Two-stage photovoltaic power forecasting method with an optimized transformer

Accurate photovoltaic(PV)power forecasting ensures the stability and reliability of power systems.To address the complex characteristics of nonlinearity,volatility,and periodicity,a novel two-stage PV forecasting method based on an optimized transformer architecture is proposed.In the first stage,an inverted transformer backbone was utilized to consider the multivariate correlation of the PV power series and capture its non-linearity and volatility.ProbSparse attention was introduced to reduce high-memory occupation and solve computational overload issues.In the second stage,a weighted series decomposition module was proposed to extract the periodicity of the PV power series,and the final forecasting results were obtained through additive reconstruction.Experiments on two public datasets showed that the proposed forecasting method has high accuracy,robustness,and computational efficiency.Its RMSE improved by 31.23%compared with that of a traditional transformer,and its MSE improved by 12.57%compared with that of a baseline model.

Enhancing photovoltaic power prediction using a CNN-LSTM-attention hybrid model with Bayesian hyperparameter optimization

Improving the accuracy of solar power forecasting is crucial to ensure grid stability,optimize solar power plant operations,and enhance grid dispatch efficiency.Although hybrid neural network models can effectively address the complexities of environmental data and power prediction uncertainties,challenges such as labor-intensive parameter adjustments and complex optimization processes persist.Thus,this study proposed a novel approach for solar power prediction using a hybrid model(CNN-LSTM-attention)that combines a convolutional neural network(CNN),long short-term memory(LSTM),and attention mechanisms.The model incorporates Bayesian optimization to refine the parameters and enhance the prediction accuracy.To prepare high-quality training data,the solar power data were first preprocessed,including feature selection,data cleaning,imputation,and smoothing.The processed data were then used to train a hybrid model based on the CNN-LSTM-attention architecture,followed by hyperparameter optimization employing Bayesian methods.The experimental results indicated that within acceptable model training times,the CNN-LSTM-attention model outperformed the LSTM,GRU,CNN-LSTM,CNN-LSTM with autoencoders,and parallel CNN-LSTM attention models.Furthermore,following Bayesian optimization,the optimized model demonstrated significantly reduced prediction errors during periods of data volatility compared to the original model,as evidenced by MRE evaluations.This highlights the clear advantage of the optimized model in forecasting fluctuating data.

Short-Term Prediction of Photovoltaic Power Based on DBSCAN-SVM Data Cleaning and PSO-LSTM Model

Accurate short-termphotovoltaic(PV)power prediction helps to improve the economic efficiency of power stations and is of great significance to the arrangement of grid scheduling plans.In order to improve the accuracy of PV power prediction further,this paper proposes a data cleaning method combining density clustering and support vector machine.It constructs a short-termPVpower predictionmodel based on particle swarmoptimization(PSO)optimized Long Short-Term Memory(LSTM)network.Firstly,the input features are determined using Pearson’s correlation coefficient.The feature information is clustered using density-based spatial clustering of applications withnoise(DBSCAN),and then,the data in each cluster is cleanedusing support vectormachines(SVM).Secondly,the PSO is used to optimize the hyperparameters of the LSTM network to obtain the optimal network structure.Finally,different power prediction models are established,and the PV power generation prediction results are obtained.The results show that the data methods used are effective and that the PSO-LSTM power prediction model based on DBSCAN-SVM data cleaning outperforms existing typical methods,especially under non-sunny days,and that the model effectively improves the accuracy of short-term PV power prediction.

Photovoltaic Power Generation Power Prediction under Major Extreme Weather Based on VMD-KELM

The output of photovoltaic power stations is significantly affected by environmental factors,leading to intermittent and fluctuating power generation.With the increasing frequency of extreme weather events due to global warming,photovoltaic power stations may experience drastic reductions in power generation or even complete shutdowns during such conditions.The integration of these stations on a large scale into the power grid could potentially pose challenges to systemstability.To address this issue,in this study,we propose a network architecture based on VMDKELMfor predicting the power output of photovoltaic power plants during severe weather events.Initially,a grey relational analysis is conducted to identify key environmental factors influencing photovoltaic power generation.Subsequently,GMM clustering is utilized to classify meteorological data points based on their probabilities within different Gaussian distributions,enabling comprehensive meteorological clustering and extraction of significant extreme weather data.The data are decomposed using VMD to Fourier transform,followed by smoothing processing and signal reconstruction using KELM to forecast photovoltaic power output under major extreme weather conditions.The proposed prediction scheme is validated by establishing three prediction models,and the predicted photovoltaic output under four major extreme weather conditions is analyzed to assess the impact of severe weather on photovoltaic power station output.The experimental results show that the photovoltaic power output under conditions of dust storms,thunderstorms,solid hail precipitation,and snowstorms is reduced by 68.84%,42.70%,61.86%,and 49.92%,respectively,compared to that under clear day conditions.The photovoltaic power prediction accuracies,in descending order,are dust storms,solid hail precipitation,thunderstorms,and snowstorms.

Reliability-BasedModel for Incomplete Preventive ReplacementMaintenance of Photovoltaic Power Systems

At present,the operation and maintenance of photovoltaic power generation systems mainly comprise regular maintenance,breakdown maintenance,and condition-based maintenance,which is very likely to lead to over-or under-repair of equipment.Therefore,a preventive maintenance and replacement strategy for PV power generation systems based on reliability as a constraint is proposed.First,a hybrid failure function with a decreasing service age factor and an increasing failure rate factor is introduced to describe the deterioration of PV power generation equipment,and the equipment is replaced when its reliability drops to the replacement threshold in the last cycle.Then,based on the reliability as a constraint,the average maintenance cost and availability of the equipment are considered,and the non-periodic incomplete maintenance model of the PV power generation system is established to obtain the optimal number of repairs,each maintenance cycle and the replacement cycle of the PV power generation system components.Next,the inverter of a PV power plant is used as a research object.The model in this paper is compared and analyzed with the equal cycle maintenance model without considering reliability and the maintenance model without considering the equipment replacement threshold,Through model comparison,when the optimal maintenance strategy is(0.80,4),the average maintenance cost of this paper’s model are decreased by 20.3%and 5.54%and the availability is increased by 0.2395% and 0.0337%,respectively,compared with the equal-cycle maintenance model without considering the reliability constraint and the maintenance model without considering the equipment replacement threshold.Therefore,this maintenance model can ensure the high reliability of PV plant operation while increasing the equipment availability to improve the system economy.

Research on the MPPT of Photovoltaic Power Generation Based on Improved WOA and P&O under Partial Shading Conditions

Partial shading conditions(PSCs)caused by uneven illumination become one of the most common problems in photovoltaic(PV)systems,which can make the PV power-voltage(P-V)characteristics curve show multi-peaks.Traditional maximum power point tracking(MPPT)methods have shortcomings in tracking to the global maximum power point(GMPP),resulting in a dramatic decrease in output power.In order to solve the above problems,intelligent algorithms are used in MPPT.However,the existing intelligent algorithms have some disadvantages,such as slow convergence speed and large search oscillation.Therefore,an improved whale algorithm(IWOA)combined with the P&O(IWOA-P&O)is proposed for the MPPT of PV power generation in this paper.Firstly,IWOA is used to track the range interval of the GMPP,and then P&O is used to accurately find the MPP in that interval.Compared with other algorithms,simulation results show that this method has an average tracking efficiency of 99.79%and an average tracking time of 0.16 s when tracking GMPP.Finally,experimental verification is conducted,and the results show that the proposed algorithm has better MPPT performance compared to popular particle swarm optimization(PSO)and PSO-P&O algorithms.

Research Progress of Photovoltaic Power Prediction Technology Based on Artificial Intelligence Methods

With the increasing proportion of renewable energy in China’s energy structure,among which photovoltaic power generation is also developing rapidly.As the photovoltaic(PV)power output is highly unstable and subject to a variety of factors,it brings great challenges to the stable operation and dispatch of the power grid.Therefore,accurate short-term PV power prediction is of great significance to ensure the safe grid connection of PV energy.Currently,the short-term prediction of PV power has received extensive attention and research,but the accuracy and precision of the prediction have to be further improved.Therefore,this paper reviews the PV power prediction methods from five aspects:influencing factors,evaluation indexes,prediction status,difficulties and future trends.Then summarizes the current difficulties in prediction based on an in-depth analysis of the current research status of physical methods based on the classification ofmodel features,statistical methods,artificial intelligence methods,and combinedmethods of prediction.Finally,the development trend ofPVpower generation prediction technology and possible future research directions are envisioned.

Discussion on the Soil and Water Conservation Model in Mountain Photovoltaic Power Generation Project

In the context of rising global energy demand and increasing awareness of environmental protection,photovoltaic power generation,as a clean and renewable form of energy,has become increasingly important and has received widespread attention and application worldwide.However,during the construction and operation of mountain photovoltaic power generation projects,water and soil erosion has become a major challenge,which not only restricts the sustainable development process of the project,but also has a significant negative impact on the local ecological environment.This article deeply analyzes the multiple causes,extensive impacts and effective prevention and control strategies of water and soil erosion in mountain photovoltaic power generation projects.The results show that rainfall intensity,terrain slope,soil type and vegetation coverage are the four key factors leading to soil erosion.Soil erosion not only causes a sharp decline in soil fertility,but also aggravates the problem of sediment deposition in rivers and reservoirs,and poses a direct threat to the stability and operating efficiency of photovoltaic equipment.In order to deal with the above problems,this paper innovatively puts forward a series of soil and water conservation technologies,covering multiple dimensions such as engineering measures,plant measures,farming measures and temporary measures,and deeply discusses the application models and management strategies of these measures in key stages such as planning and design,construction,operation and maintenance.Through specific case analysis,the successful practical experience of soil and water conservation is refined and summarized,and the key role of community cooperation,technical support and modern monitoring technology in preventing and controlling soil and water erosion is further emphasized.This article aims to achieve a win-win situation of ecological environment protection and energy development and utilization through scientific planning and effective governance,and contribute to the construction of a green,low-carbon,and sustainable energy system.

Research on the Application of New Energy Photovoltaic Power Generation Technology

With the rapid development of technology and economy,the demand for energy in society is increasing.People are gradually realizing that fossil energy is limited,and the development of new energy may also face situations where it cannot meet social needs.The problem of resource shortage is gradually exposed to people.Therefore,the development of usable new energy has become an urgent problem for society to solve.At present,electricity is the most widely used energy source worldwide and photovoltaic power generation technology is gradually becoming well-known.As an emerging industry,the development of photovoltaic power generation still requires continuous promotion by national and social policies to be extended to various industries and ensure the stability of its energy supply.This article mainly outlines the principles,characteristics,and advantages of photovoltaic power generation,and briefly explains the current technology types and application aspects of photovoltaic power generation to contribute to its promotion and better serve all aspects of social life with new energy.

Comprehensive Benefit Evaluation of SZ Distributed Photovoltaic Power Generation Project Based on AHP-Matter-Element Extension Model

With the introduction of the“dual carbon goals,”there has been a robust development of distributed photovoltaic power generation projects in the promotion of their construction.As part of this initiative,a comprehensive and systematic analysis has been conducted to study the overall benefits of photovoltaic power generation projects.The evaluation process encompasses economic,technical,environmental,and social aspects,providing corresponding analysis methods and data references.Furthermore,targeted countermeasures and suggestions are proposed,signifying the research’s importance for the construction and development of subsequent distributed photovoltaic power generation projects.

Short-term prediction of photovoltaic power generation based on LMD-EE-ESN with error correction

Considering the instability of the output power of photovoltaic(PV)generation system,to improve the power regulation ability of PV power during grid-connected operation,based on the quantitative analysis of meteorological conditions,a short-term prediction method of PV power based on LMD-EE-ESN with iterative error correction was proposed.Firstly,through the fuzzy clustering processing of meteorological conditions,taking the power curves of PV power generation in sunny,rainy or snowy,cloudy,and changeable weather as the reference,the local mean decomposition(LMD)was carried out respectively,and their energy entropy(EE)was taken as the meteorological characteristics.Then,the historical generation power series was decomposed by LMD algorithm,and the hierarchical prediction of the power curve was realized by echo state network(ESN)prediction algorithm combined with meteorological characteristics.Finally,the iterative error theory was applied to the correction of power prediction results.The analysis of the historical data in the PV power generation system shows that this method avoids the influence of meteorological conditions in the short-term prediction of PV output power,and improves the accuracy of power prediction on the condition of hierarchical prediction and iterative error correction.

Solar Energy Resource Characteristics of Photovoltaic Power Station in Shandong Province

[Objective] The aim was to analyze characters of solar energy in photo- voltaic power stations in Shandong Province. [Method] The models of total solar radiation and scattered radiation were determined, and solar energy resources in pho-tovoltaic power stations were evaluated based on illumination in horizontal plane and cloud data in 123 counties or cities and observed information in Jinan, Fushan and Juxian in 1988-2008. [Result] Solar energy in northern regions in Shandong proved most abundant, which is suitable for photovoltaic power generation; the optimal angle of tilt of photovoltaic array was at 35°, decreasing by 2°-3° compared with local latitude. Total solar radiation received by the slope with optimal angle of tilt exceeded 1 600 kw.h/（m2.a）, increasing by 16% compared with horizontal planes. The maximal irradiance concluded by WRF in different regions tended to be volatile in 1 020-1 060 W/m2. [Conclusion] The research provides references for construction of photovoltaic power stations in Shandong Province.

Short-term photovoltaic power prediction using combined K-SVD-OMP and KELM method

For photovoltaic power prediction,a kind of sparse representation modeling method using feature extraction techniques is proposed.Firstly,all these factors affecting the photovoltaic power output are regarded as the input data of the model.Next,the dictionary learning techniques using the K-mean singular value decomposition(K-SVD)algorithm and the orthogonal matching pursuit(OMP)algorithm are used to obtain the corresponding sparse encoding based on all the input data,i.e.the initial dictionary.Then,to build the global prediction model,the sparse coding vectors are used as the input of the model of the kernel extreme learning machine(KELM).Finally,to verify the effectiveness of the combined K-SVD-OMP and KELM method,the proposed method is applied to a instance of the photovoltaic power prediction.Compared with KELM,SVM and ELM under the same conditions,experimental results show that different combined sparse representation methods achieve better prediction results,among which the combined K-SVD-OMP and KELM method shows better prediction results and modeling accuracy.

Forecasting Model of Photovoltaic Power Based on KPCA-MCS-DCNN

Accurate photovoltaic(PV)power prediction can effectively help the power sector to make rational energy planning and dispatching decisions,promote PV consumption,make full use of renewable energy and alleviate energy problems.To address this research objective,this paper proposes a prediction model based on kernel principal component analysis(KPCA),modified cuckoo search algorithm(MCS)and deep convolutional neural networks(DCNN).Firstly,KPCA is utilized to reduce the dimension of the feature,which aims to reduce the redundant input vectors.Then using MCS to optimize the parameters of DCNN.Finally,the photovoltaic power forecasting method of KPCA-MCS-DCNN is established.In order to verify the prediction performance of the proposed model,this paper selects a photovoltaic power station in China for example analysis.The results show that the new hybrid KPCA-MCS-DCNN model has higher prediction accuracy and better robustness.

Interpretation and revision proposals of GB/T 29319-2012,Technical requirements for connecting photovoltaic power system to distribution network

In the context of clean and Low-carbon energy transformation and new power system,China^photovoltaic power generation will usher in great development.Its large-scale access impacts the safe and stable operation of the power grid with increasing significance.In order to strengthen the support and Leading roles of the standards,it is urgent to revise the national standard GB/T 29319-2012,Technical requirements for connecting photovoltaic power system to distribution network,based on the current development trend of photovoltaic power generation and power grid transformation needs.This paper firstly interprets the important technical provisions of the standard,then analyzes the problems in its implementation and finally proposes some revision suggestions in terms of grid adaptability,power control and fault crossing,to facilitate safe and orderly development of photovoltaic power generation in China.

Issues and Solutions for China's Photovoltaic Power Industry

Solar energy is an important renewable energy.Developing photovoltaic power will not only relieve the energy supply-demand contradiction and optimize the energy structure,but also help to restructure this industry.This paper analyzes the status quo and the development prospects of China's photovoltaic power industry and its existing issues,and puts forward some suggestions and solutions for its healthy and orderly development.

Preliminary Feasibility Study on Application of Very Large Scale-Photovoltaic Power Generation in China

Some energy experts believe that solar energy photovoltaic power generation is hopeful to be applied in a large amount and possesses a certain proportion in the structure of energy in the future. In this paper, based on the forecasting of electric load demand and energy structure of power generation in the middle of 21 century, the pictures of VLS-PV power genera- tion is composed, the operation characteristic of VLS-PV power generation and the adaptability of electric power grid for it is analyzed, the ways for transmitting large amount of PV power and the economic and technical bottlenecks for applying VLS-PV power generation are discussed. Finally, the steps and suggestions for developing VLS-PV power generation and its electric power system in China are proposed.

Solar Shutters based on Photovoltaic Power Generation

This paper introduces a set of electrical energy-saving system for commercial office buildings,aiming at making better use of solar energy and photovoltaic power generation.Solar energy is a renewable energy source,which is inexhaustible clean energy and has great commercial application value.Based on this fact,we plan to design a unique and novel solar shutter in combination with the daily observation and the shape of solar panels.The shutter blades are equipped with an automatic light tracking system,and the angle of the blades can be adjusted in time through photoresistor induction,that is,as much solar energy as possible can be converted into electric energy for load use,and at the same time,comfortable light can be provided for the house.In essence,the system is a small photovoltaic power generation system,which runs all day with high-efficiency based on automatic sun tracking.Among them,the basic operation route includes:solar position detection,computer data processing,photovoltaic and electric volt energy conversion,circuit connection,etc.From the current debugging results,the shutter has the characteristics of humanization,high efficiency,cleanliness and so on.Through this energy-saving system,we hope to maximize the use of solar energy in the premise of low cost,so as to achieve the purpose of energy saving.

Semi-asynchronous personalized federated learning for short-term photovoltaic power forecasting

Accurate forecasting for photovoltaic power generation is one of the key enablers for the integration of solar photovoltaic systems into power grids.Existing deep-learning-based methods can perform well if there are sufficient training data and enough computational resources.However,there are challenges in building models through centralized shared data due to data privacy concerns and industry competition.Federated learning is a new distributed machine learning approach which enables training models across edge devices while data reside locally.In this paper,we propose an efficient semi-asynchronous federated learning framework for short-term solar power forecasting and evaluate the framework performance using a CNN-LSTM model.We design a personalization technique and a semi-asynchronous aggregation strategy to improve the efficiency of the proposed federated forecasting approach.Thorough evaluations using a real-world dataset demonstrate that the federated models can achieve significantly higher forecasting performance than fully local models while protecting data privacy,and the proposed semi-asynchronous aggregation and the personalization technique can make the forecasting framework more robust in real-world scenarios.

Effect of Photovoltaic Power Generation on Carbon Dioxide Emission Reduction under Double Carbon Background

Increasing the efficiency and proportion of photovoltaic power generation installations is one of the best ways to reduce both CO_(2) emissions and reliance on fossil-fuel-based power supplies.Solar energy is a clean and renewable power source with excellent potential for further development and utilization.In 2021,the global solar installed capacity was about 749.7 GW.Establishing correlations between solar power generation,standard coal equivalent,carbon sinks,and green sinks is crucial.However,there have been few reports about correlations between the efficiency of tracking solar photovoltaic panels and the above parameters.This paper calculates the increased power generation achievable through the use of tracking photovoltaic panels compared with traditional fixed panels and establishes relationships between power generation,standard coal equivalent,and carbon sinks,providing a basis for attempts to reduce reliance on carbon-based fuels.The calculations show that power generation efficiency can be improved by about 26.12%by enabling solar panels to track the sun's rays during the day and from season to season.Through the use of this improved technology,global CO_(2) emissions can be reduced by 183.63 Mt,and the standard coal equivalent can be reduced by 73.67 Mt yearly.Carbon capture is worth approximately EUR 15.48 billion,and carbon accounting analysis plays a vital role in carbon trading.