Wind power probabilistic forecasting based on combined decomposition and deep learning quantile regression

Two-stage correction prediction of wind power based on numerical weather prediction wind speed superposition correction and improved clustering

Hundreds of cascading trip faults have occurred in recent years in Chinese wind farms, leading to challenges in the secure operation of the power system. To prevent cascading trip faults, the concept of a voltage security region (VSR) has been discussed in previous studies. However, the VSR proposed in previous work cannot provide decoupled voltage control ranges for each wind farm, which would be a more promising and applicable solution for wind farms in China. In addition, with increasing wind farm active power penetration, the VSR of a wind power integration area may not exist. In this situation, the control centre must curtail some wind power to guarantee security. Thus, this study proposes an iterative VSR model to guarantee secure operation and obtain a larger VSR in wind farms considering wind power curtailment. In each wind farm, the VSR is designed to guarantee normal operation of each wind turbine generator, while in the control centre, the region is designed to guarantee normal operation both under normal conditions and N–1 contingencies considering wind power curtailment. The proposed VSR demonstrated good performance in a simple test case and in a real system in Northern China using a Monte Carlo method.

As wind farm explosively develops worldwide, the interactions between wind farms and the environment attract increasing attention [1,2].Wind farms would alter the atmosphere by lifting surface roughness [3], extracting momentum and adding extra turbulence, inducing wake [4] and even climate changes [5].Local and regional changes of wind, temperature, and precipitation have been reported [1,2,6].However, it remains unclear how wind farms influence air pollution.Here we first show that wind farms in China have significant regional impacts on both climate and air pollutants especially in summer, by using a dynamic numerical weather prediction and a multi-scale air quality model.We show that although wind power cannot produce additional emissions, it is able to redistribute air pollutants via affecting the atmospheric process.Chinese wind farms influence the mesoscale circulation in summer, which increases PM2.5 in the Beijing-Tianjin-Hebei (BTH) megalopolis with a peak of 6.76 gm -3 and reduces PM2.5 in the Yangtze River Delta (YRD) region with a peak reduction of 5.25 gm -3 in the recent years from 2015 to 2018.These numbers are equivalent to those obtained by about 1.5 years of dominant contributions from anthropogenic emission abatement to reduce the national annual mean PM2.5 concentration.It indicates that the effects of investing over 2 billion US dollars per year by the central government to reduce PM2.5 might be canceled out by the impacts from Chinese wind power in certain regions.These significant variations of climate and air pollution are of the same order as the interannual variabilities in BTH and YRD, which has never been reported.More dominant impacts can be expected in the future according to the roadmap 2050 of China National Energy Administration [7].Urgent attention from the government and industry is required to comprehensively evaluate the roadmap so as to set up better policies for sustainable development.

An improved residual-based convolutional neural network for very short-term wind power forecasting

Wind power is one of the most rapidly growing clean and renewable energy sources. However, due to the uncertainty and intermittency of wind power, the increasing penetration of wind power into the electric power system will pose challenges to power system operators. Moreover, as a participant in a competitive electricity market, a wind power producer's behavior and profit will be influenced by other participants' behaviors. This paper proposes a model of using stochastic programming to generate optimal bidding strategies to maximize the total profits of wind and conventional power producers in both the energy market and a bilateral reserve market, where the reserve price is settled between wind and conventional power producers by using game theory. Case studies using real-world data for games in an electricity market with different types of players are performed to show the effectiveness of the proposed model.

The volatility and randomness of wind power generation pose significant challenges to grid integration and dispatching. Accurate wind power forecasting is crucial for ensuring grid security and optimizing dispatching strategies. This study proposes an ultra‐short‐term wind power forecasting model based on Variational Mode Decomposition (VMD) and an adaptive weighting integration mechanism. The model first employs VMD to decompose wind power time series into multiple Intrinsic Mode Functions (IMFs) to extract latent multiscale features. Subsequently, Convolutional Neural Networks (CNN) and Long Short‐Term Memory networks (LSTM) are utilized to independently model the IMFs, capturing spatial features and temporal dependencies. The outputs of CNN and LSTM are then dynamically weighted using an attention mechanism to adaptively enhance the representation of critical features. Finally, the predictions of CNN and LSTM are integrated using a Gradient Boosting Decision Tree (GBDT) meta‐learner to produce the final output. Experimental results on wind farm data from Xinjiang, China, demonstrate that the proposed model significantly outperforms traditional and state‐of‐the‐art forecasting methods in terms of accuracy and robustness, especially in handling the nonlinear and nonstationary characteristics of wind power.

Due to the randomness and intermittency of wind power, the accurate planning and dispatching of power generation is hindered. Hydro power has a certain ability to regulate, the coordination between wind farms and hydro power plants may be mutually beneficial. Wind power and hydro power collaborative optimization operation model considering the ladder electricity price is established, and the chaotic particle swarm optimization (CPSO) algorithm is used to solve the model. A case study is used to verify the reliability of the model and the feasibility of coordination strategy.

This paper presents a hybrid model constituting dynamic smoothing technique and particle swarm optimization techniques to optimally size and control battery energy storage systems for wind energy ramp rate control and power system frequency performance enhancement. In today’s modern power system, a high-proportion renewable energy grid is inevitable. This high-proportion renewable energy grid is a power system with abundant integration of renewable energy resources under the presence of energy storage tools. Energy storage tools are integrated into such power systems to balance the fluctuation and intermittence of renewable energy sources. One of the requirements in a high-proportion renewable energy grid is the fractional power balance between generation and load. One of the requirements set by power system regulators is the generation variation between two time points. A power producer is mandated to satisfy the ramp rate requirement set by the grid owner. This paper proposes dynamic smoothing techniques for initial size determination and particle swarm optimization based on optimal sizing and control of battery energy storage systems for ramp rate control and frequency regulation performance of a power system integrated with a large percentage of wind energy systems. Wind energy data taken from Zhangjiakou wind farm in China are used. The results indicate that the battery energy storage system improves the ramp rate characteristics of the wind farm. In addition, the virtual inertia capability of the battery energy storage system enabled the transient and steady-state frequency response of the test power system to improve significantly.

Overview of wind power intermittency: Impacts, measurements, and mitigation solutions

Distributed onshore wind farm siting using intelligent optimization algorithm based on spatial and temporal variability of wind energy

K-nearest neighbors and a kernel density estimator for GEFCom2014 probabilistic wind power forecasting

A novel ensemble probabilistic forecasting system for uncertainty in wind speed

Wind energy has increasingly played a vital role in mitigating conventional resource shortages. Nevertheless, the stochastic nature of wind poses a great challenge when attempting to find an accurate forecasting model for wind power. Therefore, precise wind power forecasts are of primary importance to solve operational, planning and economic problems in the growing wind power scenario. Previous research has focused efforts on the deterministic forecast of wind power values, but less attention has been paid to providing information about wind energy. Based on an optimal Adaptive-Network-Based Fuzzy Inference System (ANFIS) and Singular Spectrum Analysis (SSA), this paper develops a hybrid uncertainty forecasting model, IFASF (Interval Forecast-ANFIS-SSA-Firefly Alogorithm), to obtain the upper and lower bounds of daily average wind power, which is beneficial for the practical operation of both the grid company and independent power producers. To strengthen the practical ability of this developed model, this paper presents a comparison between IFASF and other benchmarks, which provides a general reference for this aspect for statistical or artificially intelligent interval forecast methods. The comparison results show that the developed model outperforms eight benchmarks and has a satisfactory forecasting effectiveness in three different wind farms with two time horizons.

A short-term wind power prediction model based on BiLSTM–CNN–WGAN-GP (LCWGAN-GP) is proposed in this paper, aiming at the problems of instability and low prediction accuracy of short-term wind power prediction. Firstly, the original wind energy data are decomposed into subsequences of natural mode functions with different frequencies by using the variational mode decomposition (VMD) algorithm. The VMD algorithm relies on a decision support system for the decomposition of the data into natural mode functions. Once the decomposition is performed, the nonlinear and dynamic behavior are extracted from each natural mode function. Next, the BiLSTM network is chosen as the generation model of the generative adversarial network (WGAN-GP) to obtain the data distribution characteristics of wind power’s output. The convolutional neural network (CNN) is chosen as the discrimination model, and the semi-supervised regression layer is utilized to design the discrimination model to predict wind power. The minimum–maximum game is formed by the BiLSTM and CNN network models to improve the quality of sample generation and further improve the prediction accuracy. Finally, the actual data of a wind farm in Jiuquan City, Gansu Province, China is taken as an example to prove that the proposed method has superior performance compared with other prediction algorithms.

Short-term wind power prediction based on multidimensional data cleaning and feature reconfiguration