With the increasing proportion of distributed photovoltaic (DPV) installations in county-level power grids, to improve the centralized operation and maintenance of the stations and to meet the needs of power grid dispatching, the output of the county-level regional DPV stations group needs to be predicted. In this paper, the weather prediction information is used to predict the output based on the model input average strategy. To eliminate the effect of the selected non-optimal training sample collection period on the prediction accuracy, an ensemble prediction method based on the minimum redundancy maximum relevance criterion and TabNet model is carried out. To reduce the influence of weather prediction errors on the power output prediction, a modified model based on error prediction is proposed. The ensemble prediction model is used to predict the day-ahead output, and a combination prediction model based on the proposed ensemble prediction model and the proposed modified model is established to predict the hour-ahead output. The experimental results verify the effectiveness of the proposed models. Compared with the corresponding reference models, the proposed ensemble prediction method reduces the normalized mean absolute errors (nMAEs) and the normalized root mean square errors (nRMSEs) of the day-ahead output prediction results by 2.86% and 5.51%, respectively. The combination prediction model reduces the nMAE and nRMSE of the hour-ahead output prediction results by 3.05% and 3.05%, respectively. Therefore, the prediction accuracy can be improved by the proposed models.

Traditional short-term solar power forecasting primarily focuses on individual photovoltaic (PV) plants. Recently, there has been increasing demand for power forecasting of PV clusters. In this paper, a distributed PV cluster power prediction model based on statistical upscaling and convolutional block attention module (CBAM)–bi-directional long short term memory (BiLSTM) is proposed to strike a balance between prediction accuracy and model training time. To address the issue of inaccurate cluster partitioning caused by variations in distributed PV output curves, a K-shape based cluster partitioning method is proposed. This method effectively reduces the number of prediction scenarios. Furthermore, a soft-dynamic time warping based representative power plants selection method is introduced to accurately identify representative power plants from sub-clusters. This selection method takes into account the horizontal and vertical stretching of the output curves, ensuring a comprehensive reflection of curve similarity. To minimize the conversion error during cluster transformation, a real-time statistical upscale conversion method is proposed. This method considers multiple similar output days, resulting in more accurate sub-cluster output predictions. The hybrid model, which employs CBAM for initial feature extraction and BiLSTM for output forecasting, is introduced to predict the output of representative power stations. Finally, the effectiveness of the proposed model is verified through evaluation indices, with the mean absolute percentage error value of the representative power station being less than 5%. The accuracy of the results is further supported by the confidence interval analysis.

Photovoltaic systems, as a critical component of microgrids, require accurate power forecasting for the optimization of microgrid scheduling. To enhance data feature representation and capture the fluctuating patterns of photovoltaic power, an improved power ramp feature construction method is proposed for feature engineering and selection, thereby improving the quality of photovoltaic data. Furthermore, a photovoltaic power prediction model based on a convolutional attention long short-term memory hybrid framework is employed to enhance forecasting performance. Building upon this, a novel photovoltaic power dynamic interval prediction method based on power ramp clustering is proposed to address the issues of low prediction accuracy and excessively broad ranges in existing photovoltaic power interval prediction methods. First, an improved power ramp calculation formula is used to identify and statistically analyze the power ramp, clustering daily data into four weather categories—sunny, partly cloudy, cloudy, and rainy—thus enhancing the accuracy of similar day clustering. Next, the shape dynamic time warping algorithm is applied to select similar days for the target day, and the dynamic interval calculation is performed to determine the power range. Finally, experimental analysis is conducted using real photovoltaic output historical data from a region in Xinjiang, China. The results demonstrate that the proposed method improves prediction accuracy, effectively reduces the width of the predicted range, and more accurately captures the fluctuating trends of photovoltaic power.

Frequency regulation of wind turbines can improve the stability of the power system. However, it would cause generator torque fluctuation, increasing the risk of fatigue load. Previous research works were restricted to inertial and droop responses. Less attention has been paid to striking a balance between frequency regulation and fatigue load. To overcome these, a hybrid control strategy is proposed to consider both frequency response characteristic and fatigue load mitigation. First, a small signal linearization model is built to reveal the impact of the correlation mechanism of frequency regulation on drive train torque load. Second, a multivariable cost function is constructed to optimize the proportional integral (PI) controller, which combines the total fluctuation with the dispersion of the fatigue load and frequency. Then, a hybrid controller based on PI control optimized by particle swarm optimization algorithm and active disturbance rejection control is designed to restrain rapid frequency changes as well as fatigue torque fluctuation simultaneously. Several experiments are performed to verify the significance of the proposed method under different scenarios. Compared with the existing methods, the proposed hybrid control exhibits superiority in improving frequency response and fatigue load mitigation.