利用了高斯取整函数 int(x) =[x]和定义了判断函数 F(x1 ,y1 ,x2 ,y2 ) (值为 0或 1) ,使问题转化为求 max∑n1 F(x1 ,y1 ,a,b) ,根据 F的定义使问题的求解过程线性化、规范化 .从而使问题转化成求解二元线性不等式组的问题 ,为此可方便地借助计算机求出最佳结果 .在解答问题二时 ,将角度离散化 ,通过等价距离 max{ |dx|,|dy|}≤ dx2 +dy2≤ 2 max{ dx,dy} ,使问题转化为问题一的情形 ,利用问题一的算法求解 ;最后给出 n口旧井均可利用的充要条件和简明算法 .根据算法求得的最优解是一个区域而不是一个点 ,为模型结果的实际应用提供了选择的余地 Using the Gaussian rounding function int (x) = [x] and defining the decision function F (x1, y1, x2, y2) with a value of 0 or 1, the problem is transformed to find maxΣn1 F (x1, y1 , a, b). According to the definition of F, the solution process of the problem is linearized and normalized, so that the problem can be transformed into the problem of solving a linear system of linear inequalities. For this purpose, the best result can be obtained conveniently by computer. In the second case, the angle is discretized, and the problem is transformed into problem 1 by the equivalent distance max {| dx |, | dy |} ≤ dx2 + dy2 ≤ 2 max {dx, dy} Finally, a sufficient and necessary condition and a concise algorithm are available for all n old wells. The optimal solution obtained from the algorithm is a region rather than a point, which provides a choice for the practical application of the model results