Energy load forecasting of highway facilities in response to integration transportation and energy needs
Article Text (Baidu Translation)
-
摘要: 为准确预测用能负荷，通过调研高速公路典型构造物用能负荷历史数据，分析了交通量、天气状况、星期、月份以及节假日/工作日等多重因素对用能负荷的影响，采用主成分分析(PCA)方法对这些影响因素进行降维处理，消除原始序列的冗余性；分析了节假日/工作日属性和天气状况等因素对用能负荷特征的影响，提出了高速公路用能负荷预测模型候选数据集(CD)构建策略，在此基础上，采用长短时记忆(LSTM)网络模型对多变量用能负荷预测进行了动态时间建模，并依托广西桂柳高速公路实测数据对提出的预测模型进行了验证。分析结果表明：交通量、天气状况、星期、月份、节假日/工作日这5个主成分累计贡献率为85.54%，是高速公路用能负荷主要影响因素；隧道、收费站和服务区用能负荷的高峰时段分布各异，隧道用能负荷高峰时间波动较小，收费站用能负荷高峰集中在10:00~21:00，服务区用能负荷高峰时间段最短，仅集中出现在11:00~12:00；在不同季节测试日中，采用候选数据集构建策略及PCA处理后，提升了训练集数据的针对性，预测结果的精度也得到了提高，平均绝对百分比误差不超过12.33%，均方根误差不超过3.86，提出的预测模型在不同典型场景的负荷预测中均具有良好的适用性，可为高速公路自洽能源系统设计提供理论支撑。Abstract: To accurately predict the energy load, historical energy load data of typical highway structures were investigated, and the influences of multiple factors on energy load, including traffic volume, weather condition, week, month, and holiday/workday, were analyzed. The principal component analysis (PCA) method was applied to reduce the dimensionality of these influencing factors, thereby eliminating the redundancy in original sequences. The effects of holiday/workday attributes and weather condition on energy load characteristics were examined, and a strategy for constructing a candidate dataset (CD) for energy load forecasting models of highways was proposed. On this basis, the dynamic time modeling for multivariate energy load forecasting was performed by using a long short-term memory (LSTM) network model. The proposed forecasting model was validated by using empirical data from the Guilin-Liuzhou Highway in Guangxi. Analysis results show that the cumulative contribution rate of the five principal components of traffic volume, weather condition, week, month, and holiday/workday, is 85.54%, and they are the primary influencing factors of energy load on highways. The peak energy load periods for tunnels, toll stations, and service areas vary, with tunnel energy load peaks showing minimal fluctuations, toll station peaks concentrated between 10:00-21:00, and service area peaks occurring within the shortest period, only concentrated between 11:00-12:00. Across different seasonal test days, the application of CD construction strategy and PCA processing improves the specificity of the training data, resulting in enhanced prediction accuracy, with the mean absolute percentage error not exceeding 12.33% and the root mean square error not exceeding 3.86. The proposed prediction model demonstrates strong applicability for load prediction in various typical scenarios and provides theoretical support for the design of self-consistent energy systems for highways.
-
图 6 隧道不同节假日/工作日用能负荷曲线
Figure 6. Energy load curves of tunnels on different holidays/weekdays
图 7 收费站不同节假日/工作日用能负荷曲线
Figure 7. Energy load curves of toll stations on different holidays/weekdays
图 8 服务区不同节假日/工作日用能负荷曲线
Figure 8. Energy load curves of different holidays/weekdays in service areas
表 2 不同日期属性对应天气的天数
Table 2. Numbers of days corresponding to different date attributes for weather
d 天气 晴 多云 阴 阵雨 小雨 中雨 大雨 暴雨 总计 工作日 23 116 13 21 44 14 9 3 243 周末 9 46 3 5 17 2 1 1 84 3天假期 7 2 9 4天及以上假期 3 9 1 2 4 19 特殊日 2 3 3 2 10 合计 35 180 20 31 65 18 10 6 365 
下载: 导出CSV
表 3 用能负荷影响因素
Table 3. Influencing factors of energy load
名称 描述 月份 1~12月 季节 4个季节 星期 周一至周日 节假日/工作日 工作日/非工作日 交通量 交通量数据 最高温度 单日最高温度 平均温度 单日平均温度 最低温度 单日最低温度 天气状况 晴、多云、阴、阵雨、小雨、中雨、大雨、暴雨等
下载: 导出CSV
表 4 用能负荷影响因素主成分分析
Table 4. Principal component analysis of influencing factors of energy load
主成分排名 名称 特征值 贡献率/% 1 交通量 3.75 37.60 2 天气状况 1.50 15.04 3 星期 1.26 12.64 4 月份 1.03 10.36 5 节假日/工作日 0.98 9.90
下载: 导出CSV
表 5 前5个主成分特征值对应的特征向量
Table 5. Eigenvectors corresponding to first five principal component eigenvalues
主成分1 主成分2 主成分3 主成分4 主成分5 0.368 0.142 0.256 -0.408 0.079 -0.010 0.113 0.467 0.654 0.253 0.095 -0.250 -0.621 -0.073 -0.058 0.004 -0.079 -0.260 -0.028 0.944 -0.196 -0.614 -0.083 0.382 -0.106 0.284 -0.527 0.170 0.016 -0.044 -0.489 -0.014 0.095 -0.189 0.034 -0.495 0.044 -0.001 -0.095 0.001 -0.499 0.013 0.050 -0.147 0.018 0.050 0.489 -0.462 0.430 -0.140
下载: 导出CSV
表 6 用能负荷预测模型参数
Table 6. Parameters of energy load forecasting model
参数名称 输入层时间步数 输入层维数 隐藏层数目 每个隐藏层维数 输出变量维数 参数值 5 5 1 100 24
下载: 导出CSV
表 7 隧道用能负荷预测结果
Table 7. Forecasting results of tunnels energy load
预测模型 春季 夏季 秋季 冬季 4月3日 6月3日 8月3日 10月3日 2月3日 12月3日 M/% R M/% R M/% R M/% R M/% R M/% R LSTM 46.25 32.93 31.16 41.93 41.93 28.54 39.45 29.17 41.93 30.20 36.41 27.47 PCA-LSTM 45.93 32.44 30.44 41.84 41.84 29.03 38.77 28.62 41.84 30.46 35.26 26.05 CD-LSTM 7.85 2.17 4.88 7.64 7.64 3.12 6.88 2.47 7.64 2.53 3.98 1.37 CD-PCA-LSTM 7.15 2.07 2.42 5.00 5.00 2.22 4.12 1.33 5.00 1.54 3.26 0.99
下载: 导出CSV
表 8 收费站用能负荷预测结果
Table 8. Forecasting results of toll stations energy load
预测模型 春季 夏季 秋季 冬季 4月3日 6月3日 8月3日 10月3日 2月3日 12月3日 M/% R M/% R M/% R M/% R M/% R M/% R LSTM 16.39 2.99 15.84 2.78 23.87 3.29 27.86 3.59 15.70 2.45 11.68 1.67 PCA-LSTM 15.87 2.56 15.64 2.62 23.72 3.27 26.36 3.38 15.05 2.33 10.56 1.52 CD-LSTM 13.51 1.26 4.48 0.68 3.09 0.48 4.77 0.52 10.60 1.69 2.61 0.33 CD-PCA-LSTM 12.33 1.17 3.60 0.56 2.86 0.44 4.65 0.50 10.44 1.64 2.11 0.27
下载: 导出CSV
表 9 服务区用能负荷预测结果
Table 9. Forecasting results of service area energy load
预测模型 春季 夏季 秋季 冬季 4月3日 6月3日 8月3日 10月3日 2月3日 12月3日 M/% R M/% R M/% R M/% R M/% R M/% R LSTM 26.67 13.13 29.73 12.98 21.39 11.45 34.57 14.36 19.37 11.17 21.21 11.06 PCA-LSTM 26.19 12.87 28.54 12.68 20.99 11.18 28.21 12.60 18.08 10.50 20.95 10.91 CD-LSTM 5.84 2.03 16.17 4.61 6.10 2.09 7.76 2.24 7.66 2.74 14.58 3.88 CD-PCA-LSTM 4.32 1.54 13.03 3.86 4.20 1.37 3.23 1.12 6.70 2.41 7.88 2.30
下载: 导出CSV
-
[1] 贾利民, 师瑞峰, 吉莉, 等. 我国道路交通与能源融合发展战略研究[J]. 中国工程科学, 2022, 24(3): 163-172.JIA Li-min, SHI Rui-feng, JI Li, et al. Road transportation and energy integration strategy in China[J]. Strategic Study of CAE, 2022, 24(3): 163-172. (in Chinese) [2] ZHAO Shi-yong, XIAN Rong, JIAO Li-chao, et al. Analysis of energy consumption level and influencing factors of highway service area[J]. IOP Conference Series: Earth and Environmental Science, 2021, 825(1): 012005. doi: 10.1088/1755-1315/825/1/012005 [3] QIU Feng, LI Wen-quan, XIE Qiu-feng, et al. Calculation and prediction of energy consumption for highway transportation[C]//IET. 7th Advanced Forum on Transportation of China (AFTC 2011). London: IET, 2011: 226-231. [4] CANSIZÖ F, ÜNEŞ F, ERGINER İ, et al. Modeling of highways energy consumption with artificial intelligence and regression methods[J]. International Journal of Environmental Science and Technology, 2022, 19(10): 9741-9756. doi: 10.1007/s13762-021-03813-1 [5] KOU G, YVKÇEL S, DINSER H. Inventive problem-solving map of innovative carbon emission strategies for solar energy-based transportation investment projects[J]. Applied Energy, 2022, 311: 118680. doi: 10.1016/j.apenergy.2022.118680 [6] ZHANG Tian-jian, ZHANG Jing-pei, LIU Yan-bo, et al. Design of linear regression scheme in real-time market load prediction for power market participants[C]//IEEE. 2021 11th International Conference on Power and Energy Systems (ICPES). New York: IEEE, 2021: 547-551. [7] JI Yu-qi, PANG Chen-yang, LIU Xiao-mei, et al. Combined forecasting model based on time series characteristics of power load curve[C]//IEEE. 2022 IEEE International Conference on Power Systems and Electrical Technology (PSET). New York: IEEE, 2022: 303-309. [8] SONG Feng, LIU Jun-xu, ZHANG Ting-ting, et al. The grey forecasting model for the medium-and long-term load forecasting[J]. IOP Conference Series: Materials Science and Engineering, 2020, 740(1): 012076. doi: 10.1088/1757-899X/740/1/012076 [9] MI Jian-wei, FAN Li-bin, DUAN Xue-chao, et al. Short-term power load forecasting method based on improved exponential smoothing grey model[J]. Mathematical Problems in Engineering, 2018, 2018: 3894723. [10] WANG Zheng-yu, HUO Yue-ying, LIU Zhen-yu. Prediction model on energy consumption of highway transportation in inner Mongolia based on ARMA[C]//WANG W, BENGLE K, JIANG X. International Conference on Green Intelligent Transportation System and Safety. Berlin: Springer, 2018: 105-114. [11] 李钷, 李敏, 刘涤尘. 基于改进回归法的电力负荷预测[J]. 电网技术, 2006, 30(1): 99-104.LI Po, LI Min, LIU Di-chen. Power load forecasting based on improved regression[J]. Power System Technology, 2006, 30(1): 99-104. (in Chinese) [12] DAI Li-xin, HU Fang-fang. Application optimization of grey model in power load forecasting[J]. Advanced Materials Research. 2012, 347-353: 301-305. [13] JIN Xin, DONG Yao, WU Jie, et al. An improved combined forecasting method for electric power load based on autoregressive integrated moving average model[C]//IEEE. 2010 International Conference of Information Science and Management Engineering. New York: IEEE, 2010: 476-480. [14] 赵晓华, 姚莹, 伍毅平, 等. 基于主成分分析与BP神经元网络的驾驶能耗组合预测模型研究[J]. 交通运输系统工程与信息, 2016, 16(5): 185-191, 204.ZHAO Xiao-hua, YAO Ying, WU Yi-ping, et al. Prediction model of driving energy consumption based on PCA and BP network[J]. Journal of Transportation Systems Engineering and Information Technology, 2016, 16(5): 185-191, 204. (in Chinese) [15] LIU Chang, ZHANG Yuan-liang, CHEN Wei-song, et al. A short term forecasting method for regional power consumption considering related factors[J]. Journal of Physics: Conference Series, 2022, 2195(1): 012022. doi: 10.1088/1742-6596/2195/1/012022 [16] CHEN Hua-xi. Prediction method of energy consumption in industrial production based on improved grey model[J]. International Journal of Global Energy Issues, 2023, 45(2): 101-112. doi: 10.1504/IJGEI.2023.129503 [17] 李晓娟, 张芳媛, 喻玲. 基于主成分分析-BP神经网络的风电备件需求预测[J]. 科学技术与工程, 2024, 24(1): 281-288.LI Xiao-juan, ZHANG Fang-yuan, YU Ling. Wind power spare parts demand forecasting based on PCA-BP neural network[J]. Science Technology and Engineering, 2024, 24(1): 281-288. (in Chinese) [18] BIAN Hai-hong, WANG Qian, TIAN Lin-lin. Research on short-term load forecasting based on PCA-GM[C]//Springer. International Conference on Multimedia Technology and Enhanced Learning. Berlin: Springer, 2020: 169-177. [19] 石海波. PCA-SVM在电力负荷预测中的应用研究[J]. 计算机仿真, 2010, 27(10): 279-282.SHI Hai-bo. Power load forecasting based on principal component analysis and support vector machine[J]. Computer Simulation, 2010, 27(10): 279-282. (in Chinese) [20] 党存禄, 杨海兰, 武文成. 基于LSTM和CatBoost组合模型的短期负荷预测[J]. 电气工程学报, 2021, 16(3): 62-69.DANG Cun-lu, YANG Hai-lan, WU Wen-cheng. Short-term load forecasting based on LSTM and CatBoost combined model[J]. Journal of Electrical Engineering, 2021, 16(3): 62-69. (in Chinese) [21] YU Yong, SI Xiao-sheng, HU Chang-hua, et al. A review of recurrent neural networks: LSTM cells and network architectures[J]. Neural Computation, 2019, 31(7): 1235-1270. doi: 10.1162/neco_a_01199 [22] BEZERRA F E, GRASSI F, DIAS C G, et al. A PCA-based variable ranking and selection approach for electric energy load forecasting[J]. International Journal of Energy Sector Management, 2022, 16(6): 1172-1191. [23] 商立群, 李洪波, 侯亚东, 等. 基于特征选择和优化极限学习机的短期电力负荷预测[J]. 西安交通大学学报, 2022, 56(4): 165-175.SHANG Li-qun, LI Hong-bo, HOU Ya-dong, et al. Short-term power load forecasting based on feature selection and optimized extreme learning machine[J]. Journal of Xi'an Jiaotong University, 2022, 56(4): 165-175. (in Chinese) [24] LUO Yi, WU Yong, YE Yan-feng, et al. Short-term load forecasting based on PCA-ILSTM[C]//IEEE. 2023 IEEE 2nd International Conference on Electrical Engineering, Big Data and Algorithms (EEBDA). New York: IEEE, 2023: 1075-1079. [25] CUI Can, HE Ming, DI Fang-chun, et al. Research on power load forecasting method based on LSTM model[C]//IEEE. 2020 IEEE 5th Information Technology and Mechatronics Engineering Conference (ITOEC). New York: IEEE, 2020: 1657-1660. [26] HUANG Song-tao, SHEN Jun, LYU Qing-quan, et al. A novel NODE approach combined with LSTM for short-term electricity load forecasting[J]. Future Internet, 2023, 15(1): 15010022. [27] 陆继翔, 张琪培, 杨志宏, 等. 基于CNN-LSTM混合神经网络模型的短期负荷预测方法[J]. 电力系统自动化, 2019, 43(8): 131-137.LU Ji-xiang, ZHANG Qi-pei, YANG Zhi-hong, et al. Short-term load forecasting method based on CNN-LSTM hybrid neural network model[J]. Automation of Electric Power Systems, 2019, 43(8): 131-137. (in Chinese) [28] ZHANG Qing-hu, MAO Hai-bo, SHAN Ze-long, et al. An artificial intelligent load forecast method using image based LSTM[C]//IEEE. 2022 China Automation Congress (CAC). New York: IEEE, 2022: 5318-5321. [29] LI Shan, LU Xin, OUYANG Jian-na, et al. K-means clustering algorithm and LSTM based short-term load forecasting for distribution transformer[C]//IEEE. 2023 5th Asia Energy and Electrical Engineering Symposium. New York: IEEE, 2023: 1152-1156. [30] CHEN Kun-jin, CHEN Kun-long, WANG Qin, et al. Short-term load forecasting with deep residual networks[J]. IEEE Transactions on Smart Grid, 2018, 10(4): 3943-3952. [31] 徐先峰, 赵依, 刘状壮, 等. 用于短期电力负荷预测的日负荷特性分类及特征集重构策略[J]. 电网技术, 2022, 46(4): 1548-1556.XU Xian-feng, ZHAO Yi, LIU Zhuang-zhuang, et al. Daily load characteristic classification and feature set reconstruction strategy for short-term power load forecasting[J]. Power System Technology, 2022, 46(4): 1548-1556. (in Chinese) -
点击查看大图
图(9) / 表(9)
计量
- 文章访问数: 527
- HTML全文浏览量: 213
- PDF下载量: 7
- 被引次数: 0
