一种高铁站区电网负荷的预测方法、装置及存储介质
By constructing a multi-level learning network prediction model, the complexity of power grid load prediction in high-speed railway station areas is solved, achieving high accuracy and good performance in power grid load prediction, and supporting power grid risk assessment.
CN115470964BActive Publication Date: 2026-07-17STATE GRID HEBEI ELECTRIC POWER CO LTD XIONGAN NEW DISTRICT POWER SUPPLY CO +4
Patent Information
- Authority / Receiving Office
- CN · China
- Patent Type
- Patents(China)
- Current Assignee / Owner
- STATE GRID HEBEI ELECTRIC POWER CO LTD XIONGAN NEW DISTRICT POWER SUPPLY CO
- Filing Date
- 2022-08-04
- Publication Date
- 2026-07-17
AI Technical Summary
Technical Problem
Existing power grid load forecasting methods are too complex and perform poorly, and cannot be effectively applied to the complex potential dynamic characteristics of power grid load in high-speed railway station areas.
Method used
A multi-level learning network prediction model is adopted. By preprocessing the training dataset and optimizing the two-level penalty network, a prediction model with multi-level learning mapping capability is constructed to extract the complex potential dynamic characteristics of the power grid load.
Benefits of technology
It improves the accuracy and generalization ability of power grid load forecasting in high-speed railway station areas, and provides the necessary conditions for assessing potential risks to the power grid.
✦ Generated by Eureka AI based on patent content.
Smart Images
Figure CN115470964B_ABST
Abstract
本发明公开了一种高铁站区电网负荷的预测方法、装置及存储介质，其方法包括：获取高铁站区采样点在当前采样周期的电网负荷；将当前采样周期的电网负荷输入训练好的多层次学习网络预测模型，预测下一采样周期的电网负荷；其中，所述多层次学习网络预测模型的训练包括：获取高铁站区采样点在多个连续采样周期的电网负荷，并生成训练数据集；对训练数据集进行预处理，估计最优时延和嵌入维度；根据训练数据集、最优时延和嵌入维度对预构建的多层次学习网络预测模型进行训练；本发明可以有效地提高多层次学习网络对电网负荷的泛化能力，具有良好的预测性能。
Need to check novelty before this filing date? Find Prior Art