Shujun Liu

Deep learning has become an important tool for wind power forecasting because it can help improve wind energy utilization and support reliable grid-connected operation. For wind farms, accurate turbine-level forecasting depends on spatial interactions among turbines and temporal evolution of historical operating data. In this study, a spatio-temporal forecasting framework is developed by combining a Graph Attention Network with a Temporal Convolutional Network. The graph attention module describes the neighborhood relations among turbines and learns their influence strengths adaptively, while the temporal convolution module extracts temporal patterns from multivariate SCADA sequences for multi-step prediction. On this basis, the learned attention weights are further used to define a node influence metric. This makes it possible to identify a small set of key turbines and use only their historical data to predict the future power output of the whole wind farm. The proposed framework is evaluated using one year of SCADA data from 134 turbines. A sliding-window dataset is constructed, and the model is tested on the training, validation, and test sets. The results show that the method can capture the spatio-temporal dependencies within the wind farm and still provide effective farm-wide forecasting when only limited observation nodes are available. The value of this work lies in organizing existing techniques around a practical wind farm forecasting task and in providing an interpretable prediction strategy based on key turbine selection, rather than in proposing a fundamentally new theoretical model.

The cooling with the traditional condensation method leads to huge energy consumption, while increasing attention has been paid to radiant cooling because of its characteristics of no additional energy consumption and no pollution. In order to obtain materials with higher infrared emissivity and better performance for daytime passive radiation cooling materials, the infrared emissivity of different materials was studied based on the finite-difference time-domain method. A new composite material with high emissivity has been found. The results show that the highest emissivity can reach 99.1% by adding Si₃N₄, Al₂O₃ and Fe₂O₃ particles with volume fractions of 6% and diameters of 50 nm into polydimethylsiloxane. This is the most excellent emissivity ever found. By combining the emitting layer made of polydimethylsiloxane mixed with nanoparticles with the reflecting layer made of Ag foil, the new film material can reach a solar transmissivity of 96.4% and a “sky window” mean emissivity of 94.2%. A new composite material with high emissivity and high reflectivity has been realized. The new composite material can be used as a radiation cooling material with good performance and help to solve the cooling problem caused by energy consumption.

As wind energy development increases, accurate wind energy forecasting helps to develop sensible power generation plans and ensure a balance between supply and demand. Machine-learning-based forecasting models possess exceptional predictive capabilities, and data manipulation prior to model training is also a key focus of this research. This study trained a deep Long Short-Term Memory (LSTM) neural network to learn the processing results of the Savitzky-Golay filter, which can avoid overfitting due to fluctuations and noise in measurements, improving the generalization performance. The optimum data frame length to match the second-order filter was determined by comparison. In a single-step prediction, the method reduced the root-mean-square error by 3.8% compared to the model trained directly with the measurements. The method also produced the smallest errors in all steps of the multi-step advance prediction. The proposed method ensures the accuracy of the forecasting and, on that basis, also improves the timeliness of the effective forecasts.

Spraying flue gas desulfurization wastewater into flue ducts is an emerging technology that is receiving extensive attention in thermal power plants. In order to study the evaporative performance of wastewater-atomizing droplets under variable working conditions, a combined Euler–Lagrange model was developed to demonstrate the thermal behavior of FGD wastewater spray evaporation in flue gas. The effects of several control factors under various operating conditions were numerically determined and validated against experimental data. Due to the complicated parameters and various other conditions, a least-square support vector machine (LSSVM) model relying on numerical results was used to anticipate the evaporation rate of the droplets. We prove that the LSSVM model has high prediction accuracy for the evaporation rate at different cross-sections of flue under a different operating situation. The conclusion is that for the sake of improving the quality of evaporation, the spacing between two adjacent nozzles should be increased while increasing the flow rate. However, using a higher flue gas temperature, higher initial temperature and smaller diameter of droplets can shorten the time and distance of complete evaporation. In summary, this research analysis can be used effectively to determine the design of the FGD wastewater flue gas evaporative process in thermal power plants.