基于ISSA-LSTM模型的太阳能热发电站并网发电量预测方法研究

吕达; 张超; 赵文涛

doi:doi:10.19912/j.0254-0096.tynxb.2025-0676

English Version Reader LoginJournal Login

您当前的位置：

首页 >

文章列表页 >

基于ISSA-LSTM模型的太阳能热发电站并网发电量预测方法研究

更新时间：2026-01-08

- 基于ISSA-LSTM模型的太阳能热发电站并网发电量预测方法研究
- Vol. 46, Issue 12, Pages: 359-368(2025)
- 作者机构：
- 作者简介：
- 基金信息：
- DOI：doi:10.19912/j.0254-0096.tynxb.2025-0676
  CLC： TP183
- Published Online：07 January 2026，
  
  Published：2025
- 稿件说明：
移动端阅览
吕达, 张超, 赵文涛. 基于ISSA-LSTM模型的太阳能热发电站并网发电量预测方法研究[J]. 2025, 46(12): 359-368.
DOI：

吕达, 张超, 赵文涛. 基于ISSA-LSTM模型的太阳能热发电站并网发电量预测方法研究[J]. 2025, 46(12): 359-368. DOI： doi:10.19912/j.0254-0096.tynxb.2025-0676.

摘要

为准确预测太阳能热发电站并网发电量

提出一种改进后的麻雀搜索算法与长短期记忆网络相结合的模型（ISSA-LSTM模型）。首先

基于传统LSTM模型引入麻雀搜索算法构建SSA-LSTM预测模型

以突破局部最优陷阱

提升全局搜索能力。其次

运用精英反向学习策略生成反向解并获取精英个体动态边界

对SSA-LSTM预测模型进行改进

进一步提升算法全局搜索能力与搜索精度。最后

运用太阳能热发电站真实数据对所构建模型与其他模型进行训练分析

并对比各模型预测结果。经过多次训练后结果显示：ISSA-LSTM模型预测结果精度显著高于其他模型

对比SSA-LSTM模型预测精度提升14%;同时该模型训练结果基本与实际数值相吻合。该文基于ISSA-LSTM模型

对光热发电站并网发电量进行预测

为全面提高发电量预测精度、合理布局太阳能热发电站提供一定理论借鉴。

Abstract

To accurately forecast the grid connected power generation of solar thermal power stations

an improved sparrow search algorithm combined with long short-term memory networks （ISSA-LSTM model） is proposed. Firstly

based on the traditional LSTM model

the sparrow search algorithm is introduced to construct the SSA-LSTM forecast model

in order to break through the local optimal trap and enhance the global search capability. Secondly

using the elite opposition-based learning strategy to generate reverse solutions and obtain dynamic boundaries of elite individuals

the SSA-LSTM forecast model is improved to further enhance the algorithm's global search capability and search accuracy. Finally

the constructed model is trained and analyzed using real data from the solar thermal power station

and the forecast results of each model are compared with those of other models. After multiple training sessions

the results show that the ISSA-LSTM model has significantly higher forecast accuracy than other models

with a 14% improvement in forecast accuracy compared to the SSA-LSTM model. At the same time

the training results of the model are basically consistent with the actual values. Based on the ISSA-LSTM model

this article predicts the grid connected power generation of solar thermal power plants

providing a theoretical reference for comprehensively improving the accuracy of power generation pre and rational layout of solar thermal power stations.

关键词

Keywords

references

ZHANG M J, ANABA O A, MA Z Q, et al.En route to attaining a clean sustainable ecosystem: a nexus between solar energy technology, economic expansion and carbon emissions in China[J]. Environmental science and pollution research, 2020, 27(15): 18602-18614.

方宇晨, 杜尔顺, 余扬昊, 等. 太阳能光热发电并网的综合效益量化评估方法[J]. 中国电机工程学报, 2024, 44(13): 5135-5146.

杜尔顺, 张宁, 康重庆, 等. 太阳能光热发电并网运行及优化规划研究综述与展望[J]. 中国电机工程学报, 2016, 36(21): 5765-5775, 6019.

曾光, 纪阳, 符津铭, 等. 热储能技术研究现状、热点趋势与应用进展[J]. 中国电机工程学报, 2023, 43(S1): 127-142.

ZJAVKA L.Photovoltaic power one-day and multistep-hourly AI predictions using node-by-node evolved binomial tree structures to form L-transformed PDE modules[J]. Soft computing, 2025, 29(5): 2483-2495.

余洋, 陈庚, 余佳磊, 等. 基于聚类和长短期记忆神经网络的光热电站并网电力预测[J]. 热力发电, 2021, 50(9): 128-136.

刘振路, 郭军红, 李薇, 等. 基于FCM-LSTM的光热发电出力短期预测[J]. 工程科学学报, 2024, 46(1): 178-186.

路小娟, 董海鹰. 太阳能热发电集热系统终端受限非线性模型预测控制[J]. 热力发电, 2015, 44(10): 63-67.

彭曙蓉, 陈慧霞, 孙万通, 等. 基于改进LSTM的光伏发电功率预测方法研究[J]. 太阳能学报, 2024, 45(11): 296-302.

王东风, 刘婧, 黄宇, 等. 结合太阳辐射量计算与CNN-LSTM组合的光伏功率预测方法研究[J]. 太阳能学报, 2024, 45(2): 443-450.

杨轶航, 韩璐, 史华勃, 等. 基于相似日与ISC-BiLSTM的短期光伏功率预测方法[J]. 太阳能学报, 2025, 46(1): 676-685.

王晓霞, 俞敏, 冀明, 等. 基于气候相似性与SSA-CNN-LSTM的光伏功率组合预测[J]. 太阳能学报, 2023, 44(6): 275-283.

薛建凯. 一种新型的群智能优化技术的研究与应用[D]. 上海: 东华大学, 2020.

陈立万, 赵尚飞, 曾蝶, 等. 基于混合策略麻雀搜索算法的WSN覆盖优化[J]. 电子测量技术, 2022, 45(23): 174-180.

王瑞, 闫方, 逯静, 等. 运用相似日和LSTM的短期负荷双向组合预测[J]. 电力系统及其自动化学报, 2022, 34(1): 93-99.

姚玉璧, 郑绍忠, 杨扬, 等. 中国太阳能资源评估及其利用效率研究进展与展望[J]. 太阳能学报, 2022, 43(10): 524-535.

纪文波, 徐建新, 王华, 等. 基于三态测试的生物柴油喷射火焰混沌识别及燃烧性能研究[J]. 太阳能学报, 2021, 42(11): 403-409.

李青. 基于注意力时间卷积神经网络的光伏功率概率预测[J]. 太阳能学报, 2025, 46(2): 326-332.

蒲晓云, 杨靖, 杨兴, 等. 基于分解技术的IZOA-Transformer-BiGRU短期风电功率预测[J]. 电子测量技术, 2025, 48(2): 39-48.

Views

下载量

CSCD

Alert me when the article has been cited

提交

Tools

Publicity Resources

ULTRA-SHORT-TERM WIND SPEED PREDICTION BASED ON SAMPLE ENTROPY-BASED DUAL DECOMPOSITION AND SSA-LSTM

基于多元气象信息和改进组合神经网络的分布式光伏短期功率预测模型

基于相似日聚类和SSA-LSTM的光伏组件积灰损失预测

基于BP神经网络PI和准PR控制的零序环流抑制策略

Related Author

Wang Xi

Sun Na

Su Hao

Zhang Nan

Zhang Shuai

Ji Jie

吴伟丽

米婵

Related Institution

Key Laboratory of Thermo-Fluid Science & Engineering of MOE, Xi’an Jiaotong University

Faulty of Automation, Huaiyin Institute of Technology, Huaian

School of Mechanical Engineering, Shenyang University of Technology

School of Electrical and Power Engineering, Taiyuan University of Technology, Key Laboratory of Clean Intelligent Control of Coal Power, Ministry of Education

Beijing Kedong Electric Power Control System Co., Ltd.

AI问答

⁰