论文部分内容阅读
The Charge-Density method of solving electrostatic field has many advantages over the finitedifference or finite element method,but its accuracy and fitness are still in question.By direct evaluatingelectrical potentials,the errors along surface of the electrode are plotted,which are the maximum errors foran electrostatic problem.It is shown that higher accuracy will be reached if more sub-regions are chosen atwhere the charge density is high or at the region near the area where the field is to be calculated.This will behelpful to understand and use the method efficiently.It is more convenient and accurate to use the charge-density method when we deal with the calculation of an electrostatic system having a great difference in sizebetween electrodes(such as the point emitter diode).Using the information of error on the surface ofelectrode,we can estimate the potential errors in each calculation or rearrange the sub-regions to improvethe accuracy of the next calculation.A program is set up.The difference between the calculated data and thatof E.Harting and F.H.Read(1976)is less than 1%.
The Charge-Density method of solving electrostatic field has many advantages over the finitedifference or finite element method, but its accuracy and fitness are still in question. BY direct evaluatingelectrical potentials, the errors along surface of the electrode are plotted, which are the maximum errors foran electrostatic problem .It is shown that higher accuracy will be reached reached more more sub-regions are chosen atwhere the charge density is high or at the region near the area where the field is to be calculated. This will be helpful to understand and use the method efficiently.It is more convenient and accurate to use the charge-density method when we deal with the calculation of an electrostatic system having a great difference in size between electrodes (such as the point emitter diode). Using the information of error on the surface ofelectrode , we can estimate the potential errors in each calculation or rearrange the sub-regions to improve the accuracy of the next calculation. A program is set up. The difference between the calculated data and that of E. Harting and F.H. Read (1976) is less than 1%.