卫星成本预测的蒙特卡洛优化模型研究

Analysis of Monte Carlo Optimization Model in Satellite Cost Estimation

传统的卫星成本预测模型预测结果单一,未考虑相应的不确定性因素。对此,提出了基于蒙特卡洛方法的卫星成本预测模型。该模型综合考虑了卫星成本技术的不确定性并给出了预测值的区间范围和概率分布。为了降低应用该模型所产生的误差,提出了最优分层抽样的方法,并通过划分分层区间提高了抽样的效率。最后,将收集到的相关数据代入该模型,其结果精度大约提高了10%左右。可见,使用优化的蒙特卡洛方法进行卫星成本预测具有较好的精确度。

The estimation results of the traditional satellite cost estimation models were too simple, and didn＇t consider the corresponding uncertainties. The thesis put forward satellite cost estimation model based on Monte Carlo method to solve these problems. This model considered the uncertainties of satellite cost technology and gave the interval area and probability distribution of the estimation value. In order to reduce the error caused by this model, it gave us the perfect delaminating sampling method, and improved the efficiency of sampling through dividing the delaminating interval. Finally, it put collected corresponding data into this model and compared with the traditional satellite cost estimation method, the precision almost increases 10% . Therefore, we can see that using Monte Carlo method to estimate satellite cost can get better precision.