Measures for controlling large deformations of underground caverns under high in-situ stress condition--A case study of JinpingⅠhydropower station

The Jinping I hydropower station is a huge water conservancy project consisting of the highest concrete arch dam to date in the world and a highly complex and large underground powerhouse cavern. It is located on the right bank with extremely high in-situ stress and a few discontinuities observed in surrounding rock masses. The problems of rock mass deformation and failure result in considerable challenges related to project design and construction and have raised a wide range of concerns in the fields of rock mechanics and engineering. During the excavation of underground caverns, high in-situ stress and relatively low rock mass strength in combination with large excavation dimensions lead to large deformation of the surrounding rock mass and support. Existing experiences in excavation and support cannot deal with the large deformation of rock mass effectively, and further studies are needed. In this paper, the geological conditions, layout of caverns, and design of excavation and support are first introduced, and then detailed analyses of deformation and failure characteristics of rocks are presented. Based on this, the mechanisms of deformation and failure are discussed, and the support adjustments for controlling rock large deformation and subsequent excavation procedures are proposed. Finally, the effectiveness of support and excavation adjustments to maintain the stability of the rock mass is verified. The measures for controlling the large deformation of surrounding rocks enrich the practical experiences related to the design and construction of large underground openings, and the construction of caverns in the Jinping I hydropower station provides a good case study of large-scale excavation in highly stressed ground with complex geological structures, as well as a reference case for research on rock mechanics.

Wind Power Probability Density Prediction Based on Quantile Regression Model of Dilated Causal Convolutional Neural Network

Aiming at the wind power prediction problem,a wind power probability prediction method based on the quantile regression of a dilated causal convolutional neural network is proposed.With the developed model,the Adam stochastic gradient descent technique is utilized to solve the cavity parameters of the causal convolutional neural network under different quantile conditions and obtain the probability density distribution of wind power at various times within the following 200 hours.The presented method can obtain more useful information than conventional point and interval predictions.Moreover,a prediction of the future complete probability distribution of wind power can be realized.According to the actual data forecast of wind power in the PJM network in the United States,the proposed probability density prediction approach can not only obtain more accurate point prediction results,it also obtains the complete probability density curve prediction results for wind power.Compared with two other quantile regression methods,the developed technique can achieve a higher accuracy and smaller prediction interval range under the same confidence level.

Spatial-temporal Dynamic Forecasting of EVs Charging Load Based on DCC-2D

The charging load of electric vehicles(EVs)has a strong spatiotemporal randomness.Predicting the dynamic spatiotemporal distribution of the charging load of EVs is of great significance for the grid to cope with the access of large-scale EVs.Existing studies lack a prediction model that can accurately describe the dual dynamic changes of EVs charging the load time and space.Therefore,a spatial-temporal dynamic load forecasting model,dilated causal convolution-2D neural network(DCC-2D),is proposed.First,a hole factor is added to the time dimension of the three-dimensional convolutional convolution kernel to form a two-dimensional hole convolution layer so that the model can learn the spatial dimension information.The entire network is then formed by stacking the layers,ensuring that the network can accept long-term historical input,enabling the model to learn time dimension information.The model is simulated with the actual data of the charging pile load in a certain area and compared with the ConvLSTM model.The results prove the validity of the proposed prediction model.