Xu Weight is All that Models Need! A Short-Term Power Load Forecasting Method Based on a Novel Adaptive Feature Selection Method and Xu Weight DOI

Jingqi Xu,

Xueman Wang,

Hui Hou

и другие.

Опубликована: Янв. 1, 2024

Язык: Английский

Distributed PV carrying capacity prediction and assessment for differentiated scenarios based on CNN-GRU deep learning DOI Creative Commons
Liudong Zhang, Zhen Lei, Zhigang Ye

и другие.

Frontiers in Energy Research, Год журнала: 2024, Номер 12

Опубликована: Сен. 23, 2024

The increasing penetration of distributed photovoltaic (PV) brings challenges to the safe and reliable operation distribution networks, PV access grid changes characteristics traditional grid. Therefore, assessment carrying capacity is great significance for network planning. To this end, a differentiated scenario-based method based on combination Convolutional Neural Networks (CNN) Gated Recurrent Unit (GRU) proposed. Firstly, meteorological affecting power are quantitatively analyzed using Pearson’s correlation coefficient, influence external factors assessed by integrating measured data. Then, problem high blindness clustering parameters initial centers in K-means algorithm, optimal number clusters determined combining cluster Density Based Index (DBI) hierarchical clustering. improved reduces complexity massive scenarios obtain under scenarios. On basis, prediction CNN-GRU model proposed, which employs CNN feature extraction high-dimensional data, then temporal data optimally trained GRU model. results, solution efficiency effectively accurate realized. Finally, taking into account demand network, combined with flow calculation bearing considering node voltage evaluated. In addition, verified source-grid-load IEEE 33-bus system. simulation results show that proposed fusion deep learning can accurately efficiently assess provide theoretical guidance realizing large scale.

Язык: Английский

Процитировано

1

A comparative study of different deep learning methods for time-series probabilistic residential load power forecasting DOI Creative Commons

Liangcai Zhou,

Yi Zhou, Linlin Liu

и другие.

Frontiers in Energy Research, Год журнала: 2024, Номер 12

Опубликована: Окт. 18, 2024

The widespread adoption of nonlinear power electronic devices in residential settings has significantly increased the stochasticity and uncertainty systems. original load data, characterized by numerous irregular, random, probabilistic components, adversely impacts predictive performance deep learning techniques, particularly neural networks. To address this challenge, paper proposes a time-series prediction technique based on mature network point technique, i.e., decomposing data into deterministic stochastic components. component is predicted using technology, fitted with Gaussian mixture distribution model parameters are great expectation algorithm, after which obtained generation method. Using study evaluates six different methods to forecast power. By comparing errors these methods, optimal identified, leading substantial improvement accuracy.

Язык: Английский

Процитировано

0

The phenomenon and suppression strategy of overvoltage caused by PV power reverse flow DOI Creative Commons

Yumeng Xie,

Qiying Li,

Lei Yang

и другие.

Frontiers in Energy Research, Год журнала: 2024, Номер 12

Опубликована: Ноя. 18, 2024

In the current distribution network’s energy structure, photovoltaic (PV) occupies a high proportion. However, access of proportion PV will lead to phenomenon reverse power flow in network, and then problem line overvoltage. When increases, overvoltage also worsens, which endangers normal operation system. To solve this problem, paper starts with voltage rise theory network lines. Firstly, through strict mathematical derivation, it compares influence main parameters on rise, summarizes simple calculation equation for PV. Then, according mechanism principle inverter control, considering economy practicability suppression strategy, strategy system is proposed. Finally, model simulating small village used verify effectiveness proposed strategy.

Язык: Английский

Процитировано

0

Comparison of support strategies for grid construction after failure of renewable energy power generation system DOI Creative Commons
Ke Li,

Xia Lin,

S. Zhu

и другие.

Frontiers in Energy Research, Год журнала: 2024, Номер 12

Опубликована: Ноя. 27, 2024

As the proportion of renewable energy generation continues to rise, study voltage source converter (VSC) control has become a focal point research. The concepts emulating characteristics synchronous machines have led proposals droop and virtual (VSG). However, deeper comparison these two methods is still needed, particularly in terms their ability support system when partial power sources experience fault conditions. This paper analyzes compares principles small-signal modeling, finally, based on nine-bus with 100% generation, scenarios are designed: sudden load increase disconnection. differences between compared analyzed. results indicate that VSG exhibits greater damping capable providing inertial system, making its frequency less susceptible change.

Язык: Английский

Процитировано

0

Xu Weight is All that Models Need! A Short-Term Power Load Forecasting Method Based on a Novel Adaptive Feature Selection Method and Xu Weight DOI

Jingqi Xu,

Xueman Wang,

Hui Hou

и другие.

Опубликована: Янв. 1, 2024

Язык: Английский

Процитировано

0