Historical Power Data Research Articles

Ultra-short-term photovoltaic power prediction based on similar day clustering and temporal convolutional network with bidirectional long short-term memory model: A case study using DKASC data

Due to its strong dependence on weather conditions, photovoltaic (PV) power is highly intermittent in nature. In light of this, this paper introduces a PV power prediction model based on similar-day clustering and temporal convolutional network (TCN) with bidirectional long short-term memory (BiLSTM) model. Firstly, in the data preprocessing stage, improved complete ensemble empirical mode decomposition with adaptive noise (ICEEMDAN) method is used to decompose and denoise the original PV power data to obtain subsequences of different frequencies. Subsequently, the sample entropy (SE) method is incorporated to reconstruct the subsequences and generate low-frequency and high-frequency subsequences with more pronounced temporal features. In addition, a self-organizing map (SOM) with the K-means algorithm is used to categorize historical days into four weather types: sunny, cloudy, rainy, and intermittent. Through grey relational analysis (GRA), meteorological factors significantly affecting PV power prediction are identified to construct a multidimensional feature set. During the model prediction phase, historical PV power data and related meteorological impact factors are input into the TCN-BiLSTM model to perform multi-data-driven PV power prediction. Finally, the predicted results of each subsequence are linearly combined to obtain the ultimate PV power prediction. The proposed model is compared with convolutional neural network (CNN)-LSTM, LSTM, and TCN models, achieving a reduction in the root mean square error (RMSE) of 0.129 kW, 0.238 kW, and 0.257 kW, respectively. This demonstrates that the proposed model exhibits superior overall performance in time series modeling and information capture, enhancing the understanding of seasonal, periodic, and irregular patterns in PV power generation data.

Improved informer PV power short-term prediction model based on weather typing and AHA-VMD-MPE

Precise prediction of PV power in the short term is crucial for maintaining power system stability and balance. However, the performance of conventional time series prediction models on short-term long series prediction is scarcely sufficient because of the stochastic and turbulent character of PV power data. This work suggests a PV short-term power forecast model based on weather type, AHA-VMD-MPE decomposition reconstruction, and improved Informer combination to tackle this issue. Firstly, a SUM-ApEn-K-mean++ multidimensional clustering method to group the dataset by weather conditions. Then an AHA-VMD-MPE decomposition model is proposed to decompose the historical power data Finally the Informer model is improved and the improved model is utilized to predict the PV power under various weather conditions. The model exhibits great accuracy and stability in short-term PV power prediction, as demonstrated by the experimental results, which were validated using measured data from many PV power plants.

Parameter Estimation and Preliminary Fault Diagnosis for Photovoltaic Modules Using a Three-Diode Model

Accurate estimation of photovoltaic (PV) power generation can ensure the stability of regional voltage control, provide a smooth PV output voltage and reduce the impact on power systems with many PV units. The internal parameters of solar cells that affect their PV power output may change over a period of operation and must be re-estimated to produce a power output close to the actual value. To accurately estimate the power output for PV modules, a three-diode model is used to simulate the PV power generation. The three-diode model is more accurate but more complex than single-diode and two-diode models. Different from the traditional methods, the 9 parameters of the three-diode model are transformed into 16 parameters to further provide more refined estimates. To accurately estimate the 16 parameters in the model, an optimization tool that combines enhanced swarm intelligence (ESI) algorithms and the dynamic crowing distance (DCD) index is used based on actual historical PV power data and the associated weather information. When the 16 parameters for a three-diode model are accurately estimated, the I–V (current-voltage) curves for different solar irradiances are plotted, and the possible failures of PV modules can be predicted at an early stage. The proposed method is verified using a 200 kWp PV power generation system. Three different diode models that are optimized using different ESI algorithms are compared for different weather conditions. The results affirm the reliability of the proposed ESI algorithms and the value of creating more refined estimation models with more parameters. Preliminary fault diagnosis results based on the differences between the actual and estimated I–V curves are provided to operators for early maintenance reference.

Long-Term Solar Power Generation Forecasting at Eastern West Large Scale Solar (LSS) Farm using Random Forest Regression (RFR) Method

Precise forecasting of power generation and demand is essential for effective resource allocation and energy trading in contemporary energy systems. Power forecasting accuracy has increased dramatically since Random Forest Regression (RFR) techniques were used. The study's primary objective is to forecast electricity generation in Malaysia's Eastern West region, with a concentration on solar energy. The research process entails gathering and examining pertinent factors, weather information, and historical power data. To evaluate the accuracy and predictive potential of RFR models, a specific power grid is used for training, validation, and testing. One of the anticipated results is the creation of an accurate model for power generation predictions, which will help to optimise energy operations and smoothly incorporate renewable sources. The paper examines the advantages, disadvantages, and best practices related to RFR-based power forecasting. The dataset, which spans the years 2019 to 2023, includes 30-minute interval records for the following variables: average output power, ambient temperature, PV module temperature, global horizontal irradiance, and wind speed. Using the RandomForestRegressor class from the scikit-learn library, the RFR model is implemented. In order to assess the model's overall fit, average deviation, and sensitivity to outliers, measures such as root mean square error (RMSE), mean square error (MSE), and mean absolute error (MAE) are used on the test set. The temperature, irradiance, and AC power output of PV modules are found to be strongly correlated.

Long-Term Solar Power Generation Forecasting in the Eastern Coast Region of Malaysia using Artificial Neural Network (ANN) Method

Accurate prediction of power demand and generation is crucial for modern energy systems to efficiently allocate resources and facilitate energy trading. The integration of artificial intelligence (AI) and machine learning techniques has significantly improved the precision of power forecasting. This study focuses on the application of Artificial Neural Networks (ANN) for forecasting power generation in the Eastern Coast region of Malaysia, with a specific emphasis on solar power. The research methodology involves collecting and analyzing historical power data, weather data, and relevant variables. ANN models are trained, validated, and tested on a selected power grid to assess their accuracy and predictive capabilities. The expected outcomes aim to include the development of a precise power generation forecasting model, providing valuable insights for decision-makers to optimize energy operations and seamlessly integrate renewable sources. Additionally, the study explores potential challenges, limitations, and best practices associated with ANN-based power forecasting. The dataset covers the period from 2020 to 2023, with variables such as average output power, ambient temperature, PV module temperature, global horizontal irradiance, and wind speed recorded at 30-minute intervals. The architecture of the ANN model, implemented using the Keras framework, is described as a Sequential model with layers utilizing the 'ReLU' activation function. Model evaluation employs metrics like root mean square error (RMSE), mean square error (MSE), and mean absolute error (MAE) on the test set, offering insights into the model's overall fit, average deviation, and sensitivity to outliers. Results reveal strong correlations between PV module temperature, irradiance, and AC power generated.

Optimum Power Forecasting Technique for Hybrid Renewable Energy Systems Using Deep Learning

Power forecasting in large-scale electrical systems, comprising photovoltaic (PV), solar, and wind power, faces challenges due to geographical diffusion and temporal variations. Despite numerous studies, the disparity between predicted and actual generation remains a significant issue. This study utilizes historical power and atmospheric data from diverse plants, employing preprocessing techniques to enhance quality and reduce noise. K-Means clustering is applied to the dataset, optimizing deep learning training periods and increasing accuracy. A resilient hybrid deep learning model is proposed for microgrid (MG) power forecasting, encompassing preprocessing, model training, and assessment stages. Mathematical models for PV systems, battery storage, and wind systems, along with a K-means clustering algorithm, contribute to accurate forecasting. The recurrent neural network based on gated recurrent unit architecture outperforms traditional algorithms, demonstrating superior accuracy, and reduced errors in extensive experimental analyses. Pearson coefficients reveal associations between different power production forms, emphasizing the potential of hybrid renewable energy clusters to enhance forecasting. Case studies illustrate the partial controllability of concentrated solar power production, reducing overall renewable energy cluster unpredictability. The proposed method showcases the efficacy of the hybrid model in addressing challenges and improving accuracy in large-scale power forecasting.

Multiphase Solar Photovoltaic Prediction Model Based on Season, Hierarchical k‐Means Clustering, GRA‐PCC, SVM, and Neural Network

Solar photovoltaic (PV) has accounted for the highest percentage of power generation capacity among other renewables. However, solar PV power generation is highly variable because of different factors; therefore, accurate forecasting is critical for reliable integration into the power system. This paper proposes a multiphase solar PV prediction model that includes grouping, clustering, linking, classifying, and predicting using historical solar PV power and weather data. Seasonal variation is considered in the grouping phase, followed by hybrid hierarchical k‐means clustering to enhance data division in the clustering phase. A hybrid gray relational analysis‐Pearson correlation coefficient identifies significant weather factors impacting solar PV power in the linking phase. The classification phase employs a support vector machine to establish the relationship between the clusters and the relevant weather factors. Lastly, a neural network (NN) is trained to predict solar PV power. The solar PV power profiles are presented to show the variability in season and time. The simulation results of the proposed model showed relatively accurate forecasting results, including MAE of 0.408 MW, MSE of 460.51 MW, RMSE of 0.679 MW, nRMSE of 4.345%, and MRE of 2.266%. These results represent that the uncertainties of the proposed model are 6 and 12 times lower than those of the conventional methods (i.e., conventional NN and ARMAX). These results assure that the proposed model can provide more accurate solar PV power profiles for reliable power system integration.

Wind power forecasting based on manifold learning and a double-layer SWLSTM model

With the increase of wind power installed capacity, wind energy is playing an increasingly important role in power grids. However, this also aggravates the volatility of a power grid. Accurate wind power prediction is particularly important for grid dispatching and grid connection modes. However, the existing forecasting methods mostly consider the meteorological environmental factors and historical power data and do not consider the impact of wind turbine generator parameters. Therefore, this study proposes a double-layer Shared Weight Long Short-Term Memory Network (SWLSTM) prediction model based on manifold learning considering meteorological environment factors, wind turbine generator parameters, and historical wind power. First, 17 indicators such as meteorological environment factors and wind turbine parameters are determined as influencing factors to construct high-dimensional data sets. Second, manifold learning is used to reduce the dimension of high-dimensional data sets and determine the best dimension reduction. The data after dimensionality reduction and historical wind power are used as the input of the prediction model. Finally, considering the influence of errors on the prediction results, a double-layer SWLSTM prediction model is constructed. The double-layer model is constructed simultaneously, and the final power prediction result is obtained by adding the prediction error of the lower model and the prediction wind power of the upper model. Moreover, a wind farm in Xinjiang is used to verify the effectiveness of the proposed prediction model. The results show that the prediction strategy and the model proposed in this paper have high prediction accuracy.

Short‐term prediction of wind power based on temporal convolutional network and the informer model

AbstractIn this study, a new short‐term wind power prediction model based on a temporal convolutional network (TCN) and the Informer model is proposed to solve the problem of low prediction accuracy caused by large wind speed fluctuations in short‐term prediction. First, an input feature selection method based on the maximum information coefficient is proposed after considering the problem of information interference caused by excessively large input features. A dynamic time planning method is used to select the optimal input step of historical power. Then, the combined forecasting model composed of TCN and the Informer is constructed in accordance with the numerical weather forecast and historical power data. Lastly, the pinball loss function is used to expand the prediction model into a quantile regression model, measure the effect of volatility, quantify the volatility range of prediction, and finally, obtain a deterministic prediction result. The actual measured data of wind farms in the Bohai Sea area are selected for analysis and calculation. The results show that the prediction model proposed in this study achieves better accuracy in deterministic prediction and interval prediction than the traditional model.

Optimization of Artificial Intelligence Algorithm based on Neural Network in Complex System

The purpose of this paper is to discuss the application of AI (Artificial Intelligence) algorithm with NN (Neural Network) in complex system optimization. Therefore, this paper takes the optimization of power system as the research object and designs a hybrid NN model. The model can effectively extract features from historical power data and predict future power demand and supply. At the same time, the NN model is optimized by genetic algorithm. The algorithm can efficiently search the optimal solution in a large solution space and has the ability to deal with multi-objective optimization problems. Finally, it is verified by experiments. By using test data sets to evaluate the model, it is found that the algorithm in this paper has high accuracy and applicability in dealing with power system optimization problems. At the same time, the model can effectively reduce the cost of power generation and improve the stability of the system. These achievements provide new ideas and methods for future complex system optimization, and provide useful reference for promoting the development and application of AI technology. In order to provide some guidance for practical application.

Short-term wind power forecasting based on multivariate/multi-step LSTM with temporal feature attention mechanism

Precision enhancement for short-term wind power forecasting can alleviate negative impact of the forecasting results on wind power generation. Due to complexities and nonlinearities among factors and facets in wind power, it is essential to achieve reliable and stable power generation via the long short-term memory (LSTM) forecasting. To this purpose, multi-task temporal feature attention (MTTFA) based LSTM, namely MTTFA-LSTM, is proposed for multivariate/multi-step wind power forecasting with historical power and meteorological data, in which task-sharing and task-specifying layers are designed for task co-features extracting and task specifics discriminating, respectively. More specifically, in the task-sharing layer, multi-dimensional inputs are fed into LSTM to extract long-term trends, while in the task-specifying layer, one-dimensional convolution operations extract temporal features hidden in each and all time steps. Furthermore, an attention mechanism is adopted to adaptively tune weights for temporal features. Finally, the proposed model is leveraged to cope with different short-term wind power forecasting (SWPF) problems based on the national renewable energy laboratory’s (NREL) wind power data. Simulation results show that the proposed MTTFA-LSTM achieves persistent excellent forecasting accuracy, comparing its backbone STL model, TFA-LSTM as well as the benchmark MTL models in the same setting, which indicate that the complex and non-linear interdependencies among multi-dimensional data can be well depicted by the proposed model.

An upscaling minute-level regional photovoltaic power forecasting scheme

Along with the increasing penetration of photovoltaic (PV) power generation, regional power forecasting becomes more and more critical for stable and economical operation of power system. The key challenge of regional PV power forecasting technology is the lack of complete and accurate historical power data since not all PV plants are equipped with the precise real-time output power monitoring system. Besides, the computation burden will be heavy when the number of PV plants in the target region is large. This paper therefore proposes an upscaling minute-level regional PV power forecasting scheme using the data of the selected reference PV plants. In this paper, a novel method of reference PV plants selection is proposed by comprehensively considering the prediction accuracy of artificial neural network (ANN) as well as Pearson correlation coefficient. The reference PV plant selection coefficient μ is introduced as the comprehensive indicator for reference PV plant selection, which incorporates Pearson correlation coefficient and MAPE. In addition, a PV output power correction method is assumed to guarantee the proper operation of regional power forecasting. Besides, this paper proposes a flexible approach to effectively decrease the accumulated error of rolling forecasting by integrating the forecasting results under different temporal resolutions. In specific, the power forecasting results in temporal resolutions of 1 min, 5 min and 15 min are simultaneously derived and the performance between the traditional rolling forecasting and the proposed method is compared. The validity of the proposed method is finally verified using the collected historical power data of PV plants installed in a city of Eastern China. For time resolution of 1 min, 5 mins and 10 mins, the corresponding RMSE are 6.56, 5.73 and 4.85 and corresponding MAPE are 4.04%, 3.45% and 2.86%.

Short‐term photovoltaic power forecasting using hybrid contrastive learning and temporal convolutional network under future meteorological information absence

AbstractPhotovoltaic (PV) power generation is widely utilized to satisfy the increasing energy demand due to its cleanness and inexhaustibility. Accurate PV power forecasting can improve the penetration of PV power in the grid. However, it is pretty challenging to predict PV power in short‐term under precious future meteorological information absence conditions. To address this problem, this study proposes the hybrid Contrastive Learning and Temporal Convolutional Network (CL‐TCN), and this forecasting approach consists of two parts, including model training and adaptive processes of forecasting models. In the model training stage, this forecasting method firstly trains 18 TCN models for 18 time points from 9:00 a.m. to 17:30 p.m. These TCN models are trained by only using historical PV power data samples, and each model is used to predict the next half‐hour power output. The adaptive process of models means that, in a practical forecasting stage, PV power samples from historical data are firstly evaluated and scored by a CL based data scoring mechanism to search for the most similar data samples to current measured samples. Then these similar samples are further applied to training a single above‐mentioned well‐trained TCN model to improve its performance in forecasting the next half‐hour PV power. The experimental results tested at the time resolution of 30 min demonstrate that the proposed approach has superior performance in forecasting accuracy not only in smooth PV power samples but also in fluctuating PV power samples. Moreover, the proposed CL based data scoring mechanism can filter useless data samples effectively accelerating the forecasting process.

A design of ultra-short-term power prediction algorithm driven by wind turbine operation and maintenance data for LSTM-SA neural network

Due to factors such as meteorology and geography, the generated power of wind turbines fluctuates frequently. In this way, power changes should be predicted in grid connection to take control measures in time. In this paper, an operation and maintenance data-driven LSTM-SA (long short-term memory with self-attention) prediction algorithm is designed to predict the ultra-short-term power of wind turbines. First, the wind turbine operation and maintenance data, including wind speed, blade deflection angle, yaw angle, humidity, and temperature, are subjected to feature selection by using the Pearson correlation coefficient method and the Lasso algorithm, thereby establishing the correlation between wind speed, blade deflection angle, and out power. Then, full-connect neural network is trained to establish a mapping model of wind speed, blade deflection angle, and out power. The power change rate k is calculated by the derivative of output power to wind speed. Finally, based on the historical power data and the power change rate k, the LSTM neural network power prediction model is trained to calculate the output power prediction value. In order to increase the training efficiency and reduce the delay, the self-attention mechanism is used to optimize the hidden layer of the LSTM model. The test results show that, compared with similar prediction algorithms, this algorithm has higher prediction accuracy, faster convergence speed, and better stability, which can solve the problem of accurately predicting ultra-short-term power when wind power training data is inadequate.

Bayesian uncertainty quantification framework for wake model calibration and validation with historical wind farm power data

AbstractThe expected growth in wind energy capacity requires efficient and accurate models for wind farm layout optimization, control, and annual energy predictions. Although analytical wake models are widely used for these applications, several model components must be better understood to improve their accuracy. To this end, we propose a Bayesian uncertainty quantification framework for physics‐guided data‐driven model enhancement. The framework incorporates turbulence‐related aleatoric uncertainty in historical wind farm data, epistemic uncertainty in the empirical parameters, and systematic uncertainty due to unmodeled physics. We apply the framework to the wake expansion parametrization in the Gaussian wake model and employ historical power data of the Westermost Rough Offshore Wind Farm. We find that the framework successfully distinguishes the three sources of uncertainty in the joint posterior distribution of the parameters. On the one hand, the framework allows for wake model calibration by selecting the maximum a posteriori estimators for the empirical parameters. On the other hand, it facilitates model validation by separating the measurement error and the model error distribution. In addition, the model adequacy and the effect of unmodeled physics are assessable via the posterior parameter uncertainty and correlations. Consequently, we believe that the Bayesian uncertainty quantification framework can be used to calibrate and validate existing and upcoming physics‐guided models.

A novel GRU-TCN network based Interactive Behavior Learning of multi-energy Microgrid under incomplete information

The interactive operation of multi-energy Microgrid (MEMG) in energy market is facing more and more incomplete information decision-making requirements due to the enhanced privacy of users, which brings great challenges to the participation flexibility of MEMG in demand response programs. Meanwhile, considering the uncertainty of the external MEMG unit combination, the randomness of the environmental meteorology, the interactive characteristics of the external MEMGs are complex and time-varying. To solve these problems, a novel GRU-TCN network based Interactive Behavior Learning method for MEMG interactive operation is proposed. The proposed Interactive Behavior Learning framework needs merely externally available input information and historical interaction power data to efficiently predict the interaction power, which protects the data privacy of users. Besides, the GRU-TCN network also leverages the strengths of Recurrent Neural Network and Convolutional Neural Networks while incorporating the self-attention mechanism. This combination enables the GRU-TCN network to effectively capture the relationship between interactive power and the available resources data beyond the MEMG. The example analysis and test are carried out on a typical MEMG system, and the case study turns out that the proposed GRU-TCN network has the capability to capture coupling mechanisms even in the presence of incomplete information and present more accurate prediction of multi-energy interactive data.

Hybrid model based on K-means++ algorithm, optimal similar day approach, and long short-term memory neural network for short-term photovoltaic power prediction

Photovoltaic (PV) power generation is characterized by randomness and intermittency due to weather changes. Consequently, large-scale PV power connections to the grid can threaten the stable operation of the power system. An effective method to resolve this problem is to accurately predict PV power. In this study, an innovative short-term hybrid prediction model (i.e., HKSL) of PV power is established. The model combines K-means++, optimal similar day approach, and long short-term memory (LSTM) network. Historical power data and meteorological factors are utilized. This model searches for the best similar day based on the results of classifying weather types. Then, the data of similar day are inputted into the LSTM network to predict PV power. The validity of the hybrid model is verified based on the datasets from a PV power station in Shandong Province, China. Four evaluation indices, mean absolute error, root mean square error (RMSE), normalized RMSE, and mean absolute deviation, are employed to assess the performance of the HKSL model. The RMSE of the proposed model compared with those of Elman, LSTM, HSE (hybrid model combining similar day approach and Elman), HSL (hybrid model combining similar day approach and LSTM), and HKSE (hybrid model combining K-means++, similar day approach, and LSTM) decreases by 66.73%, 70.22%, 65.59%, 70.51%, and 18.40%, respectively. This proves the reliability and excellent performance of the proposed hybrid model in predicting power.

A Novel Forecasting Model for Solar Power Generation by a Deep Learning Framework With Data Preprocessing and Postprocessing

Photovoltaic power has become one of the most popular forms of energy owing to the growing consideration of environmental factors; however, solar power generation has brought many challenges for power system operations. With regard to optimizing safety and reducing the costs of power system operations, an accurate and reliable solar power forecasting model would be a significant step forward. This study proposes a deep learning method to improve the performance of short-term one-hour-ahead solar power forecasting, which includes data preprocessing, feature engineering, kernel principal component analysis, a gated recurrent unit network training mode based on time-of-day classification, and postprocessing with error correction. Both historical solar power, solar irradiance, and numerical weather prediction (NWP) data, such as temperature, irradiance, rainfall, wind speed, air pressure, and humidity, were used as the input dataset in this work. As a case study, the measured power from ten PV sites in Taiwan were collected and predicted with a one-hour resolution. The normalized root mean squared error and normalized mean absolute percent error were chosen to evaluate the performance of the forecasting models. Compared with other benchmark models, including ANN, LSTM, XGBoost, and single GRU, the experimental results showed the proposed model's superior performance. Furthermore, the importance of data preprocessing and postprocessing based on error correction was demonstrated.

An Effective Hybrid Symbolic Regression–Deep Multilayer Perceptron Technique for PV Power Forecasting

The integration of Photovoltaic (PV) systems requires the implementation of potential PV power forecasting techniques to deal with the high intermittency of weather parameters. In the PV power prediction process, Genetic Programming (GP) based on the Symbolic Regression (SR) model has a widespread deployment since it provides an effective solution for nonlinear problems. However, during the training process, SR models might miss optimal solutions due to the large search space for the leaf generations. This paper proposes a novel hybrid model that combines SR and Deep Multi-Layer Perceptron (MLP) for one-month-ahead PV power forecasting. A case study analysis using a real Australian weather dataset was conducted, where the employed input features were the solar irradiation and the historical PV power data. The main contribution of the proposed hybrid SR-MLP algorithm are as follows: (1) The training speed was significantly improved by eliminating unimportant inputs during the feature selection process performed by the Extreme Boosting and Elastic Net techniques; (2) The hyperparameters were preserved throughout the training and testing phases; (3) The proposed hybrid model made use of a reduced number of layers and neurons while guaranteeing a high forecasting accuracy; (4) The number of iterations due to the use of SR was reduced. The presented simulation results demonstrate the higher forecasting accuracy (reductions of more than 20% for Root Mean Square Error (RMSE) and 30 % for Mean Absolute Error (MAE) in addition to an improvement in the R2 evaluation metric) and robustness (preventing the SR from converging to local minima with the help of the ANN branch) of the proposed SR-MLP model as compared to individual SR and MLP models.

An approach for day-ahead interval forecasting of photovoltaic power: A novel DCGAN and LSTM based quantile regression modeling method

In order to effectively quantify the uncertainty of PV power and improve the forecasting accuracy, a day-ahead interval forecasting method of PV power based on multi-correlation parameter scenarios generation is proposed. The historical PV power data is divided into a limited set of scenarios representing different output and fluctuation characteristics through the K-means clustering algorithm; Combined with the strong correlation feature parameters determined by the correlation coefficient, the characteristic curves of PV power and multi-correlation parameter in different scenarios are generated through the deep convolutional generative adversarial networks (DCGAN) model; The average value of each characteristic curve and other explanatory variables are input together into a quantile regression long short-term memory (QRLSTM) model to achieve day-ahead PV power forecasting, and obtain the forecasting interval under different confidence levels. The results of the experiment of a PV power station in Yangzhou, Jiangsu Province, China show that the forecasting performance of the proposed method under different evaluation indicators is 17.9755% and 18.1571% higher than that of the Gaussian process regression (GPR) model and the single QRLSTM model, respectively, which has obvious advantages.