Photovoltaic Power Prediction and SHAP Interpretability Analysis Based on Multi-Model Comparative Learning

With the increasing global warming and enormous pollution, it is very obvious to generate power from renewable energy sources. In fact, photo voltaic (PV) power is a main renewable energy source for sustainable power generation. But PV power generation is uncertain and intermittent in nature. Stable and reliable power supply may not be feasible with PV power. Power supply reliability is important as much as its availability. To handle the inconsistency of PV power generation, power sector is highly dependent on forecasting methods and techniques. In this paper the feed forward neural network (FFNN) of artificial neural networks(ANN) with optimization technique particle swarm optimization (PSO) and support vector regression (SVR) are used in short term PV power generation forecasting and their performance is compared with the error calculation in terms of the mean absolute error (MAE) and root mean squared error (RMSE). The PV power and meteorological data of solar irradiation and temperature of Kolkata region of India was used for forecasting. Similar days of the year sunny days, cloudy days and rainy days are clustered using k-means clustering technique. Thereafter, a feed forward neural network is implemented to forecast a day head PV power and week ahead photo voltaic power. PSO is used to optimize the weights of the neural network. The performance of the proposed forecasting method is compared to a data mining approach, support vector regression (SVR).In this paper, the significance of proper selection of input parameters in PV power forecasting is emphasized. With the application of K-means clustering, the accuracy of ANN-PSO approach is improved significantly in a day ahead PV power forecasting and a week ahead PV power forecasting with a good margin. Even the accuracy of ANN-PSO approach after K-means clustering is better than SVR model in a week ahead forecasting

Short-term PV power prediction based on meteorological similarity days and SSA-BiLSTM

Traditional photovoltaic (PV) power prediction methods in microgrids are insensitive to weather characteristics, and fluctuations in weather data can lead to insufficient PV power prediction accuracy. Therefore, this paper proposes a short-term PV power prediction method for microgrids based on FCM–QPSO–LSTM, which is used to predict PV power under different seasons and weather types. First, the fuzzy [Formula: see text]-means clustering algorithm (FCM) is used to perform clustering analysis on multi-source weather data, and the initial dataset is clustered into sunny, cloudy, and rainy days. Second, to address the shortcomings of a single long-short-term memory (LSTM) model in terms of prediction accuracy and stability, the quantum particle swarm optimization (QPSO) algorithm is used to optimize the model of the LSTM network, and the QPSO–LSTM prediction model is constructed. By introducing quantum dynamic inertia weights and adaptive learning factors, the QPSO algorithm enhances the global search capability and local search capability, which greatly improves the prediction accuracy and stability of the prediction model. Finally, the average absolute percentage error and root mean square error (RMSE) are applied as evaluation indexes to assess the performance of the prediction model. The simulation results show that the QPSO–LSTM model is better than the single LSTM model and the traditional PSO–LSTM model in terms of the average absolute percentage error and RMSE indicators and is able to make accurate predictions of short-term PV power under different seasons and weather types.

In view of the large fluctuation of photovoltaic output power affected by different weather, accurate prediction of photovoltaic output power is particularly important for the safe and stable operation of power system. Firstly, the pelican optimization algorithm ( POA ) is improved in the following three aspects : adding Circle chaotic map to make the population evenly distributed, introducing mutation factor to expand the search range of pelican when approaching prey, adding adaptive weight and firefly disturbance to avoid falling into local optimum in the water surface flight stage ; then, in order to improve the prediction accuracy of BP algorithm, the improved pelican algorithm ( IPOA ) is used to optimize the weights and thresholds of BP neural network, and the IPOA-BP photovoltaic power prediction model is built to improve the accuracy of power prediction. Finally, this paper tests the prediction performance of IPOA-BP, POA-BP and basic BP power prediction models in sunny, cloudy and rainy days through experiments. The experimental results demonstrate that the IPOA-BP prediction model outperforms both the POA-BP and traditional BP neural network models under various weather conditions.

Photovoltaic (PV) power forecasting plays a crucial role in optimizing renewable energy integration into the grid, necessitating accurate predictions to mitigate the inherent variability of solar energy generation. We propose a novel forecasting model that combines improved variational mode decomposition (IVMD) with the temporal convolutional network-gated recurrent unit (TCN-GRU) architecture, enriched with a multi-head attention mechanism. By focusing on four key environmental factors influencing PV output, the proposed IVMD-TCN-GRU framework targets a significant research gap in renewable energy forecasting methodologies. Initially, leveraging the sparrow search algorithm (SSA), we optimize the parameters of VMD, including the mode component K-value and penalty factor, based on the minimum envelope entropy principle. The optimized VMD then decomposes PV power, while the TCN-GRU model harnesses TCN’s proficiency in learning local temporal features and GRU’s capability in rapidly modeling sequence data, while leveraging multi-head attention to better utilize the global correlation information within sequence data. Through this design, the model adeptly captures the correlations within time series data, demonstrating superior performance in prediction tasks. Subsequently, the SSA is employed to optimize GRU parameters, and the decomposed PV power mode components and environmental feature attributes are inputted into the TCN-GRU neural network. This facilitates dynamic temporal modeling of multivariate feature sequences. Finally, the predicted values of each component are summed to realize PV power forecasting. Validation using real data from a PV station corroborates that the novel model demonstrates a substantial reduction in RMSE and MAE of up to 55.1% and 54.5%, respectively, particularly evident in instances of pronounced photovoltaic power fluctuations during inclement weather conditions. The proposed method exhibits marked improvements in accuracy compared to traditional PV power prediction methods, underscoring its significance in enhancing forecasting precision and ensuring the secure scheduling and stable operation of power systems.

Deep attention ConvLSTM-based adaptive fusion of clear-sky physical prior knowledge and multivariable historical information for probabilistic prediction of photovoltaic power

Accurate prediction of photovoltaic power is of great significance to the safe operation of power grids. In order to improve the prediction accuracy, a similar day clustering convolutional neural network (CNN)–informer model was proposed to predict the photovoltaic power. Based on correlation analysis, it was determined that global horizontal radiation was the meteorological factor that had the greatest impact on photovoltaic power, and the dataset was divided into four categories according to the correlation between meteorological factors and photovoltaic power fluctuation characteristics; then, a CNN was used to extract the feature information and trends of different subsets, and the features output by CNN were fused and input into the informer model. The informer model was used to establish the temporal feature relationship between historical data, and the final photovoltaic power generation power prediction result was obtained. The experimental results show that the proposed CNN–informer prediction method has high accuracy and stability in photovoltaic power generation prediction and outperforms other deep learning methods.

Solar photovoltaic power generation as an inexhaustible, inexhaustible clean energy has become the focus of future energy development. Along with photovoltaic power generation incorporated into the power grid, in order to make power generation plan reasonably, ensure the stable operation of power system, need to forecast the photovoltaic power output. In this paper solar photovoltaic power generation forecasting methods are analyzed and summarized. According to the application of solar photovoltaic power generation and demand, mainly on photovoltaic power generation system power prediction research method has carried on the comprehensive elaboration, hoping for the researches play an important role in promoting and advancing the development of solar photovoltaic prediction methods.

Accurate photovoltaic (PV) power forecasting is essential for the stable and reliable operation of PV power generation systems. Recently, various deep learning- (DL-) based forecasting models have been proposed for accurate forecasting, but newly built systems cannot benefit from them due to the absence of PV power data. Although zero-shot methods based on single site can be used for PV power forecasting, they suffer from performance degradation problems when the characteristics of the source data and target data are different. To address this issue, we propose a novel zero-shot PV power forecasting scheme that leverages historical data from multiple PV generation systems at different sites. The proposed scheme constructs an individual forecasting model using historical data from each PV generation system. Then, two correlation coefficients are calculated for each forecasting model: one based on the correlation between the input variables of the source data and target data and the other on the correlation between the input variables and output variables of the source data. Lastly, the final forecasting value is calculated as a weighted sum of the predicted values of the constructed forecasting models for the input variables of the target data. In the extensive experiments for diverse DL models for forecasting, correlation coefficient types for weights, and data time intervals, the combination of recurrent neural network, Pearson’s correlation coefficient, and solar-noon time yielded the best prediction performance, with an improvement of up to 34.47% in mean absolute error and up to 15.94% in root mean square error compared to the best single-site zero-shot prediction. In addition, in experiments on PV power data from 9 cities in Korea using this combination, the proposed scheme achieved the best predictive performance in almost all cases and the second-best performance with a very narrow margin only in a few cases.

With the continuous increase of solar penetration rate, it has brought challenges to the smooth operation of the power grid. Therefore, to make photovoltaic power generation not affect the smooth operation of the grid, accurate photovoltaic power prediction is required. And short-term forecasting is essential for the deployment of daily power generation plans. In this paper, A short-term photovoltaic power generation forecast method based on K-means++, grey relational analysis (GRA) and support vector regression (SVR) (Hybrid Kmeans-GRA-SVR, HKGSVR) was proposed. The historical power data was clustered through the multi-index K-means++ algorithm. And the similar days and the nearest neighbor similar day of the prediction day were selected by the GRA algorithm. Then, similar days and nearest neighbor similar days were used to train SVR to obtain an accurate photovoltaic power prediction model. Under ideal weather, the average values of MAE, RMSE, and R2 were 0.8101 kW, 0.9608 kW, and 99.66%, respectively. The average computation time was 1.7487 s, which was significantly better than the SVR model. Thus, the demonstrated numerical results verify the effectiveness of the proposed model for short-term PV power prediction.

The prevalence of extreme weather events gives rise to a significant degree of prediction bias in the forecasting of photovoltaic (PV) power. In order to enhance the precision of forecasting outcomes, this study examines the interrelationships between China’s 24 conventional solar terms and extreme meteorological events. Additionally, it proposes a methodology for estimating the short-term generation of PV power based on the division of solar term time series. Firstly, given that the meteorological data from the same festival is more representative of the climate state at the current prediction moment, the sample data are grouped according to the 24 festival time nodes. Secondly, a convolutional neural network–long short-term memory (CNN-LSTM) PV power prediction model based on an Attention mechanism is proposed. This model extracts temporal change information from nonlinear sample data through LSTM, and a CNN link is added at the front end of LSTM to address the issue of LSTM being unable to obtain the spatial linkage of multiple features. Additionally, an Attention mechanism is incorporated at the back end of the CNN to obtain the feature information of crucial time steps, further reducing the multi-step prediction error. Concurrently, a PV power error prediction model is constructed to rectify the outcomes of the aforementioned prediction model. The examination of the measured data from PV power stations and the comparison and analysis with other prediction models demonstrate that the model presented in this paper can effectively enhance the accuracy of PV power predictions.

With the rapid growth of the global renewable energy industry, photovoltaic (PV) power generation has become a crucial component of clean energy systems. However, PV power output exhibits significant variability and uncertainty due to weather conditions, seasonal shifts, and geographical differences. Developing an accurate and robust PV power prediction model is essential for improving the operational efficiency of PV plants, reducing power system regulation costs, and increasing the utilization rate of renewable energy. This study proposes a novel PV power forecasting model based on the State-Frequency Memory (SFM) recurrent network. The model incorporates a time-varying transformation mechanism for specific sequences, combined with discrete Fourier transform-based frequency decomposition and weight adjustment techniques, to effectively handle the frequency information in PV power output, thereby enhancing prediction accuracy. To evaluate the model’s effectiveness, historical datasets from the National Renewable Energy Laboratory in the United States were used for both short-term and seasonal PV forecasting. Comparative experiments with conventional short-term prediction models demonstrate that the proposed SFM-based model achieves superior performance, with the lowest root mean square error and mean absolute error values across various time scales and seasonal conditions. These results confirm the model’s high accuracy, robustness, and practical value in supporting reliable PV power forecasting.

The accuracy of photovoltaic (PV) power prediction is significantly influenced by the high complexity and volatility of the PV sequence. The existing methods for predicting photoelectric power are difficult to effectively mine and analyze the internal variation law of data. To improve the accuracy of PV power prediction, a new method is proposed that first performs variational mode decomposition (VMD) and empirical mode decomposition (EMD), and then establishes a bi-directional long and short-term memory neural network (BiLSTM) for PV output power prediction. The proposed method extracts the amplitude and frequency characteristics of the PV output power series through VMD. After that, the residual term with strong non-stationarity is generated, which still has more sequence characteristics. The residual term is then decomposed by EMD for the second time to extract more features. Finally, the BiLSTM model is established to conduct bidirectional mining for PV power data and predict PV output power. The actual PV data is used to test the experimental results, which show that the proposed VMD-EMD-BiLSTM prediction model has better prediction performance.

Novel applications of various neural network models for prediction of photovoltaic system power under outdoor condition of mountainous region

A photovoltaic (PV)-powered electric motor is used for hose-drawn traveler driving instead of a water turbine to achieve high transmission efficiency. PV power generation (PVPG) is affected by different meteorological conditions, resulting in different power generation of PV panels for a hose-drawn traveler. In the above situation, the hose-drawn traveler may experience deficit power generation. The reasonable determination of the PV panel capacity is crucial. Predicting the PVPG is a prerequisite for the reasonable determination of the PV panel capacity. Therefore, it is essential to develop a method for accurately predicting PVPG. Extreme gradient boosting (XGBoost) is currently an outstanding machine learning model for prediction performance, but its hyperparameters are difficult to set. Thus, the XGBoost model based on particle swarm optimization (PSO-XGBoost) is applied for PV power prediction in this study. The PSO algorithm is introduced to optimize hyperparameters in XGBoost model. The meteorological data are segmented into four seasons to develop tailored prediction models, ensuring accurate prediction of PVPG in four seasons for hose-drawn travelers. The input variables of the models include solar irradiance, time, and ambient temperature. The prediction accuracy and stability of the model is then assessed statistically. The predictive accuracy and stability of PV power prediction by the PSO-XGBoost model are higher compared to the XGBoost model. Finally, application of the PSO-XGBoost model is implemented based on meteorological data.