http://blog.sina.com.cn/s/blog_76c31b8e0100qskf.html
在编程过程中,由于数据仿真模拟的需要,我们经常需要产生一些随机数,在C#中,产生一般随机数用Random即可,但是,若要产生服从特定分布的随机数,就需要一定的算法 来支持了,为了方便广大编程人员,我将我再做项目过程中用到的产生服从正态分布、泊松分布、指数分布以及负指数分布随机数的方法与大家共享,希望会有所帮助!
using System; using System.Collections.Generic; using System.Linq; using System.Text; using System.IO;
namespace PZ { class Model { /// <summary> /// 正态分布随机数 /// </summary> const int N = 100; const int MAX = 50; const double MIN = 0.1; const int MIU = 40; const int SIGMA = 1; static Random aa = new Random((int)(DateTime.Now.Ticks / 10000)); public double AverageRandom(double min, double max)//产生(min,max)之间均匀分布的随机数 { int MINnteger = (int)(min * 10000); int MAXnteger = (int)(max * 10000); int resultInteger = aa.Next(MINnteger, MAXnteger); return resultInteger / 10000.0; } public double Normal(double x, double miu, double sigma) //正态分布概率密度函数 { return 1.0 / (x * Math.Sqrt(2 * Math.PI) * sigma) * Math.Exp(-1 * (Math.Log(x) - miu) * (Math.Log(x) - miu) / (2 * sigma * sigma)); } public double Random_Normal(double miu, double sigma, double min, double max)//产生正态分布随机数 { double x; double dScope; double y; do { x = AverageRandom(min, max); y = Normal(x, miu, sigma); dScope = AverageRandom(0, Normal(miu, miu, sigma)); } while (dScope > y); return x; }
/// <summary> /// 指数分布随机数 /// </summary> /// <param name="const_a"></param> /// <returns></returns> public double RandExp(double const_a)//const_a是指数分布的参数λ { Random rand = new Random(Guid.NewGuid().GetHashCode()); double p; double temp; if (const_a != 0) temp = 1 / const_a; else throw new System.InvalidOperationExceptio n("除数不能为零!不能产生参数为零的指数分布!"); double randres; while (true) //用于产生随机的密度,保证比参数λ小 { p = rand.NextDouble(); if (p < const_a) break; } randres = -temp * Math.Log(temp * p, Math.E); return randres; }
/// <summary> /// 负指数分布随机数产生 /// </summary> /// <param name="lam">参数</param> /// <returns></returns> Random ran;
public Model() { ran = new Random(); } public double ngtIndex(double lam) { double dec = ran.NextDouble(); while (dec == 0) dec = ran.NextDouble(); return -Math.Log(dec) / lam; }
/// <summary> /// 泊松分布产生 /// </summary> /// <param name="lam">参数</param> /// <param name="time">时间</param> /// <returns></returns> public double poisson(double lam, double time) { int count = 0;
while (true) { time -= ngtIndex(lam); if (time > 0) count++; else break; } return count; } }
}