Railway Equipment Data Fusion Technology based On "One Map"

The vast amounts of data generated by today's rail operations are often underutilized. As an important part of smart city and intelligent transportation construction, railway equipment informatization is the trend of future intelligent railway development. There is a wide variety of railway equipment, and the data generated by it is usually stored in separate systems with decentralized data sources. To solve these problems, this paper proposes a data fusion method for integrated management and integration of equipment and facility data, business thematic data and real-time weather data by using artificial neural network to ensure unique data sources and unified data standards. After the railway data fusion is realized, the fused data is trained by the machine learning algorithm, and then a comprehensive railway portrait with multi-dimensional information is drawn. Using a comprehensive picture of the railway, railway decision-makers can make decisions according to this picture. The experimental results show that when the training sample size is 1e6, compared with the training decision provided by a single data source, the precision and recall of the training decision can be improved by about 20% by using multi-source data fusion.