A VAE-Bayesian deep learning scheme for solar power generation forecasting based on dimensionality reduction

The advancements in distributed generation(DG)technologies such as solar panels have led to a widespread integration of renewable power generation in modern power systems.However,the intermittent nature of renewable energy poses new challenges to the network operational planning with underlying uncertainties.This paper proposes a novel probabilistic scheme for renewable solar power generation forecasting by addressing data and model parameter uncertainties using Bayesian bidirectional long short-term memory(BiLSTM)neural networks,while handling the high dimensionality in weight parameters using variational auto-encoders(VAE).The forecasting performance of the proposed method is evaluated using various deterministic and probabilistic evaluation metrics such as root-mean square error(RMSE),Pinball loss,etc.Furthermore,reconstruction error and computational time are also monitored to evaluate the dimensionality reduction using the VAE component.When compared with benchmark methods,the proposed method leads to significant improvements in weight reduction,i.e.,from 76,4224 to 2,022 number of weight parameters,quantifying to 97.35%improvement in weight parameters reduction and 37.93%improvement in computational time for 6 months of solar power generation data.

CT-NET: A Novel Convolutional Transformer-Based Network for Short-Term Solar Energy Forecasting Using Climatic Information

Photovoltaic(PV)systems are environmentally friendly,generate green energy,and receive support from policies and organizations.However,weather fluctuations make large-scale PV power integration and management challenging despite the economic benefits.Existing PV forecasting techniques(sequential and convolutional neural networks(CNN))are sensitive to environmental conditions,reducing energy distribution system performance.To handle these issues,this article proposes an efficient,weather-resilient convolutional-transformer-based network(CT-NET)for accurate and efficient PV power forecasting.The network consists of three main modules.First,the acquired PV generation data are forwarded to the pre-processing module for data refinement.Next,to carry out data encoding,a CNNbased multi-head attention(MHA)module is developed in which a single MHA is used to decode the encoded data.The encoder module is mainly composed of 1D convolutional and MHA layers,which extract local as well as contextual features,while the decoder part includes MHA and feedforward layers to generate the final prediction.Finally,the performance of the proposed network is evaluated using standard error metrics,including the mean squared error(MSE),root mean squared error(RMSE),and mean absolute percentage error(MAPE).An ablation study and comparative analysis with several competitive state-of-the-art approaches revealed a lower error rate in terms of MSE(0.0471),RMSE(0.2167),and MAPE(0.6135)over publicly available benchmark data.In addition,it is demonstrated that our proposed model is less complex,with the lowest number of parameters(0.0135 M),size(0.106 MB),and inference time(2 ms/step),suggesting that it is easy to integrate into the smart grid.

Forecasting method of monthly wind power generation based on climate model and long short-term memory neural network

Predicting wind power gen eration over the medium and long term is helpful for dispatchi ng departme nts,as it aids in constructing generation plans and electricity market transactions.This study presents a monthly wind power gen eration forecast!ng method based on a climate model and long short-term memory(LSTM)n eural n etwork.A non linear mappi ng model is established between the meteorological elements and wind power monthly utilization hours.After considering the meteorological data(as predicted for the future)and new installed capacity planning,the monthly wind power gen eration forecast results are output.A case study shows the effectiveness of the prediction method.

Forecasting promising technology using analysis of patent information:Focused on next generation mobile communications

In order to forecast promising technologies in the field of next generation mobile communication, various patent indicators were analyzed such as citation per patent, patent family information, patent share, increase rate, and patent activity. These indicators were quantified into several indexes and then integrated into an evaluation score to provide promising technologies. As a result of the suggested patent analysis, four technologies out of twenty two in details classification were selected, which showed outstanding technology competitiveness, high patent share and increasing rates as well as high recent-patent-ratios and triad-patent-family-ratios. Each of the selected technologies scored more than 10 points in total, and the following four technologies were suggested as promising ones in the field of next generation mobile communication: 1) 3GPP based mobile communication, 2) beyond 4G mobile communication, 3) IEEE 802.16 based mobile communication, which are in medium classification of broadband mobile communication system, and 4) testing/certification system of mobile communication, which is in medium classification of mobile communication testing/certification system.

The Hidden-Layers Topology Analysis of Deep Learning Models in Survey for Forecasting and Generation of the Wind Power and Photovoltaic Energy

As wind and photovoltaic energy become more prevalent,the optimization of power systems is becoming increasingly crucial.The current state of research in renewable generation and power forecasting technology,such as wind and photovoltaic power(PV),is described in this paper,with a focus on the ensemble sequential LSTMs approach with optimized hidden-layers topology for short-term multivariable wind power forecasting.The methods for forecasting wind power and PV production.The physical model,statistical learningmethod,andmachine learning approaches based on historical data are all evaluated for the forecasting of wind power and PV production.Moreover,the experiments demonstrated that cloud map identification has a significant impact on PV generation.With a focus on the impact of photovoltaic and wind power generation systems on power grid operation and its causes,this paper summarizes the classification of wind power and PV generation systems,as well as the benefits and drawbacks of PV systems and wind power forecasting methods based on various typologies and analysis methods.

Comprehensive applicability evaluation of four precipitation products at multiple spatiotemporal scales in Northwest China

Precipitation plays a crucial role in the water cycle of Northwest China.Obtaining accurate precipitation data is crucial for regional water resource management,hydrological forecasting,flood control and drought relief.Currently,the applicability of multi-source precipitation products for long time series in Northwest China has not been thoroughly evaluated.In this study,precipitation data from 183 meteorological stations in Northwest China from 1979 to 2020 were selected to assess the regional applicability of four precipitation products(the fifth generation of European Centre for Medium-Range Weather Forecasts(ECMWF)atmospheric reanalysis of the global climate(ERA5),Global Precipitation Climatology Centre(GPCC),Climatic Research Unit gridded Time Series Version 4.07(CRU TS v4.07,hereafter CRU),and Tropical Rainfall Measuring Mission(TRMM))based on the following statistical indicators:correlation coefficient,root mean square error(RMSE),relative bias(RB),mean absolute error(MAE),probability of detection(POD),false alarm ratio(FAR),and equitable threat score(ETS).The results showed that precipitation in Northwest China was generally high in the east and low in the west,and exhibited an increasing trend from 1979 to 2020.Compared with the station observations,ERA5 showed a larger spatial distribution difference than the other products.The overall overestimation of multi-year average precipitation was approximately 200.00 mm and the degree of overestimation increased with increasing precipitation intensity.The multi-year average precipitation of GPCC and CRU was relatively close to that of station observations.The trend of annual precipitation of TRMM was overestimated in high-altitude regions and the eastern part of Lanzhou with more precipitation.At the monthly scale,GPCC performed well but underestimated precipitation in the Tarim Basin(RB=-4.11%),while ERA5 and TRMM exhibited poor accuracy in high-altitude regions.ERA5 had a large bias(RB≥120.00%)in winter months and a strong dispersion(RMSE≥35.00 mm)in summer months.TRMM showed a relatively low correlation with station observations in winter months(correlation coefficients≤0.70).The capture performance analysis showed that ERA5,GPCC,and TRMM had lower POD and ETS values and higher FAR values in Northwest China as the precipitation intensity increased.ERA5 showed a high capture performance for small precipitation events and a slower decreasing trend of POD as the precipitation intensity increased.GPCC had the lowest FAR values.TRMM was statistically ineffective for predicting the occurrence of daily precipitation events.The findings provide a reference for data users to select appropriate datasets in Northwest China and for data developers to develop new precipitation products in the future.

Multi-dimensional scenario forecast for generation of multiple wind farms

A novel multi-dimensional scenario forecast approach which can capture the dynamic temporal-spatial interdependence relation among the outputs of multiple wind farms is proposed.In the proposed approach,support vector machine(SVM)is applied for the spot forecast of wind power generation.The probability density function(PDF)of the SVM forecast error is predicted by sparse Bayesian learning(SBL),and the spot forecast result is corrected according to the error expectation obtained.The copula function is estimated using a Gaussian copula-based dynamic conditional correlation matrix regression(DCCMR)model to describe the correlation among the errors.And the multidimensional scenario is generated with respect to the estimated marginal distributions and the copula function.Test results on three adjacent wind farms illustrate the effectiveness of the proposed approach.

Bayesian bootstrap quantile regression for probabilistic photovoltaic power forecasting

Photovoltaic(PV)systems are widely spread across MV and LV distribution systems and the penetration of PV generation is solidly growing.Because of the uncertain nature of the solar energy resource,PV power forecasting models are crucial in any energy management system for smart distribution networks.Although point forecasts can suit many scopes,probabilistic forecasts add further flexibility to an energy management system and are recommended to enable a wider range of decision making and optimization strategies.This paper proposes methodology towards probabilistic PV power forecasting based on a Bayesian bootstrap quantile regression model,in which a Bayesian bootstrap is applied to estimate the parameters of a quantile regression model.A novel procedure is presented to optimize the extraction of the predictive quantiles from the bootstrapped estimation of the related coefficients,raising the predictive ability of the final forecasts.Numerical experiments based on actual data quantify an enhancement of the performance of up to 2.2%when compared to relevant benchmarks.

Energy Management System Design and Testing for Smart Buildings Under Uncertain Generation (Wind/Photovoltaic) and Demand

This study provides details of the energy management architecture used in the Goldwind microgrid test bed. A complete mathematical model, including all constraints and objectives, for microgrid operational management is first described using a modified prediction interval scheme. Forecasting results are then achieved every 10 min using the modified fuzzy prediction interval model, which is trained by particle swarm optimization.A scenario set is also generated using an unserved power profile and coverage grades of forecasting to compare the feasibility of the proposed method with that of the deterministic approach. The worst case operating points are achieved by the scenario with the maximum transaction cost. In summary, selection of the maximum transaction operating point from all the scenarios provides a cushion against uncertainties in renewable generation and load demand.