中文摘要：

随着磁头滑块的飞行高度不断降低，给气体润滑方程的数值求解带来了诸如计算时间过长、甚至计算发散等方面的问题。为了获得1Tbit／in^2的存储密度，磁头滑块尾部的最小飞行高度接近1．5nm。本文基于作者提出的修正气膜润滑方程的线性流率（LFR）模型，考虑磁头滑块表面高度的不连续性，建立了基于有限体积法的气膜润滑方程离散格式，并把网格自适应技术与多重网格法应用到离散方程的迭代算法中，发展了可模拟最小飞行高度为0．5nm时磁头滑块压力分布的数值模拟方法与有效算法。文中以一个具有复杂表面形状的磁头滑块为例，检验了计算方法与算法的有效性。数值结果表明：在磁头滑块最小飞行高度较低时，必须要考虑滑块表面高度的不连续性，否则就得不到收敛的数值计算结果；与FK—Boltzmann模型相比，LFR模型具有较高的计算效率，采用网格自适应技术与多重网格法能有效地提高求解气膜润滑方程的计算效率。

英文摘要：

With the decreasing of the flying height of the air bearing slider in hard disk drives, many problems appear in the numerical solution of the gas lubrication equation, such as too long computational time or even computing divergence. In order to get the storage density of 1 Tbit/in2 ,the flying height of the air bearing slider will approximate 1.5 nm. In this paper,based on the linearized flow rate （LFR） model of the corrected Reynolds equation proposed by the authors,the gas lubrication equation is discretized by using finite volume method with consideration of height discontinuities of the air bearing surface. Mesh adaptive technique and Multigrid method are used in the iteration algorithm to improve the computational efficiency. A numerical method with efficient algorithm is developed to simulate pressure distribution of the slider with 0.5 nm flying height. The validity of the numerical method is verified by using an air bearing slider with complex surface shapes. Numerical results show that when the minimum flying height is relatively low,the numerical results can not be obtained without considering the height discontinuities of the slider surface. The computational efficiency of the LFR model is higher than that of FK-Boltzmann model. The computational efficiency of solving the gas lubrication equation can be reduced efficiently by using mesh adaptive technique and multigird method.