Wind Power Grid Research Articles

Structures, Submodule Topologies and Control Strategies for Modular Multilevel Converter

Global energy demand is rapidly increasing, and the limitations and disadvantages of traditional energy sources are becoming more evident. Traditional energy sources are finite, and their sustainability is facing significant challenges. Compared to traditional multi-level converters, modular multi-level converters (MMC) adopt a modular design that divides the entire converter system into several independent sub-modules, each responsible for converting specific voltage levels. This article provides an overview of the theoretical aspects, topology structures, basic operation principles, and output characteristics of multi-level converters. This paper reviews the characteristics and application ranges of different topology structures, highlighting the differences in response speed, output quality, and control complexity. This paper also discusses the advantages and limitations of multi-level converters in various application scenarios. Explored the application of multi-level converters in wind power grid integration and electric vehicle charging systems, emphasizing their energy-saving and environmentally friendly features. In conclusion, this paper emphasizes the critical role of multi-level converters in power systems and future energy applications, stressing the importance of their reliability and promising future developments.

An innovative interpretable combined learning model for wind speed forecasting

Wind energy is taken as one of the most potential green energy sources, whose accurate and stable prediction is important to improve the efficiency of wind turbines as well as to guarantee the power balance and economic dispatch of power systems and equipment safety. However, the random and fluctuating nature of wind speed poses a great risk to wind power grid connections. To address the issues of low prediction performance and lack of interpretable analysis in most past studies, this research proposes an interpretable combined learning model for wind speed time series prediction by combining linear models, different neural networks, and deep learning by introducing interpretable TFT models. To test the effectiveness of the forecasting models, the presented combined model is verified using eight wind speed datasets covering four seasons collected from two wind farms in Shaanxi, China. The experimental results show that the average root mean squared error of the one-step, two-step, and three-step predictions on the eight datasets for proposed model are 0.3448, 0.4586 and 0.6164, respectively, which are much better than the six single models and the six combined models with different strategies. And proposed model outperforms the single model and combined model in most cases, with 86.80% and 92.01% of the DM values greater than the corresponding critical values when the significance level is set to 0.01 and 0.1, respectively. Finally, the proposed model is discussed and analyzed in depth through interpretability analysis of the combined model, which further validates the potential of the model and also provides a reference for other time series forecasting studies.

The univariate model for long-term wind speed forecasting based on wavelet soft threshold denoising and improved Autoformer

Accurate wind speed prediction is crucial for effective wind power grid integration and energy dispatching. Recent research has explored the combination of decomposition algorithms with forecast models to form hybrid models, aiming to enhance wind speed prediction accuracy. However, these traditional decomposition techniques often lead to high time costs in practical applications, as they require new wind speed sequences to be appended to historical long sequences for decomposition before entering the forecast model. To overcome this challenge, this study introduces and improves the Autoformer model, applying it for the first time to long-term univariate wind speed forecasting. By incorporating decomposition technology as a sub-module of the forecast model, Autoformer not only solves the high time cost issue found in conventional hybrid models but also retains the benefits of decomposition technology in time series processing.Furthermore, in this paper, the decomposition module of Autoformer is replaced with the Mixture of Expert Decomposition Module (MOEDecomp) to better extract complex trend elements of wind speed series. Combined with the auto-correlation mechanism, sequential attention is paid to wind speed for extracting time dependencies in long series. Additionally, the Wavelet Soft Threshold Denoising (WSTD) algorithm is utilised for noise reduction in wind speed sequences. To evaluate the model's performance, two multi-step forecasting strategies were employed to predict wind speeds for the forthcoming 24, 48, 72, 96, and 120 h using four datasets. Experimental results demonstrate that the proposed model surpasses 15 comparative models in terms of prediction accuracy, generalization ability, and handling uncertain data.

Two-stage Robust Optimization Scheduling Considering Wind Uncertainty and Multi-area Frequency Dynamic Safety

The large-scale wind power grid connection exacerbates the problem of uneven distribution of inertia between different regions, posing challenges to the frequency security defense of multi-area systems. This article proposes a two-stage robust optimization scheduling model that considers both the uncertainty of wind power output and the dynamic safety of multi-area frequencies. Firstly, according to the conditional value at risk theory, the operating cost and operating risk of the multi-region system are optimized, and the acceptable safety boundary of wind power is obtained. Secondly, multi-zone frequency dynamic safety verification is carried out in harsh wind power scenarios. Decouple the reserve capacity into non-accident reserve and accident reserve, and nest them separately in a two-stage model to achieve clear solution ideas. Finally, the two-stage optimization problem is solved through the column and constraint generation algorithm. The simulation results of numerical examples show that the model and method can effectively improve the frequency response capability of the multi-region system, so as to ensure the safe and economical operation of the system.

An advanced multi-objective collaborative scheduling strategy for large scale EV charging and discharging connected to the predictable wind power grid

With the large-scale EV connected into the wind power grid, the intermittent, fluctuating and stability bring rigorous challenges for power quality and dispatch, thus wind curtailment becomes more serious. This paper proposes an advanced multi-objective collaborative scheduling strategy to improve the accuracy of the predictable wind grid and ensure the stable operation of the grid by optimizing wind power and EV synergistically. Firstly, a hybrid power prediction method is proposed based on density-based spatial clustering applied to noise (DBSCAN) and partial least squares regression (PLSR) combined with long and short-term memory neural network (LSTM). It establishes a linear component prediction mode to predict the nonlinear wind power efficiently. Secondly, a multi-objective optimization scheduling model is established according to the predicted wind power, considering the vehicle-to-grid (V2G) characteristics of EV, grid load fluctuations, wind curtailment capacity, EV charging costs, and grid losses. The global control aims to achieve overall energy balance and optimize the charging and discharging of EV in a time dimension under constrained conditions. Additionally, based on the total number of EV charging and discharging determined by the global level control, a spatial scheduling optimization is conducted for EV and wind power by the distribution network control, with the goal of minimizing network losses in the distribution network system. Finally, through simulation and comparison of four scenarios, the results demonstrate that the proposed method achieves synergistic optimization of wind power and EV. This paper provides an efficient solution for the optimized scheduling of large-scale EV connected to the predictable wind power grid.

Analysis of energy control system in wind farm

This article first introduces the structure and principle of two mainstream wind turbine models, analyses the different power generation principles and energy transfer structures of these two different wind turbines, and briefly introduces the network topology structure of wind power stations, as well as the direction of command data from the control end to the controlled end. Then, it introduces and analyses several widely used maximum power point tracking (MPPT) control algorithms and frequency control methods, summarizes and predicts these algorithms for single wind turbines, and lays the foundation for future research. Large-scale wind power and other new energy grid connection is one of the most effective solutions to address the world energy crisis. With continuous development of technology, more and more power electronic control components are being closely integrated with traditional power systems, and we are facing problems different from those in the past. We must change our thinking and work hard to address these issues.

GCNInformer: A combined deep learning model based on GCN and Informer for wind power forecasting

AbstractWind power energy is green, clean, and renewable, which is random and volatile. The integration of unstable wind energy severely threatens the security and constant operation of the power system. The need to enhance the reliability of wind power grid integration, mitigate the impact of wind power uncertainty, and develop a robust prediction model has become a pressing issue. However, only some people have considered the correlations among the power of multiple adjacent wind turbine arrays. In this paper, we propose GCNInformer to construct these relationships. Furthermore, we analyze the relationships among multiple features of individual wind turbines. GCNInformer is composed of two main components. The first component employs a graph convolutional network (GCN) to establish relationships among multiple wind turbine arrays, enhancing the correlation of the data. The second part employs Informer to extract the time information from the data and predict long‐term sequences. For training and testing, GCNInformer utilizes two data sets: Data_CQ and Data_DL. The evaluation of the model's performance is conducted using various metrics such as mean absolute percentage error, mean absolute error, root mean square error, and mean square error. Numerous experimental findings have validated the effectiveness of the GCNInformer.

Large-scale wind power grid integration challenges and their solution: a detailed review.

Despite global warming, renewable energy has gained much interest worldwide due to its ability to generate large-scale energy without emitting greenhouse gases. The availability and low cost of wind energy and its high efficiency and technological advancements make it one of the most promising renewable energy sources. Hence, capturing large amounts of wind energy is essential today. The large-scale integration of wind power sources must be evaluated and mitigated to develop a sustainable future power system. Wind energy research and the government are working together to overcome the potential barriers associated with its penetration into the power grid. This paper reviews the social, environmental, and cost-economic impacts of installing large-scale wind energy plants. A valuable review of wind energy technology and its challenges is also presented in this paper, including the effects of wind farms on nearby communities, generation uncertainty, power quality issues, angular and voltage stability, reactive power support, and fault ride-through capability. Besides, socioeconomic, environmental, and electricity market challenges due to the grid integration of wind power are also investigated. Finally, potential technical challenges to integrating large-scale wind energy into the power grid are reviewed regarding current research and their available mitigation techniques.

Short-term wind speed forecasting based on a hybrid model of ICEEMDAN, MFE, LSTM and informer.

Wind energy, as a kind of environmentally friendly renewable energy, has attracted a lot of attention in recent decades. However, the security and stability of the power system is potentially affected by large-scale wind power grid due to the randomness and intermittence of wind speed. Therefore, accurate wind speed prediction is conductive to power system operation. A hybrid wind speed prediction model based on Improved Complete Ensemble Empirical Mode Decomposition with Adaptive Noise (ICEEMDAN), Multiscale Fuzzy Entropy (MFE), Long short-term memory (LSTM) and INFORMER is proposed in this paper. Firstly, the wind speed data are decomposed into multiple intrinsic mode functions (IMFs) by ICEEMDAN. Then, the MFE values of each mode are calculated, and the modes with similar MFE values are aggregated to obtain new subsequences. Finally, each subsequence is predicted by informer and LSTM, each sequence selects the one with better performance than the two predictors, and the prediction results of each subsequence are superimposed to obtain the final prediction results. The proposed hybrid model is also compared with other seven related models based on four evaluation metrics under different prediction periods to verify its validity and applicability. The experimental results indicate that the proposed hybrid model based on ICEEMDAN, MFE, LSTM and INFORMER exhibits higher accuracy and greater applicability.

Research on the Stability of Grid Connected Wind Turbine Combined with Energy Storage Power System Based on a New Wind Power Fluctuation Smoothing Strategy

Wind power equipped with an energy storage system (ESS) has been demonstrated as the best potential configuration for a rapid global energy transition in the future. In the traditional anti-pulse average filtering control strategy for controlling wind power fluctuations, the smoothing window interval is a constant value, when the forecast output power error is considerable, the fixed window interval width may not be able to meet the smoothing demand, so the traditional strategy has the problems of bad smoothing effect and high probability for power fluctuations exceeding the limit. We propose an optimization strategy for energy storage operation in this paper. Firstly, this strategy can dynamically adjust the window interval width according to the power prediction error, and find the optimal window interval width that makes the output power volatility satisfy requirements for grid connection, so as to achieve the effect of improving the smoothing effect. Secondly, by optimizing hydrogen storage systems operation to reduce the demand for storage system capacity, the odds of output power volatility exceeding the limits are reduced. The outcomes indicate that this new smoothing strategy is effective in improving the stability of the wind power grid connection.

An Optimal Configuration Model for Supercapacitor Capacity to Suppress Wind Power Fluctuations

This paper presents an optimization configuration scheme for energy storage capacity by taking into account the operational characteristics of supercapacitors. The scheme utilizes Empirical Mode Decomposition (EMD) to address power fluctuations during the integration of wind power into the grid. To begin with, the wind power data undergoes EMD to decompose it into different-order intrinsic mode functions. These functions are then reconstructed as low-frequency components and high-frequency components. The low-frequency components are directly connected to the grid, while the high-frequency components are employed for power smoothing through energy storage. Subsequently, the impact of various wind power grid limits on the directly connected power is compared to identify the optimal value. A model is established to configure the capacity of supercapacitors, aiming to mitigate wind power fluctuations. The model considers an objective function that minimizes the sum of energy storage investment and operational maintenance costs, along with the compensation cost associated with wind power fluctuation opportunities. Finally, the proposed model’s effectiveness and economic feasibility are validated through case studies. This ensures that the model’s efficacy in reducing power fluctuations and its viability from a financial standpoint are thoroughly examined.

Fast Frequency Control Strategy for Direct Drive Wind Turbines Considering the Suppression of Secondary Frequency Drop in the Power Grid

Inertia and damping in the power system are important parameters in frequency regulation and oscillation suppression. Inertia will reduce the rate and amplitude of frequency changes after active power sudden changes. Damping in the system can effectively suppress low-frequency oscillations after grid disturbances. As the penetration rate of wind turbines increases, it will significantly reduce the effective inertia in the power grid and reduce the friendliness of wind power grid connection. To provide better frequency support for the power grid in the event of large disturbances in the power system, this paper proposes a fast frequency control strategy for wind farms that considers the suppression of secondary frequency dips in the power grid. The strategy combines frequency droop control with grid frequency bias control and evaluates the kinetic energy reserve online based on wind power prediction and hourly speed, thus balancing the frequency regulation and suppression of secondary frequency dips in the power grid. A 2.5 MW direct drive wind turbine simulation model was built based on the PSCAD simulation platform, verifying the effectiveness of the proposed comprehensive control strategy.

Wind Power Group Prediction Model Based on Multi-Task Learning

Large-scale wind power grid connection increases the uncertainty of the power system, which reduces the economy and security of power system operations. Wind power prediction technology provides the wind power sequence for a period of time in the future, which provides key technical support for the reasonable development of the power generation plan and the arrangement of spare capacity. For large-scale wind farm groups, we propose a cluster model of wind power prediction based on multi-task learning, which can directly output the power prediction results of multiple wind farms. Firstly, the spatial and temporal feature matrix is constructed based on the meteorological forecast data provided by eight wind farms, and the dimensionality of each attribute is reduced by the principal component analysis algorithm to form the spatial fusion feature set. Then, a network structure with bidirectional gated cycle units is constructed, and a multi-output network structure is designed based on the Multi-gate Mixture-of-Experts (MMoE) framework to design the wind power group prediction model. Finally, the data provided by eight wind farms in Jilin, China, was used for experimental analysis, and the predicted average normalized root mean square error is 0.1754, meaning the prediction precision meets the scheduling requirement, which verifies the validity of the wind power prediction model.

Integrated Reactive Power Optimisation for Power Grids Containing Large-Scale Wind Power Based on Improved HHO Algorithm

Large-scale wind power grid integration will greatly change the system current distribution, making it difficult for the reactive power regulator to adjust to the optimal state. In this paper, an integrated reactive power optimisation method based on the improved Harris Hawk (HHO) algorithm is proposed. Firstly, a reactive power regulation model is constructed to solve the reactive power regulation interval of wind turbines, and the reactive power margin of wind turbines is used to participate in the system’s reactive power optimisation. Finally, a reactive power compensation capacity allocation optimisation model considering nodal voltage deviation, line loss and equipment investment cost, is established, and a reactive power optimisation scheme is obtained using the Harris Hawk optimisation algorithm on the basis of considering the constraints of the wind turbine reactive power output interval. The improved HHO algorithm is used to solve the reactive power optimisation scheme considering the constraints of tidal power, machine end voltage, a conventional generator and wind farm reactive power. In the simulation, the effects of the improved Harris Hawk optimisation algorithm and the particle swarm optimisation algorithm are compared, and the experimental results prove that compared to the particle swarm algorithm, the optimisation result of the improved Harris Hawk optimisation algorithm reduces the average loss of the system by 42.6% and reduces the average voltage deviation by 30.3%, which confirms that the improved Harris Hawk intelligent optimisation algorithm is effective in proving its superiority and solving the multi-objective model for reactive power optimisation.

Deep Learning for Variable Renewable Energy: A Systematic Review

In recent years, both fields, AI and VRE, have received increasing attention in scientific research. Thus, this article’s purpose is to investigate the potential of DL-based applications on VRE and as such provide an introduction to and structured overview of the field. First, we conduct a systematic literature review of the application of Artificial Intelligence (AI), especially Deep Learning (DL), on the integration of Variable Renewable Energy (VRE). Subsequently, we provide a comprehensive overview of specific DL-based solution approaches and evaluate their applicability, including a survey of the most applied and best suited DL architectures. We identify ten DL-based approaches to support the integration of VRE in modern power systems. We find (I) solar PV and wind power generation forecasting, (II) system scheduling and grid management, and (III) intelligent condition monitoring as three high potential application areas.

Low-carbon economic dispatch of integrated energy system containing electric hydrogen production based on VMD-GRU short-term wind power prediction

The integration of energy systems (IES) enhances the interaction between the electricity, gas, and heat systems, and the concept of low-carbon development can further reduce the carbon emissions of IES. However, the uncertainty of wind power output and the complexity of different energy chains pose significant challenges to the low-carbon operation of IES. Therefore, this paper proposes a low-carbon economic dispatching strategy for IES containing electric hydrogen production based on short-term wind power prediction. Firstly, a variational mode decomposition- gate recurrent unit network prediction model is employed to enhance the accuracy of ultra-short wind power forecasting, which reduces the impact of wind power output uncertainty on wind power grid connection. Secondly, a refined two-stage P2G operation model for refining the electric hydrogen production device is constructed to decrease the carbon emissions of IES. Thirdly, a thermoelectric integrated demand response model is established to adjust the heat and power demand proportion, further reducing the IES's carbon emissions. Finally, the case study is performed based on IEEE standard system to verify the effectiveness of the proposed strategy. Simulation analysis shows that introducing electrolysis for hydrogen production can reduce IES carbon emissions by 12.90%. In addition, introducing an adjustable thermal-electric comprehensive demand response can reduce IES carbon emissions by 1.543% while lowering the overall cost by 5.24%. The proposed strategies can simultaneously consider the low-carbon and economic aspects of IES.

A design of new wind power forecasting approach based on IVMD-WSA-IC-LSTM model

The wind power forecasting (WPF) technology can reduce the adverse impact of wind power grid connection. Based on the characteristics of wind power data, an algorithm based on improved variational mode decomposition (IVMD) and long short-term memory (LSTM) Network is proposed to predict the wind power, and hyper parameter optimization search of LSTM using Whale Swarm Algorithm with Iterative Counter (WSA-IC). Firstly, through correlation analysis, the characteristics of 10 different wind power data are screened, and two kinds of data with large correlation with wind power are determined as input of the mode. Secondly, IVMD is used to calculate the maximum envelope kurtosis, determine the best decomposition parameters of the variational mode decomposition (VMD), and the original wind power and wind speed sequences are decomposed to obtain the IMF with different time scales. Finally, to address the problems of difficult optimization of hyper parameter and difficulty in obtaining optimal solutions for LSTM neural network modes, the WSA-IC algorithm is proposed to optimize its key hyper parameter, and the IVMD-WSA-IC-LSTM forecasting mode is established to obtain the short-term forecasting results of wind power. The algorithm is tested with the data of China Longyuan Power Group Corporation Limited. Compared with other common forecasting approaches using same data, the mean absolute error (MAE) of the forecasting approach is reduced to 0.007859, the mean square error (MSE) is reduced to 0.00011, and the determination coefficient is improved to 0.998828, which has higher forecasting accuracy.

Design of a Wind Power Forecasting System Based on Deep Learning

In recent years, wind energy, as a ubiquitous, easy-to-capture cost-effective clean energy, is accounting for a sharp increase in the installed capacity of China’s new energy grid. However, its stochastic volatility has always brought challenges to the generation scheduling of wind farms. To better optimize the energy management level of wind farms and improve the stability of wind power grid connection, this paper proposes a design of a wind power forecasting system based on deep learning. Our system architecture mainly includes data pre-processing, power prediction, and data application modules. The data pre-processing module will correct the numerical weather forecast (NWP) updated twice daily to get the wind turbine generator (WTG) hub height weather forecast and resample the SCADA data in a 15-minute time resolution. The power prediction module will periodically perform the ultra-short-term and short-term forecasting results and record them in the database. On one hand, the data application module will present the power prediction results and SCADA data to the web page for display. On the other hand, it will provide data to the energy dispatching department for reference. The experiment shows the effect of using a residual channel attention network (RCAN) to correct the NWP and the influence of two different RNN cores, including GRU and LSTM, on the prediction of ultra-short-term wind power results of the DeepAR model. The experimental results show that the proposed RCAN can effectively correct the results of the numerical weather forecast to a single WTG, and the DeepAR model using the GRU core can achieve better performance in the test set than the DeepAR model using the LSTM core. Thus, we choose GRU as the core of the model of our power prediction module.

Research on the side converter system of wind power grid based on fractional LCL filter

AbstractThe power electronic converters and grid‐connected filters were important components of the permanent magnet direct drive wind power generation system whose performance directly determines the quality of wind power generation. In past modeling, analysis, and control studies, capacitive and inductive components were often treated as integral‐order components. However, the inductance and capacitance components in the actual permanent magnet direct drive wind generator were fractional‐order components, and their electrical characteristics will change with the change of order, which had an important impact on the dynamic and static characteristics of the system. In this paper, the mathematical model of the fractional‐order LCL (FOLCL) filter was derived. Through simulation, it could be seen that the FOLCL filter can avoid resonance fundamentally. At the same time, the fractional‐order PI (FOPI) controller was introduced into the machine‐side current converter, and the parameters of the FOPI controller of the outer speed loop and the inner current loop were adjusted by using the time‐domain optimization method. The results showed that the efficiency of the FOPI controller was significantly better than that of the integer‐order PI controller in realizing maximum wind energy capture. It provides theoretical support and practical application value for the stable operation of the wind power generation system.

Nonlinear operational optimization of an industrial power-to-heat system with a high temperature heat pump, a thermal energy storage and wind energy

The heat demand for industrial processes is often provided in the form of steam generated by various fossil fueled equipment. In order to reduce CO2 emissions, the heat demand has to be covered by renewable energy sources. Electrified steam generation relies on complex energy systems, that can be operated according to energy availability and cost developments. However, such a multi component industrial energy system poses a challenge in modeling and determining the cost- or emission-optimal operation of the system. This study develops a methodology to model a multi component industrial energy system on the basis of a case study. By optimal system operation, either costs or emissions are minimized in response to fluctuating renewable wind energy and electricity prices.A high temperature heat pump (HTHP), a sensible thermal energy storage (TES) and a wind turbine are combined to create an electrified energy system to supply super-heated steam. During periods of low wind speed, additional grid electricity is purchased to ensure a steady heat supply. The HTHP offers a high operational flexibility and thus, enables the charging and discharging of the TES. A model of the closed reverse Brayton cycle HTHP, which is able to simulate part load behavior, is created in a process simulation software and consolidated in nonlinear surrogate models. The component behavior of a TES is represented by a combination of equations based on heat exchanger relations. Finally, the resulting algebraic nonconvex, nonlinear optimization problem based on the proposed system is solved using the local interior point optimizer (IPOPT) solver equipped with a multi-start approach to determine an optimal operation over a reference week with respect to the current wind power generation, grid emissions and electricity prices.The results of the optimization show, that optimal operating strategies enable a high potential to decarbonize future industries at minimum operational costs or emissions.