论文部分内容阅读
从铁路行车事故的预测问题出发,试图找到稳定有效的方法对铁路行车事故进行预测。首先引入BX数据生成法对原始数据序列进行处理,以弱化原始数据之间的随机性。建立了单因子系统云灰色SCGM(1,1)。模型,揭示铁路行车事故时序变化的发展趋势。参照原始数据的中心趋势曲线,来划分铁路行车事故状态,得到了状态转移概率矩阵。据此计算自相关系数并进行归一化,作为各步马尔可夫链的权重,提出了加权马尔可夫SCGM(1,1)。模型,以修正SCGM(1,1)。模型的预测值,对铁路行车事故总数进行了拟合和预测。结果表明:相比较而言,加权马尔可夫SCGM(1,1)。模型在对铁路行车事故的拟合和预测中均有较好的效果,拟合精度和预测精度分别达到了98.92%和96.36%。
Starting from the prediction of railway traffic accidents, trying to find a stable and effective method to predict railway traffic accidents. Firstly, BX data generation method is introduced to process the original data sequence to weaken the randomness between the original data. One-factor system cloud gray SCGM (1,1) was established. Model to reveal the development trend of railway traffic accident sequence changes. With reference to the central tendency curve of the original data, the status of railway traffic accidents is divided and the state transition probability matrix is obtained. Based on this, the autocorrelation coefficients are calculated and normalized, and weighted Markovian SCGM (1,1) is proposed as the weight of each step Markov chain. Model to correct SCGM (1,1). The predicted values of the model fit and predict the total number of railway traffic accidents. The results show that, compared with the weighted Markov SCGM (1,1). The model has a good effect on the fitting and prediction of railway traffic accidents, and the fitting accuracy and prediction accuracy reach 98.92% and 96.36% respectively.