolson's method, a following equation can be obtained : [numerical formula] In the case of two dimensional diffusion, a following equation can be obtained as in the case of one dimensional : [numerical formula] where subscripts j, k ; j, k-1 ; j, k+1 ; j-1, k ; j+1, k denote space x=jh, y=kh ; x=jh, y=(k-1) h ; x=jh, y=(k+1) h ; x=(j-1) h, y=kh ; x=(j+1) h, y=kh, respectively. It was found that Crank-Nicolson's method is available for the theoretical study of natural ventilation of polluted air diffused heterogeneously in room as well as Schmidt's and Dusinberre's methods.