基于隶属度的模糊时间序列取水量预测研究
PDF下载 (110)毛小燕,戴俊杰,曹 磊.基于隶属度的模糊时间序列取水量预测研究[J].宁波大学学报(理工版),2025,38(6):82-90.DOI:10.20098/j.cnki.1001-5132.2024.1212
MAO Xiaoyan,DAI Junjie,CAO Lei.Research on fuzzy time series water consumption prediction method based on membership degree[J].Journal of Ningbo University(Natural Science & Engineering Edition),2025,38(6):82-90.DOI:10.20098/j.cnki.1001-5132.2024.1212
| Title: | Research on fuzzy time series water consumption prediction method based on membership degree |
| 作者: | 毛小燕, 戴俊杰, 曹 磊 |
| Author(s): | MAO Xiaoyan, DAI Junjie, CAO Lei |
| 关键词: | 取水量; 隶属度; 模糊时间序列; 模糊C均值聚类; 循环神经网络; 门控循环单元 |
| Keywords: | water consumption; membership; fuzzy time series; fuzzy C-means clustering; recurrent neural network; gated recurrent unit |
| 分类号: | TP301 |
| DOI: | 10.20098/j.cnki.1001-5132.2024.1212 |
| 文献标识码: | A |
| 摘要: | 科学准确地预测水厂取水量是实现水资源高效管理与科学决策的关键环节，更是推进智慧水利建设的重要前提。由于取水量序列具有非线性、模糊性等复杂特征，且受多种不确定因素影响，使传统预测模型的精度受到一定限制。本文结合模糊理论和门控循环单元（GRU）神经网络的特性，提出一种基于隶属度的模糊时间序列取水量预测新方法。首先，利用模糊C均值聚类（FCM）算法从原始序列中构建子隶属度序列；其次，采用两种不同策略，利用GRU模型对各子隶属度序列进行预测；最后，通过去模糊化处理得出取水量预测值。将新模型应用于浙江省某水厂的取水量序列预测，并与传统GRU模型和AM-GRU模型进行对比分析，结果显示新模型在均方误差（MSE）、均方根误差（RMSE）和平均绝对百分比误差（MAPE）上的表现均优于其他模型，显著提高了取水量的预测精度。 |
| Abstract: | The scientific and accurate prediction of water consumption from water plants is crucial for the efficient management and scientific decision-making of water resources, and is a prerequisite for promoting smart water conservancy construction. Due to the nonlinearity, fuzziness, and other complex characteristics in the water consumption sequence, and those affected by a variety of uncertainties, the accuracy of traditional prediction models is limited to some extent. This paper combines the characteristics of fuzzy theory and gated recurrent unit neural network to propose a new method for predicting water consumption in fuzzy time series based on membership degree. First, the fuzzy C-means clustering (FCM) is used to construct sub membership sequences from the original sequence; second, the GRU model is used with two strategies to predict each sub membership degree sequence; finally, the predicted value of water consumption is obtained by defuzzification. The new model proposed in this paper is employed to predict the water consumption sequence of a water plant in Zhejiang Province, with its results being compared with those from the traditional GRU model and AM-GRU model for analysis. The experimental results show that the new model outperforms other models in terms of MSE, RMSE, and MAPE, and it improves prediction accuracy. |
| 参考文献 /References: | [1] 尹学康. 城市需水量预测系统的开发研究[D]. 长沙: 湖南大学, 2003. [2] 练庭宏, 刘秋娟, 王景成. 基于ARIMA时序辨识的需水量预测[J]. 控制工程, 2008, 15(增刊1):162-164. [3] 景亚平, 张鑫, 罗艳. 基于灰色神经网络与马尔科夫链的城市需水量组合预测[J]. 西北农林科技大学学报(自然科学版), 2011, 39(7):229-234. [4] 苟非洲, 程玉婷. 基于长短期记忆网络的日供水量预测方法研究[J]. 中国给水排水, 2019, 35(17):79-83. [5] 陆维佳, 朱建文, 叶圣炯, 等. 基于多因素长短时神经网络的日用水量预测方法研究[J]. 给水排水, 2020, 56 (1):125-129. [6] PANIGRAHI S, BEHERA H S. A study on leading machine learning techniques for high order fuzzy time series forecasting[J]. Engineering applications of artificial intelligence, 2020, 87:103245. [7] 孙平, 王丽萍, 陈凯, 等. 基于时间序列模型ARMA的水厂逐日需水量过程预测方法[J]. 中国农村水利水电, 2013(11):139-142. [8] SONG Q, CHISSOM B S. Forecasting enrollments with fuzzy time series-Part I[J]. Fuzzy sets and systems, 1993, 54(1):1-9. [9] SONG Q, CHISSOM B S. Forecasting enrollments with fuzzy time series-Part II[J]. Fuzzy sets and systems, 1994, 62(1):1-8. [10] ÖZNUR Ö K, KAYMAK Y. Prediction of crude oil prices in COVID-19 outbreak using real data[J]. Chaos, solitons & fractals, 2022, 158:111990. [11] 蔺玉佩, 杨一文. 基于模糊时间序列模型的股票市场预测[J]. 统计与决策, 2010, 26(8):34-37. [12] SEVERIANO C A, CANDIDO D L E S P, COHEN M W, et al. Evolving fuzzy time series for spatio-temporal forecasting in renewable energy systems[J]. Renewable energy, 2021, 171:764-783. [13] SADAEI H J, CANDIDO D L E S P, GUIMARAES F G, et al. Short-term load forecasting by using a combined method of convolutional neural networks and fuzzy time series[J]. Energy, 2019, 175(5):365-377. [14] KUMAR N, SUSAN S. Particle swarm optimization of partitions and fuzzy order for fuzzy time series forecasting of COVID-19[J]. Applied soft computing, 2021, 110:107611. [15] 鲜思东, 李堂金. 基于改进狼群算法的模糊时间序列预测模型[J]. 控制理论与应用, 2020, 37(7):1637-1643. [16] 董文超, 郭强, 张彩明. 一种模糊时间序列概率预测方法[J]. 计算机工程与科学, 2024, 46(8):1493-1502. [17] 王鹏, 田宗浩. 基于直觉模糊化的广义模糊时间序列预测模型[J]. 运筹与管理, 2020, 29(3):128-134. [18] 杨烨, 卢建刚. 基于融合Transformer的门尼系数预测建模研究[J]. 化工学报, 2025, 76(1):266-282. [19] 戴邵武, 陈强强, 刘志豪, 等. 基于EMD-LSTM的时间序列预测方法[J]. 深圳大学学报(理工版), 2020, 37 (3):265-270. [20] 管业鹏, 苏光耀, 盛怡. 双向长短期记忆网络的时间序列预测方法[J]. 西安电子科技大学学报, 2024, 51(3):103-112. [21] 万红, 钱锐. 模糊C-均值聚类引导的Kinect深度图像修复算法[J]. 计算机应用研究, 2019, 36(5):1564-1568. [22] HOUIMLI R, ZMAMI M, BEN-SALHA O. Short-term electric load forecasting in Tunisia using artificial neural networks[J]. Energy systems, 2020, 11(2):357-375. [23] KONG W C, DONG Z Y, JIA Y W, et al. Short-term residential load forecasting based on LSTM recurrent neural network[J]. IEEE transactions on smart grid, 2017, 10(1):841-851. [24] 李鹏, 何帅, 韩鹏飞, 等. 基于长短期记忆的实时电价条件下智能电网短期负荷预测[J]. 电网技术, 2018, 42 (12):4045-4052. [25] 赵家庆, 赵裕啸, 丁宏恩, 等. 电网调度自动化主备系统间模型正确性校验技术方案[J]. 电力系统保护与控制, 2014, 42(19):139-144. [26] KABIR G, TESFAMARIAM S, HEMSING J, et al. Handling incomplete and missing data in water network database using imputation methods[J]. Sustainable and resilient infrastructure, 2020, 5(6):365-377. [27] VELASCO-GALLEGO C, LAZAKIS I. Real-time data-driven missing data imputation for short-term sensor data of marine systems: a comparative study[J]. Ocean engineering, 2020, 218:108261. [28] LIM B, ZOHREN S. Time-series forecasting with deep learning: a survey[J]. Philosophical transactions of the royal society A, 2021, 379(2194):20200209. [29] 王志良, 黄珊, 陈海涛. 黄河流域水文数据插补方法比较及应用[J]. 人民黄河, 2020, 47(7):14-18. [30] 沐年国, 姚洪刚. 基于注意力机制的循环神经网络对金融时间序列的应用[J]. 现代电子技术, 2021, 44(14):1-5. [31] 张扬. 基于改进深度神经网络的短期电力负荷预测[J]. 科技创新与应用, 2022, 12(25):12-15. [32] 鲜思东, 李堂金. 基于改进狼群算法的模糊时间序列预测模型[J]. 控制理论与应用, 2020, 37(7):1637-1643. [33] 高凯悦, 牟莉. 基于二次分解和GRU-attention的时间序列预测研究[J]. 国外电子测量技术, 2023, 42(2):80-87. [34] 丁欣, 谢祥俊, 赵春兰, 等. 基于动态隶属度的模糊时间序列在我国居民消费水平预测上的应用[J]. 模糊系统与数学, 2019, 33(1):164-174. [35] 林俊亭, 王帅, 刘恩东, 等. 基于模糊聚类和CNN- BIGRU的轨道电路故障预测方法[J]. 振动、测试与诊断, 2023, 43(3):500-507. |
| 备注/Memo: | 收稿日期：2024−12−23 宁波大学学报（理工版）网址：http://journallg.nbu.edu.cn/ 基金项目：浙江省自然科学基金（LY20A010012） 第一作者：毛小燕，副教授，主要研究方向为模糊数学与Rough集理论。E-mail: maoxiaoyan@nbu.edu.cn 宁波大学学报（理工版）网址：http://journallg.nbu.edu.cn/ |
