Significant wave height (SWH) is a vital parameter in marine science research and engineering application. The up-to-date reanalysis product ERA5 SWH brings new possibilities to the long-term analysis. However, a systematic assessment of ERA5 SWH on a large scale is still lacking. This paper presents an evaluation of ERA5 SWH against observations from 103 buoys in the North American Atlantic and Pacific sourced by the National Data Buoy Center over the period 1979 to 2019. Overall, the ERA5 SWH has a good agreement with the in-situ observations, with a bias of −0.058 m, root mean squared error of 0.325 m, correlation coefficient of 0.961 and scatter index of 18.54%. The accuracy of ERA5 SWH is satisfactory under the most typical sea states (0.5 m < SWH < 4 m). The monthly analysis shows the performance of ERA5 SWH in summer is the best. The water depth and offshore distance have also been identified to impact the reliability of ERA5 SWH. Although the statistics vary at different locations, the performances of ERA5 SWH at most stations are reasonable. In addition, an evident improvement in the validity over time is observed, which can be attributed to the assimilation of the altimeter wave height.
Significant wave height (SWH) forecasting is a key process for offshore and costal engineering. However, accurate prediction of the SWH is quite challenging due to the randomness and fluctuation features of waves. This paper employs a novel deep learning method, the gated recurrent unit network (GRU), to forecast SWH with lead times of 3, 6, 12 and 24 h. The data sets used in this study include the wind speed of the past 3 h and the current SWH as inputs, which were obtained from six buoy stations in the Taiwan Strait and its adjacent waters. The GRU results are compared with those of back propagation neural network (BP), extreme learning machine (ELM), support vector machine (SVM), and random forest (RF). Although the error indices of the six stations are different, the general performance of GRU is satisfactory, with a faster forecasting speed, smaller volatility and better adaptability. Using buoy station 46714D as an example, the root mean square error (RMSE) predicted by GRU reaches 0.234, 0.299, 0.371, and 0.479 with lead times of 3, 6, 12, and 24 h, respectively.
1 July 2014 in Journal of Natural Gas Science and Engineering
As wind energy and other renewable energy sources are valued by various countries, it is very important to estimate and predict the wind energy level. The accuracy of wind energy prediction mainly depends on the accuracy of wind speed prediction. Therefore, to seek ways of improvement the accuracy of wind speed prediction has become the most important issue. In this paper, three different decomposition methods and commonly used wind speed prediction methods are used to compose the corresponding combined models, and to study which combined prediction model has higher accuracy. According to data research conducted by the National Meteorological Science Center, experiments show that the prediction accuracy of the combined prediction model using the Variational mode decomposition (VMD) method is higher than that of the combined prediction model using empirical mode decomposition (EMD) and Ensemble Empirical Mode Decomposition (EEMD).