Short-term photovoltaic power forecasting based on a new hybrid deep learning model incorporating transfer learning strategy

The accurate prediction of photovoltaic(PV)power generation is an important basis for hybrid grid scheduling.With the expansion of the scale of PV power plants and the popularization of distributed PV,this study proposes a multilayer PV power generation prediction model based on transfer learning to solve the problems of the lack of data on new PV bases and the low accuracy of PV power generation prediction.The proposed model,called DRAM,concatenates a dilated convolutional neural network(DCNN)module with a bidirectional long short-term memory(BiLSTM)module,and integrates an attention mechanism.First,the processed data are input into the DCNN layer,and the dilation convolution mechanism captures the spatial features of the wide sensory field of the input data.Subsequently,the temporal characteristics between the features are extracted in the BiLSTM layer.Finally,an attention mechanism is used to strengthen the key features by assigning weights to efficiently construct the relationship between the features and output variables.In addition,the power prediction accuracy of the new PV sites was improved by transferring the pre-trained model parameters to the new PV site prediction model.In this study,the pre-training of models using data from different source domains and the correlations between these pre-trained models and the target domain were analyzed.

Photovoltaic Models Parameters Estimation Based on Weighted Mean of Vectors

Renewable energy sources are gaining popularity,particularly photovoltaic energy as a clean energy source.This is evident in the advancement of scientific research aimed at improving solar cell performance.Due to the non-linear nature of the photovoltaic cell,modeling solar cells and extracting their parameters is one of the most important challenges in this discipline.As a result,the use of optimization algorithms to solve this problem is expanding and evolving at a rapid rate.In this paper,a weIghted meaN oF vectOrs algorithm(INFO)that calculates the weighted mean for a set of vectors in the search space has been applied to estimate the parameters of solar cells in an efficient and precise way.In each generation,the INFO utilizes three operations to update the vectors’locations:updating rules,vector merging,and local search.The INFO is applied to estimate the parameters of static models such as single and double diodes,as well as dynamic models such as integral and fractional models.The outcomes of all applications are examined and compared to several recent algorithms.As well as the results are evaluated through statistical analysis.The results analyzed supported the proposed algorithm’s efficiency,accuracy,and durability when compared to recent optimization algorithms.

Evaluation of model fidelity for solar analysis in the context of distributed PV integration at urban scale

This paper examines the level of model fidelity required to support design phases in the urban solar planning process.The two modelling features crucial to the fidelity of the photovoltaic(PV)yield prediction on urban surfaces are(1)a level of fidelity for modelling urban shading and solar reflection and(2)a level of fidelity for modelling PV system operation.The paper compares three different models for predicting urban shading and reflection and two different PV models for predicting PV system operation.The relevance of the model fidelities is investigated through a case study of an urban area in Wuhan,China under three decision-making contexts:setting a solar target,place-making,and economic assessment for urban-scale distributed PV integration.Predictions for the decision-makings are generated using the selected models through computational simulation under the same annual weather profile.The results show that the relatively less accurate canyon-based method tends to overpredict with 57 buildings identified as suitable for PV installation for walls in the studied urban area;the more accurate vector-based model predicts only 14 suitable buildings.The results predicted with additional consideration of dynamic PV system operation exhibit differences from those predicted by the static PV system model,with a difference of roughly 13 buildings on average within each payback-time category.The differences are noticeable but can be regarded as incremental for urban-scale economic assessment compared with the significant difference due to the fidelity level of modelling urban shading and reflection.

Dynamic characteristics and improved MPPT control of PV generator

This paper presents a mathematical model of photovoltaic （PV） module and gives a strategy to calculate online the maximum power point （MPP）. The variation of series and shunt resistor are taken into account in the model and are dynamically identified using the Newton-Raphson algorithm. The effectiveness of the proposed model is verified by laboratory experiments obtained by implementing the model on the dSPACE DS1104 board.

Locomotion-based Hybrid Salp Swarm Algorithm for Parameter Estimation of Fuzzy Representation-based Photovoltaic Modules

Identifying the parameters of photovoltaic(PV)modules is significant for their design and simulation.Because of the instabilities in the weather action and land surface of the earth,which cause errors in measuring,a novel fuzzy representation-based PV module is formulated and developed.In this paper,a novel locomotion-based hybrid salp swarm algorithm(LHSSA)is presented to identify the parameters of PV modules accurately and reliably.In the LHSSA,better leader salps based on particle swarm optimization(PSO)are incorporated to the traditional salp swarm algorithm(SSA)in a serialized scheme with the aim of providing more valuable information for the leader salps of the SSA.By this integration,the proposed LHSSA can escape the local optima as well as guide the seeking process to attain the promising region.The proposed LHSSA is investigated on different PV models,i.e.,single-diode(SD),double-diode(DD),and PV module in crisp and fuzzy aspects.By comparing with different algorithms,the comprehensive results affirm that the LHSSA can achieve a highly competitive performance,especially on quality and reliability.