java生成正态分布方法

//正态分布函数，从三种生成法中随机选择一种，更加随机
public class Normal {
public Normal() {
}

public double normalRandom1(double a, double b) {//注意这里的b是方差，等于标准差的平方
double temp = 12;
double x = 0;
for (int i = 0; i < temp; i++)
x = x + (Math.random());
x = (x - temp / 2) / (Math.sqrt(temp / 12));
x = a + x * Math.sqrt(b); return x;
}

public double normalRandom2(double a, double b) {
double pi = 3.1415926535;
double r1 = Math.random();
Math.random();
Math.random();
Math.random();
Math.random();
Math.random();
Math.random();
Math.random();
double r2 = Math.random();
double u = Math.sqrt((-2) * Math.log(r1)) * Math.cos(2 * pi * r2);
double z = a + u * Math.sqrt(b); return (z);
}

public double normalRandom3(double a, double b) {
double f = 0;
double c0 = 2.515517, c1 = 0.802853, c2 = 0.010328;
double d1 = 1.432788, d2 = 0.189269, d3 = 0.001308; double w;
double r = Math.random();
if (r <= 0.5) w = r;
else w = 1 - r;
if ((r - 0.5) > 0) f = 1;
else if ((r - 0.5) < 0) f = -1;
double y = Math.sqrt((-2) * Math.log(w));
double x = f * (y - (c0 + c1 * y + c2 * y * y) / (1 + d1 * y + d2 * y * y + d3 * y * y * y));
double z = a + x * Math.sqrt(b); return (z);
}

//然后判断用哪个正态分布随机数函数的方法如下：
public double normalRandom(double a, double b) {
double r = Math.random() * 9;
switch ((int) r / 3) {
case 0: return normalRandom1(a, b);
case 1: return normalRandom2(a, b);
case 2: return normalRandom3(a, b);
}
return 0.0;
}
}