Search

Estimating the total wind power output from the meteorological information at a province level (called Provincial Regional Wind Power Conversion Model, PRWPCM) plays vital and fundamental roles in energy modeling community and regional wind power forecasting. How to construct a reliable PRWPCM is a real challenge, since PRWPCM involves a large number of widely distributed wind turbines, massive meteorological data across the whole province, and complex nonlinear correlations. This paper proposes a lightweight PRWPCM by integrating Geographic Information System (GIS) analysis technology and Convolutional Neural Network (CNN). First, we conduct the land suitability analysis for wind turbine sites through the multi-criteria GIS layer overlay method to make the provincial wind turbine land suitability map (WTLSM) with scored divisions from the least suitable to the most suitable areas. On this basis, a new fusion mechanism for geographic and meteorological information is proposed, through which the raw meteorological data matrix can be reconstructed to filter and amplify the meteorological information that is more relevant to the total wind power output of the province, and avoid the time-consuming and labor-intensive data collection and processing, large-size model construction and validation. Second, a CNN-based regression architecture is designed to further capture the mapping relationship between the reconstructed meteorological data and total wind power output of the province; each type of meteorological factor is considered as an input channel and the attention modules are introduced to adaptively enhance useful channels and suppress less useful ones. Numerical experiments based on the wind power operation data of Yunnan Province, China, are conducted to validate the superiority of the proposed PRWPCM via benchmarking against 13 classical methods. • Mapping relationship between wind power output and meteorological information at a province level is recognized by integrating GIS and CNN. • A new fusion mechanism for geographic and meteorological information is proposed to achieve lightweight modeling. • Comparative experimental studies are conducted in Yunnan province, China. [ABSTRACT FROM AUTHOR]

A major difficulty in integrating large scale wind power generation in an electrical power system is that wind generated power appears to be erratic, intermittent, and volatile. In this paper, we demonstrate the efficacies of a novel ultra short term 1-step ahead wind generated power prediction model, by combining two best of breed machine learning models in their respective areas of applications: a deep convolutional neural network (CNN) model, known to be effective in classification problems, and a bi-directional long short term memory (Bi-LSTM) model, known to be effective in 1-step ahead time series prediction problems, using a cross attention (XA) mechanism on three challenging practical datasets: the East-China dataset, the Yalova (Turkey) dataset, and the 16 MW dataset. There are two alternative cross attention models: (1) using the CNN features as the key–value pair, and the Bi-LSTM features as the query, this we called a XA model, and (2) using the Bi-LSTM features as the key–value pair, and the CNN features as the query, this we called a XA_I model; we found empirically for a properly chosen sliding window length, whichever CNN model or Bi-LSTM model performs well in the 1-step ahead prediction, then its features should be used as the query in the cross attention mechanism, thus, the XA achieves better results than XA_I on the East-China dataset, while XA_I achieves better results than XA on the Yalova dataset, and on the 16 MW dataset. Where comparable results using other state-of-the-art models are available on the Yalova dataset and the 16 MW dataset, we found that our XA_I results are superior. Moreover, it is found empirically, that the cross attention mechanism model produces stable k -step ahead predictions, for k ≤ W 2 , where W is the sliding window length. • A novel prediction model for wind generated power based on cross attention mechanism has been proposed. • The proposed method demonstrated on three practical datasets to have superior performance. • The cross attention mechanism based models give a stable multi-step ahead prediction. [ABSTRACT FROM AUTHOR]

• SL, DL and DRL models were constructed for multi performance prediction of gymnasiums under different weather conditions. • Accuracy, time, interpretability, ease-of-use, hyperparameters, and robustness of the metamodels have been investigated. • Energy consumption, temperature and CO 2 time series vectors prediction were modeling with nine training sets. • Methodology for hyperparameters comparison using exhaustive grid search and sensitivity analysis. • RDPG is most accurate but time-consuming. LSTM is more efficient and also has relatively excellent accuracy. Building performance simulation (BPS) is essential for testing energy demand and indoor environment quality of different building designs. However, software for BPS is computationally intensive and impractical to run thousands even millions of simulations for performance analysis and optimization. Especially for the large space buildings, which usually have complex forms and high energy consumption. The computational problem could be overcome by the adoption of metamodels. Most correlational research focuses on single performance prediction of a particular building, which makes the model less robust when applied to multi vectors prediction for different buildings under multiple weather conditions. This paper leveraged six metamodels, which include recurrent deterministic policy gradient (RDPG), asynchronous advantage actor-critic (A3C), long short-term memory (LSTM), convolution neural network (CNN), artificial neural network (ANN) and support vector regression (SVR), to predict hourly-based multi performance vectors of gymnasiums under various design parameters and multiple weather conditions. Six metamodels are trained and tested on a large scale of datasets simulated by EnergyPlus over four gymnasium cases in different cities of China. The accuracy, efficiency, ease-of-use, robustness and interpretability of the models are investigated. To conduct a fair and detailed comparison, a methodological approach using grid searches for model settings selection assisted by sensitivity analysis is pursued. Principal component analysis (PCA) is also adopted to interpret the work process of the metamodels. The comparison showed that the RDPG model provides the most accurate prediction results with R2 converges at 0.993, 0.982 and 0.941 for energy, temperature and CO 2 , respectively. LSTM model is more efficient than RDPG, and suitable for users who need emphasis on both time and accuracy. ANN is suitable for users with limited time and require models of ease-of-use and robustness. SVR and ANN could be used for the automatically co-simulation with BPS software. In future research, the influence of occupant behavior also should be investigated. [ABSTRACT FROM AUTHOR]

• The seasonal decomposition is used to acquire various frequency components. • The temporal neural network extracts the nonlinear features of the natural gas. • Two hybrid SDTCN models were proposed to predict the natural gas load in DHS. • Detailed experiments were conducted to validate the performance of the SDTCN. In recent years, natural gas was widely used as a primary clean energy source to replace coal-fired in northern cities in China, whose objective is to reduce the severe environmental pollution caused by coal-fired central heating in winter. The rapid growth of natural gas consumption has brought a significant burden for natural gas production and transportation, affecting residents' regular heating demand. Therefore, accurately predicting natural gas consumption is of great significance to the district heating system (DHS). However, accurately predicting natural gas consumption is challenging due to the complex nonlinear time-varying feature for the large-scale DHS system. A novel hybrid model was proposed to predict the daily natural gas consumption in the DHS based on the seasonal decomposition and temporal convolution network (SDTCN) in this paper, under the principle of Divide and Conquers strategy and deep learning algorithm. The seasonal decomposition of the natural gas consumption produces three distinct subsequences: the trend item, the seasonal item, and the residual item. The spatiotemporal features of these three subsequences are then modeled and predicted based on the TCN model, combining the advantages of recurrent neural networks (RNN) and convolution neural network (CNN) characteristics. Besides, we compare the two SDTCN models with state-of-the-art algorithms, such as support vector machine (SVM), Adaboost, extreme tree regression (ETR), passive-aggressive regression (PassAgg), nu support vector regression (NuSVR), and bootstrap aggregating (Bagging). The experimental results show that the proposed SDTCN model is superior to other algorithms, and the prediction accuracy can reach 94.4%. [ABSTRACT FROM AUTHOR]