A Novel Hybrid Ensemble Learning Approach for Enhancing Accuracy and Sustainability in Wind Power Forecasting

Accurate wind power forecasting is critical for system integration and stability as renewable energy reliance grows.Traditional approaches frequently struggle with complex data and non-linear connections. This article presentsa novel approach for hybrid ensemble learning that is based on rigorous requirements engineering concepts.The approach finds significant parameters influencing forecasting accuracy by evaluating real-time Modern-EraRetrospective Analysis for Research and Applications (MERRA2) data from several European Wind farms usingin-depth stakeholder research and requirements elicitation. Ensemble learning is used to develop a robust model,while a temporal convolutional network handles time-series complexities and data gaps. The ensemble-temporalneural network is enhanced by providing different input parameters including training layers, hidden and dropoutlayers along with activation and loss functions. The proposed framework is further analyzed by comparing stateof-the-art forecasting models in terms of Root Mean Squared Error (RMSE) and Mean Absolute Error (MAE),respectively. The energy efficiency performance indicators showed that the proposed model demonstrates errorreduction percentages of approximately 16.67%, 28.57%, and 81.92% for MAE, and 38.46%, 17.65%, and 90.78%for RMSE for MERRAWind farms 1, 2, and 3, respectively, compared to other existingmethods. These quantitativeresults show the effectiveness of our proposed model with MAE values ranging from 0.0010 to 0.0156 and RMSEvalues ranging from 0.0014 to 0.0174. This work highlights the effectiveness of requirements engineering in windpower forecasting, leading to enhanced forecast accuracy and grid stability, ultimately paving the way for moresustainable energy solutions.

Primary frequency control considering communication delay for grid-connected offshore wind power systems

Offshore wind farms are becoming increasingly distant from onshore centralized control centers,and the communication delays between them inevitably introduce time delays in the measurement signal of the primary frequency control.This causes a deterioration in the performance of the primary frequency control and,in some cases,may even result in frequency instability within the power system.Therefore,a frequency response model that incorporates communication delays was established for power systems that integrate offshore wind power.The Padéapproximation was used to model the time delays,and a linearized frequency response model of the power system was derived to investigate the frequency stability under different time delays.The influences of the wind power proportion and frequency control parameters on the system frequency stability were explored.In addition,a Smith delay compensation control strategy was devised to mitigate the effects of communication delays on the system frequency dynamics.Finally,a power system incorporating offshore wind power was constructed using the MATLAB/Simulink platform.The simulation results demonstrate the effectiveness and robustness of the proposed delay compensation control strategy.

Generation of input spectrum for electrolysis stack degradation test applied to wind power PEM hydrogen production

Hydrogen production by proton exchange membrane electrolysis has good fluctuation adaptability,making it suitable for hydrogen production by electrolysis in fluctuating power sources such as wind power.However,current research on the durability of proton exchange membrane electrolyzers is insufficient.Studying the typical operating conditions of wind power electrolysis for hydrogen production can provide boundary conditions for performance and degradation tests of electrolysis stacks.In this study,the operating condition spectrum of an electrolysis stack degradation test cycle was proposed.Based on the rate of change of the wind farm output power and the time-averaged peak-valley difference,a fluctuation output power sample set was formed.The characteristic quantities that played an important role in the degradation of the electrolysis stack were selected.Dimensionality reduction of the operating data was performed using principal component analysis.Clustering analysis of the data segments was completed using an improved Gaussian mixture clustering algorithm.Taking the annual output power data of wind farms in Northwest China with a sampling rate of 1 min as an example,the cyclic operating condition spectrum of the proton-exchange membrane electrolysis stack degradation test was constructed.After preliminary simulation analysis,the typical operating condition proposed in this paper effectively reflects the impact of the original curve on the performance degradation of the electrolysis stack.This study provides a method for evaluating the degradation characteristics and system efficiency of an electrolysis stack due to fluctuations in renewable energy.

Research on the IL-Bagging-DHKELM Short-Term Wind Power Prediction Algorithm Based on Error AP Clustering Analysis

With the continuous advancement of China’s“peak carbon dioxide emissions and Carbon Neutrality”process,the proportion of wind power is increasing.In the current research,aiming at the problem that the forecasting model is outdated due to the continuous updating of wind power data,a short-term wind power forecasting algorithm based on Incremental Learning-Bagging Deep Hybrid Kernel Extreme Learning Machine(IL-Bagging-DHKELM)error affinity propagation cluster analysis is proposed.The algorithm effectively combines deep hybrid kernel extreme learning machine(DHKELM)with incremental learning(IL).Firstly,an initial wind power prediction model is trained using the Bagging-DHKELM model.Secondly,Euclidean morphological distance affinity propagation AP clustering algorithm is used to cluster and analyze the prediction error of wind power obtained from the initial training model.Finally,the correlation between wind power prediction errors and Numerical Weather Prediction(NWP)data is introduced as incremental updates to the initial wind power prediction model.During the incremental learning process,multiple error performance indicators are used to measure the overall model performance,thereby enabling incremental updates of wind power models.Practical examples show the method proposed in this article reduces the root mean square error of the initial model by 1.9 percentage points,indicating that this method can be better adapted to the current scenario of the continuous increase in wind power penetration rate.The accuracy and precision of wind power generation prediction are effectively improved through the method.

Review of the Analysis and Suppression for High-frequency Oscillations of the Grid-connected Wind Power Generation System

High-frequency oscillation(HFO)of gridconnected wind power generation systems(WPGS)is one of the most critical issues in recent years that threaten the safe access of WPGS to the grid.Ensuring the WPGS can damp HFO is becoming more and more vital for the development of wind power.The HFO phenomenon of wind turbines under different scenarios usually has different mechanisms.Hence,engineers need to acquire the working mechanisms of the different HFO damping technologies and select the appropriate one to ensure the effective implementation of oscillation damping in practical engineering.This paper introduces the general assumptions of WPGS when analyzing HFO,systematically summarizes the reasons for the occurrence of HFO in different scenarios,deeply analyses the key points and difficulties of HFO damping under different scenarios,and then compares the technical performances of various types of HFO suppression methods to provide adequate references for engineers in the application of technology.Finally,this paper discusses possible future research difficulties in the problem of HFO,as well as the possible future trends in the demand for HFO damping.

The Correlation between the Power Quality Indicators and Entropy Production Characteristics of Wind Power+Energy Storage Systems

Power quality improvements help guide and solve the problems of inefficient energy production and unstable power output in wind power systems.The purpose of this paper is mainly to explore the influence of different energy storage batteries on various power quality indicators by adding different energy storage devices to the simulated wind power system,and to explore the correlation between systementropy generation and various indicators,so as to provide a theoretical basis for directly improving power quality by reducing loss.A steady-state experiment was performed by replacing the wind wheel with an electric motor,and the output power qualities of the wind power systemwith andwithout energy storagewere compared and analyzed.Moreover,the improvement effect of different energy storage devices on various indicatorswas obtained.Then,based on the entropy theory,the loss of the internal components of the wind power system generator is simulated and explored by Ansys software.Through the analysis of power quality evaluation indicators,such as current harmonic distortion rate,frequency deviation rate,and voltage fluctuation,the correlation between entropy production and each evaluation indicator was explored to investigate effective methods to improve power quality by reducing entropy production.The results showed that the current harmonic distortion rate,voltage fluctuation,voltage deviation,and system entropy production are positively correlated in the tests and that the power factor is negatively correlated with system entropy production.In the frequency range,the frequency deviationwas not significantly correlated with the systementropy production.

Optimal Scheduling Strategy of Source-Load-Storage Based onWind Power Absorption Benefit

In recent years,the proportion of installed wind power in the three north regions where wind power bases are concentrated is increasing,but the peak regulation capacity of the power grid in the three north regions of China is limited,resulting in insufficient local wind power consumption capacity.Therefore,this paper proposes a two-layer optimal scheduling strategy based on wind power consumption benefits to improve the power grid’s wind power consumption capacity.The objective of the uppermodel is tominimize the peak-valley difference of the systemload,which ismainly to optimize the system load by using the demand response resources,and to reduce the peak-valley difference of the system load to improve the peak load regulation capacity of the grid.The lower scheduling model is aimed at maximizing the system operation benefit,and the scheduling model is selected based on the rolling schedulingmethod.The load-side schedulingmodel needs to reallocate the absorbed wind power according to the response speed,absorption benefit,and curtailment penalty cost of the two DR dispatching resources.Finally,the measured data of a power grid are simulated by MATLAB,and the results show that:the proposed strategy can improve the power grid’s wind power consumption capacity and get a large wind power consumption benefit.

A Wind Power Prediction Framework for Distributed Power Grids

To reduce carbon emissions,clean energy is being integrated into the power system.Wind power is connected to the grid in a distributed form,but its high variability poses a challenge to grid stability.This article combines wind turbine monitoring data with numerical weather prediction(NWP)data to create a suitable wind power prediction framework for distributed grids.First,high-precision NWP of the turbine range is achieved using weather research and forecasting models(WRF),and Kriging interpolation locates predicted meteorological data at the turbine site.Then,a preliminary predicted power series is obtained based on the fan’s wind speed-power conversion curve,and historical power is reconstructed using variational mode decomposition(VMD)filtering to form input variables in chronological order.Finally,input variables of a single turbine enter the temporal convolutional network(TCN)to complete initial feature extraction,and then integrate the outputs of all TCN layers using Long Short Term Memory Networks(LSTM)to obtain power prediction sequences for all turbine positions.The proposed method was tested on a wind farm connected to a distributed power grid,and the results showed it to be superior to existing typical methods.

Research on the Control Strategy of Micro Wind-Hydrogen Coupled System Based on Wind Power Prediction and Hydrogen Storage System Charging/Discharging Regulation

This paper addresses the micro wind-hydrogen coupled system,aiming to improve the power tracking capability of micro wind farms,the regulation capability of hydrogen storage systems,and to mitigate the volatility of wind power generation.A predictive control strategy for the micro wind-hydrogen coupled system is proposed based on the ultra-short-term wind power prediction,the hydrogen storage state division interval,and the daily scheduled output of wind power generation.The control strategy maximizes the power tracking capability,the regulation capability of the hydrogen storage system,and the fluctuation of the joint output of the wind-hydrogen coupled system as the objective functions,and adaptively optimizes the control coefficients of the hydrogen storage interval and the output parameters of the system by the combined sigmoid function and particle swarm algorithm(sigmoid-PSO).Compared with the real-time control strategy,the proposed predictive control strategy can significantly improve the output tracking capability of the wind-hydrogen coupling system,minimize the gap between the actual output and the predicted output,significantly enhance the regulation capability of the hydrogen storage system,and mitigate the power output fluctuation of the wind-hydrogen integrated system,which has a broad practical application prospect.

Source-Load Coordinated Optimal Scheduling Considering the High Energy Load of Electrofused Magnesium and Wind Power Uncertainty

In fossil energy pollution is serious and the“double carbon”goal is being promoted,as a symbol of fresh energy in the electrical system,solar and wind power have an increasing installed capacity,only conventional units obviously can not solve the new energy as the main body of the scheduling problem.To enhance the systemscheduling ability,based on the participation of thermal power units,incorporate the high energy-carrying load of electro-melting magnesiuminto the regulation object,and consider the effects on the wind unpredictability of the power.Firstly,the operating characteristics of high energy load and wind power are analyzed,and the principle of the participation of electrofusedmagnesiumhigh energy-carrying loads in the elimination of obstructedwind power is studied.Second,a two-layer optimization model is suggested,with the objective function being the largest amount of wind power consumed and the lowest possible cost of system operation.In the upper model,the high energy-carrying load regulates the blocked wind power,and in the lower model,the second-order cone approximation algorithm is used to solve the optimizationmodelwithwind power uncertainty,so that a two-layer optimizationmodel that takes into account the regulation of the high energy-carrying load of the electrofused magnesium and the uncertainty of the wind power is established.Finally,the model is solved using Gurobi,and the results of the simulation demonstrate that the suggested model may successfully lower wind abandonment,lower system operation costs,increase the accuracy of day-ahead scheduling,and lower the final product error of the thermal electricity unit.

Improved AVOA based on LSSVM for wind power prediction

Improving the prediction accuracy of wind power is an effective means to reduce the impact of wind power on power grid.Therefore,we proposed an improved African vulture optimization algorithm(AVOA)to realize the prediction model of multi-objective optimization least squares support vector machine(LSSVM).Firstly,the original wind power time series was decomposed into a certain number of intrinsic modal components(IMFs)using variational modal decomposition(VMD).Secondly,random numbers in population initialization were replaced by Tent chaotic mapping,multi-objective LSSVM optimization was introduced by AVOA improved by elitist non-dominated sorting and crowding operator,and then each component was predicted.Finally,Tent multi-objective AVOA-LSSVM(TMOALSSVM)method was used to sum each component to obtain the final prediction result.The simulation results show that the improved AVOA based on Tent chaotic mapping,the improved non-dominated sorting algorithm with elite strategy,and the improved crowding operator are the optimal models for single-objective and multi-objective prediction.Among them,TMOALSSVM model has the smallest average error of stroke power values in four seasons,which are 0.0694,0.0545 and 0.0211,respectively.The average value of DS statistics in the four seasons is 0.9902,and the statistical value is the largest.The proposed model effectively predicts four seasons of wind power values on lateral and longitudinal precision,and faster and more accurately finds the optimal solution on the current solution space sets,which proves that the method has a certain scientific significance in the development of wind power prediction technology.

Dispersed Wind Power Planning Method Considering Network Loss Correction with Cold Weather

In order to play a positive role of decentralised wind power on-grid for voltage stability improvement and loss reduction of distribution network,a multi-objective two-stage decentralised wind power planning method is proposed in the paper,which takes into account the network loss correction for the extreme cold region.Firstly,an electro-thermal model is introduced to reflect the effect of temperature on conductor resistance and to correct the results of active network loss calculation;secondly,a two-stage multi-objective two-stage decentralised wind power siting and capacity allocation and reactive voltage optimisation control model is constructed to take account of the network loss correction,and the multi-objective multi-planning model is established in the first stage to consider the whole-life cycle investment cost of WTGs,the system operating cost and the voltage quality of power supply,and the multi-objective planning model is established in the second stage.planning model,and the second stage further develops the reactive voltage control strategy of WTGs on this basis,and obtains the distribution network loss reduction method based on WTG siting and capacity allocation and reactive power control strategy.Finally,the optimal configuration scheme is solved by the manta ray foraging optimisation(MRFO)algorithm,and the loss of each branch line and bus loss of the distribution network before and after the adoption of this loss reduction method is calculated by taking the IEEE33 distribution system as an example,which verifies the practicability and validity of the proposed method,and provides a reference introduction for decision-making for the distributed energy planning of the distribution network.

The Short-Term Prediction ofWind Power Based on the Convolutional Graph Attention Deep Neural Network

The fluctuation of wind power affects the operating safety and power consumption of the electric power grid and restricts the grid connection of wind power on a large scale.Therefore,wind power forecasting plays a key role in improving the safety and economic benefits of the power grid.This paper proposes a wind power predicting method based on a convolutional graph attention deep neural network with multi-wind farm data.Based on the graph attention network and attention mechanism,the method extracts spatial-temporal characteristics from the data of multiple wind farms.Then,combined with a deep neural network,a convolutional graph attention deep neural network model is constructed.Finally,the model is trained with the quantile regression loss function to achieve the wind power deterministic and probabilistic prediction based on multi-wind farm spatial-temporal data.A wind power dataset in the U.S.is taken as an example to demonstrate the efficacy of the proposed model.Compared with the selected baseline methods,the proposed model achieves the best prediction performance.The point prediction errors(i.e.,root mean square error(RMSE)and normalized mean absolute percentage error(NMAPE))are 0.304 MW and 1.177%,respectively.And the comprehensive performance of probabilistic prediction(i.e.,con-tinuously ranked probability score(CRPS))is 0.580.Thus,the significance of multi-wind farm data and spatial-temporal feature extraction module is self-evident.

A method for cleaning wind power anomaly data by combining image processing with community detection algorithms

Current methodologies for cleaning wind power anomaly data exhibit limited capabilities in identifying abnormal data within extensive datasets and struggle to accommodate the considerable variability and intricacy of wind farm data.Consequently,a method for cleaning wind power anomaly data by combining image processing with community detection algorithms(CWPAD-IPCDA)is proposed.To precisely identify and initially clean anomalous data,wind power curve(WPC)images are converted into graph structures,which employ the Louvain community recognition algorithm and graph-theoretic methods for community detection and segmentation.Furthermore,the mathematical morphology operation(MMO)determines the main part of the initially cleaned wind power curve images and maps them back to the normal wind power points to complete the final cleaning.The CWPAD-IPCDA method was applied to clean datasets from 25 wind turbines(WTs)in two wind farms in northwest China to validate its feasibility.A comparison was conducted using density-based spatial clustering of applications with noise(DBSCAN)algorithm,an improved isolation forest algorithm,and an image-based(IB)algorithm.The experimental results demonstrate that the CWPAD-IPCDA method surpasses the other three algorithms,achieving an approximately 7.23%higher average data cleaning rate.The mean value of the sum of the squared errors(SSE)of the dataset after cleaning is approximately 6.887 lower than that of the other algorithms.Moreover,the mean of overall accuracy,as measured by the F1-score,exceeds that of the other methods by approximately 10.49%;this indicates that the CWPAD-IPCDA method is more conducive to improving the accuracy and reliability of wind power curve modeling and wind farm power forecasting.

Capacity Optimization Configuration of Hydrogen Production System for Offshore Surplus Wind Power

To solve the problem of residual wind power in offshore wind farms,a hydrogen production system with a reasonable capacity was configured to enhance the local load of wind farms and promote the local consumption of residual wind power.By studying the mathematical model of wind power output and calculating surplus wind power,as well as considering the hydrogen production/storage characteristics of the electrolyzer and hydrogen storage tank,an innovative capacity optimization allocation model was established.The objective of the model was to achieve the lowest total net present value over the entire life cycle.The model took into account the cost-benefit breakdown of equipment end-of-life cost,replacement cost,residual value gain,wind abandonment penalty,hydrogen transportation,and environmental value.The MATLAB-based platform invoked the CPLEX commercial solver to solve the model.Combined with the analysis of the annual average wind speed data from an offshore wind farm in Guangdong Province,the optimal capacity configuration results and the actual operation of the hydrogen production system were obtained.Under the calculation scenario,this hydrogen production system could consume 3,800 MWh of residual electricity from offshore wind power each year.It could achieve complete consumption of residual electricity from wind power without incurring the penalty cost of wind power.Additionally,it could produce 66,500 kg of green hydrogen from wind power,resulting in hydrogen sales revenue of 3.63 million RMB.It would also reduce pollutant emissions from coal-based hydrogen production by 1.5 tons and realize an environmental value of 4.83 million RMB.The annual net operating income exceeded 6 million RMB and the whole life cycle NPV income exceeded 50 million RMB.These results verified the feasibility and rationality of the established capacity optimization allocation model.The model could help advance power system planning and operation research and assist offshore wind farm operators in improving economic and environmental benefits.

A low-carbon economic dispatch model for electricity market with wind power based on improved ant-lion optimisation algorithm

Introducing carbon trading into electricity market can convert carbon dioxide into schedulable resources with economic value.However,the randomness of wind power generation puts forward higher requirements for electricity market transactions.Therefore,the carbon trading market is introduced into the wind power market,and a new form of low-carbon economic dispatch model is developed.First,the economic dispatch goal of wind power is be considered.It is projected to save money and reduce the cost of power generation for the system.The model includes risk operating costs to account for the impact of wind power output variability on the system,as well as wind farm negative efficiency operating costs to account for the loss caused by wind abandonment.The model also employs carbon trading market metrics to achieve the goal of lowering system carbon emissions,and analyze the impact of different carbon trading prices on the system.A low-carbon economic dispatch model for the wind power market is implemented based on the following two goals.Finally,the solution is optimised using the Ant-lion optimisation method,which combines Levi's flight mechanism and golden sine.The proposed model and algorithm's rationality is proven through the use of cases.

Missing interpolation model for wind power data based on the improved CEEMDAN method and generative adversarial interpolation network

Randomness and fluctuations in wind power output may cause changes in important parameters(e.g.,grid frequency and voltage),which in turn affect the stable operation of a power system.However,owing to external factors(such as weather),there are often various anomalies in wind power data,such as missing numerical values and unreasonable data.This significantly affects the accuracy of wind power generation predictions and operational decisions.Therefore,developing and applying reliable wind power interpolation methods is important for promoting the sustainable development of the wind power industry.In this study,the causes of abnormal data in wind power generation were first analyzed from a practical perspective.Second,an improved complete ensemble empirical mode decomposition with adaptive noise(ICEEMDAN)method with a generative adversarial interpolation network(GAIN)network was proposed to preprocess wind power generation and interpolate missing wind power generation sub-components.Finally,a complete wind power generation time series was reconstructed.Compared to traditional methods,the proposed ICEEMDAN-GAIN combination interpolation model has a higher interpolation accuracy and can effectively reduce the error impact caused by wind power generation sequence fluctuations.

Bio-Inspired Optimal Dispatching of Wind Power Consumption Considering Multi-Time Scale Demand Response and High-Energy Load Participation

Bio-inspired computer modelling brings solutions fromthe living phenomena or biological systems to engineering domains.To overcome the obstruction problem of large-scale wind power consumption in Northwest China,this paper constructs a bio-inspired computer model.It is an optimal wind power consumption dispatching model of multi-time scale demand response that takes into account the involved high-energy load.First,the principle of wind power obstruction with the involvement of a high-energy load is examined in this work.In this step,highenergy load model with different regulation characteristics is established.Then,considering the multi-time scale characteristics of high-energy load and other demand-side resources response speed,a multi-time scale model of coordination optimization is built.An improved bio-inspired model incorporating particle swarm optimization is applied to minimize system operation and wind curtailment costs,as well as to find the most optimal energy configurationwithin the system.Lastly,we take an example of regional power grid in Gansu Province for simulation analysis.Results demonstrate that the suggested scheduling strategy can significantly enhance the wind power consumption level and minimize the system’s operational cost.

Physics-Informed AI Surrogates for Day-Ahead Wind Power Probabilistic Forecasting with Incomplete Data for Smart Grid in Smart Cities

Due to the high inherent uncertainty of renewable energy,probabilistic day-ahead wind power forecasting is crucial for modeling and controlling the uncertainty of renewable energy smart grids in smart cities.However,the accuracy and reliability of high-resolution day-ahead wind power forecasting are constrained by unreliable local weather prediction and incomplete power generation data.This article proposes a physics-informed artificial intelligence(AI)surrogates method to augment the incomplete dataset and quantify its uncertainty to improve wind power forecasting performance.The incomplete dataset,built with numerical weather prediction data,historical wind power generation,and weather factors data,is augmented based on generative adversarial networks.After augmentation,the enriched data is then fed into a multiple AI surrogates model constructed by two extreme learning machine networks to train the forecasting model for wind power.Therefore,the forecasting models’accuracy and generalization ability are improved by mining the implicit physics information from the incomplete dataset.An incomplete dataset gathered from a wind farm in North China,containing only 15 days of weather and wind power generation data withmissing points caused by occasional shutdowns,is utilized to verify the proposed method’s performance.Compared with other probabilistic forecastingmethods,the proposed method shows better accuracy and probabilistic performance on the same incomplete dataset,which highlights its potential for more flexible and sensitive maintenance of smart grids in smart cities.

Dynamic grouping control of electric vehicles based on improved k-means algorithm for wind power fluctuations suppression

To address the significant lifecycle degradation and inadequate state of charge(SOC)balance of electric vehicles(EVs)when mitigating wind power fluctuations,a dynamic grouping control strategy is proposed for EVs based on an improved k-means algorithm.First,a swing door trending(SDT)algorithm based on compression result feedback was designed to extract the feature data points of wind power.The gating coefficient of the SDT was adjusted based on the compression ratio and deviation,enabling the acquisition of grid-connected wind power signals through linear interpolation.Second,a novel algorithm called IDOA-KM is proposed,which utilizes the Improved Dingo Optimization Algorithm(IDOA)to optimize the clustering centers of the k-means algorithm,aiming to address its dependence and sensitivity on the initial centers.The EVs were categorized into priority charging,standby,and priority discharging groups using the IDOA-KM.Finally,an two-layer power distribution scheme for EVs was devised.The upper layer determines the charging/discharging sequences of the three EV groups and their corresponding power signals.The lower layer allocates power signals to each EV based on the maximum charging/discharging power or SOC equalization principles.The simulation results demonstrate the effectiveness of the proposed control strategy in accurately tracking grid power signals,smoothing wind power fluctuations,mitigating EV degradation,and enhancing the SOC balance.