Lee, J.-W. and Park, S.-C., 2016. Artificial neural network-based data recovery system for the time series of tide stations.

Accurate prediction of missing water levels attributable to reasons ranging from recording failure and transmission problems to mistakes made by field staff is essential in coastal and oceanic areas. This paper presents a new system for data recovery based on the artificial neural network soft-computing technique and an end-point fixing method. Only the past time series of tide stations are used, therefore, the data recovery system is not only simple and fast but also effective in situations where all the neighboring tide stations simultaneously malfunction. To verify the efficacy of the proposed data recovery system, we utilized the system at the tide stations located in the eastern part of Korea. Further, we calibrated parameters such as window size and thresholds and conducted performance tests in terms of statistical parameter. Our results indicate that although the performance of the proposed system declines marginally as the gap size increases, it performs creditably in alleviating the gap-filling problem.