No abstract

The icing problem of wind turbine blades in northern China has a serious impact on the normal and safe operation of the unit. In order to effectively predict the icing conditions of wind turbine blades, a deep fully connected neural network optimized by machine learning (ML) algorithms based on big data from the wind farm is proposed to diagnose the icing conditions of wind turbine blades. This study first uses the random forest model to reduce the features of the supervisory control and data acquisition (SCADA) data that affect blade icing, and then uses the K-nearest neighbor (KNN) algorithm to enhance the active power feature. The features after the random forest reduction and the active power mean square error (MSE) feature enhanced by the KNN algorithm are combined and used as the input of the fully connected neural network (FCNN) to perform and an empirical analysis for the diagnosis of blade icing. The simulation results show that the proposed model has better diagnostic accuracy than the ordinary back propagation (BP) neural network and other methods.

Summary As the installed capacity of wind power continues to increase, the problem of curtailed wind power is becoming serious in China, especially in the northern region during the winter heating season. To solve the problem of wind‐heat conflict during the heating period in the Three North area, an electric boiler with thermal storage (EBTS) is installed at the end of the grid where wind power is difficult to accommodate and using curtailed wind power to supply heat promotes local accommodation. In this paper, a multi‐objective optimization model of wind power accommodation based on the wind power–EBTS system for heating is established. The goals of maximizing wind power accommodation, minimizing the number of times EBTS must be adjusted, and minimizing operating costs are presented, and a bi‐level optimization scheme is designed. An improved multi‐objective particle swarm optimization algorithm is used to solve these functions, and an optimal compromise solution from the generated Pareto solution set is filtered using the fuzzy membership method. Based on actual data from a demonstration project in China's Jilin Province, the simulation results verify that this method can effectively reduce operating costs and improve wind power accommodation.

In-situ exploitation of oil shale by electric heating consumes large amounts of electricity. Under the existing dispatch system, using wind power output and photovoltaic power output to support the exploitation of oil shale can promote renewable energy use, reduce the consumption of coal and other fossil fuels, and protect the environment from pollution. In this study, the characteristics of the wind power and photovoltaic power output are analyzed, and the correlation between the power outputs is evaluated using the copula function. The load of exploiting oil shale is presented. In order to match the heating load characteristics of oil shale exploitation, a particle swarm optimization algorithm is used to optimize the heating temperature of the heated well to minimize the cost. An economic analysis is conducted of five different power supply combinations, including wind power, photovoltaic power, and the existing power grid. The income ratio of the five modes is calculated using actual data of a project in Jilin province in China, and the feasibility of in-situ electric heating by wind power, photovoltaic power, and the power grid is determined. The results of this study provide useful references for decision makers to plan the power supply scheme for in-situ oil shale exploitation.

Blade icing of a wind turbine will affect the startup performance of the blades, resulting in the loss of power generation of the wind turbine, and even affect the safety of production and operation. In order to reflect the blade icing state of wind turbines as truthfully and objectively as possible, this paper proposes a wind turbine blade icing state recognition model based on the combination of vine-Copula network model and Long Short-Term Memory (LSTM)-Autoencoder algorithm. First, the vine-Copula model is used to analyze the correlation between the various parameters in supervisory control and data acquisition (SCADA) system and the blade icing state, thereby constructing a high-dimensional vine-Copula structure. Then, removing the features that are not directly related to the blade icing state, the final vine-Copula model and related features are obtained. The filtered features are input into the LSTM-Autoencoder algorithm, then the “memory” function and non-linear feature extraction capabilities of the LSTM-Autoencoder algorithm are used to obtain the evaluation results of the blade icing state of wind turbines. The experimental results show that the indicators of the wind turbine blade icing state recognition based on this method are overall better than the indicators of the Recurrent Neural Network-Autoencoder algorithm without feature reduction and the LSTM-Autoencoder algorithm without feature reduction and traditional classification algorithms.