说明：  若非奇异矩阵A能够分解为一个正交矩阵Q与非奇异上三角矩阵R的乘积，即： A=QR。则称其为A的QR分解。 实现QR分解的方法有很多种，包括Givens变换法，Householder变换法，Schemit正交化法。具体原理可以参考《矩阵论》（程云鹏，西工大出版）教材，这里仅给出三种实现QR分解的程序。
(If nonsingular matrix A can be decomposed into a product of orthogonal matrix Q and nonsingular upper triangular matrix R, that is: A=QR. It is called QR decomposition of A. There are many methods to realize QR decomposition, including givens transformation, householder transformation, and scheme orthogonalization. The specific principle can refer to the teaching materials of matrix theory (Cheng Yunpeng, published by Xigong University). Here, only three kinds of QR decomposition procedures are given.)