A Processing Method of Missing Value for Industrial Big Data Based on Improved Neural Network Algorithm

With the explosive growth of industrial big data, it is particularly important to deal with the missing value of big data. There are various deficiencies in the traditional methods of handling missing values. To improve the efficiency and accuracy of handling missing values, an improved idea based on Back-Propagation (BP) neural network algorithm is proposed to deal with the missing value in this thesis. Besides, the power data of a steam turbine unit is selected to test and verify the effectiveness of the improved algorithm, and the feasibility of this improved algorithm is well verified by test and prediction calculation.