[1]胡晶晶,罗永明,张纲强,等.基于邻域粗糙集的气象因子选择在虾塘水温预测中的应用[J].江苏农业学报,2023,(03):732-740.[doi:doi:10.3969/j.issn.1000-4440.2023.03.014]
　HU Jing-jing,LUO Yong-ming,ZHANG Gang-qiang,et al.Application of meteorological factor selection based on neighborhood rough set in shrimp pond water temperature prediction[J].,2023,(03):732-740.[doi:doi:10.3969/j.issn.1000-4440.2023.03.014]
点击复制

基于邻域粗糙集的气象因子选择在虾塘水温预测中的应用()
分享到：

江苏农业学报[ISSN:1006-6977/CN:61-1281/TN]

卷:
期数:
2023年03期
页码:
732-740
栏目:
农业信息工程
出版日期:
2023-06-30

文章信息/Info

Title:
Application of meteorological factor selection based on neighborhood rough set in shrimp pond water temperature prediction
作者:
胡晶晶1; 罗永明2; 张纲强3; 匡昭敏2; 谢映2; 曾行吉4
(1.广西民族大学电子信息学院，广西南宁530006；2.广西壮族自治区气象科学研究所，广西南宁530022；3.广西民族大学人工智能学院，广西南宁530006；4.广西壮族自治区气象信息中心，广西南宁530022)
Author(s):
HU Jing-jing1; LUO Yong-ming2; ZHANG Gang-qiang3; KUANG Zhao-min2; XIE Ying2; ZENG Xing-ji4
(1.College of Electronic Information, Guangxi Minzu University, Nanning 530006, China；2.Guangxi Zhuang Autonomous Region Institute of Meteorological Sciences, Nanning 530022, China；3.School of Artificial Intelligence, Guangxi Minzu University, Nanning 530006, China；4.Guangxi Zhuang Autonomous Region Meteorological Information Center, Nanning 530022, China)
关键词:
虾塘水温; 气象因子; 邻域粗糙集; SFNN模型
Keywords:
shrimp pond water temperature; meteorological factors; neighborhood rough set; single hidden layer feed forward neural network (SFNN) model
分类号:
S966.12
DOI:
doi:10.3969/j.issn.1000-4440.2023.03.014
文献标志码:
A
摘要:
基于邻域粗糙集对影响虾塘水温变化的气象因子进行选择，并选取模型预测虾塘水温，为南美白对虾养殖趋利避害提供科学参考。首先，将平均气温、最高气温、最低气温、降水量、气压、2 min风速、10 min风速和瞬时风速等8个气象因子组合输入SFNN模型（单隐层前馈神经网络模型）、高斯回归模型和岭回归模型进行虾塘水温预测，选取预测效果最好的SFNN模型为本研究预测模型。然后，运用邻域粗糙集和熵理论，考虑气象因子和虾塘水温之间的相关性、冗余性和交互性，选出影响虾塘水温变化的主要气象因子。最后，利用选出的主要气象因子和SFNN模型实现虾塘水温预测。将基于邻域粗糙集选出的5个气象因子组合与8个气象因子组合，以及8个气象单因子分别输入SFNN模型，预测结果表明：邻域粗糙集选出的5个气象因子组合预测结果最好，其预测均方根误差、均方误差、平均绝对误差最小，分别为1.121 1、1.256 9和0.893 8，决定系数（R2）为0.791 6；在气象单因子中，气压对虾塘水温的预测结果较好。因此，基于邻域粗糙集选出的5个气象因子组合，通过SFNN模型进行虾塘水温预测结果最好，此方法在南美白对虾养殖趋利避害、防灾减灾中具有一定的实用价值。
Abstract:
Based on the neighborhood rough set, the meteorological factors affecting the change of shrimp pond water temperature were selected. A model was selected to predict the water temperature of shrimp pond, which provided a scientific reference for shrimp culture to seek advantages and avoid disadvantages. Firstly, eight meteorological factors, including average temperature, highest temperature, lowest temperature, amount of precipitation, barometric pressure, two-minute wind speed, ten-minute wind speed, and instantaneous wind speed, were combined into the single hidden layer feed forward neural network (SFNN) model, Gaussian regression model, and bridge regression model to predict the water temperature of shrimp ponds. SFNN model with the best prediction effect was selected as the prediction model. Then considering the correlation, redundancy and interaction between meteorological factors and shrimp pond water temperature, the main meteorological factors affecting the shrimp pond water temperature were selected by using neighborhood rough set and entropy theory. Finally, the water temperature of the shrimp pond was predicted by using the selected meteorological factors and SFNN model. Combination of five meteorological factors selected based on neighborhood rough set, the combination of eight meteorological factors, and eight single meteorological factors were input into the SFNN model, respectively. The prediction results showed that, the combination of five meteorological factors selected by neighborhood rough set had the best prediction results, and its root mean square error, mean square error and mean absolute error were the smallest, which were 1.121 1, 1.256 9 and 0.893 8, respectively. And the determination coefficient was 0.791 6. Among single meteorological factors, atmospheric pressure predicted best on shrimp pond water temperature. Therefore, the combination of five meteorological factors selected based on the neighborhood rough set had the best prediction in shrimp pond water temperature by the SFNN model. This method had practical value in seeking advantages and avoiding disadvantages, and disaster prevention and reduction in prawn breeding.

参考文献/References:

[1]陈美佳,孔杰,谭建,等. 凡纳滨对虾育种基础群体收获体重的未知亲本组效应分析[J]. 中国水产科学, 2021, 28(7): 863-870.
[2]张志东,陈爱华,吴杨平,等. 不同饲料投喂模式对脊尾白虾生长、消化酶、体成分及养殖水环境的影响[J]. 中国水产科学, 2020, 27(9): 1075-1084.
[3]农成万,张幸,李斌喜,等. 钦州对虾养殖灾害性天气分析和气象服务要领[J]. 气象研究与应用, 2013, 34(1): 60-63.
[4]王立超,申子彬,周溥佳. 南美白对虾养殖气象指数预报方法[J]. 浙江农业科学, 2017, 58(7): 1254-1256.
[5]杨栋,姚日升,金志凤,等. 不同类型海水养殖水体温度日变化谐波分析[J]. 中国农业气象, 2017, 38(9): 558-566.
[6]吴迪生,张娟,刘增宏,等. 赤道西太平洋暖池次表层水温与热带气旋的关系[J]. 热带气象学报, 2010, 26(2): 242-249.
[7]叶舟,龚梦园. 基于下游生境水温需求的水库分层取水调度研究[J]. 人民长江, 2022,53(6): 105-110.
[8]刘杏红,韩云童,常晓林,等. 考虑淤沙影响的高拱坝水库水温数值分析[J]. 水利学报, 2010, 31(2): 6-8.
[9]邓爱娟,刘敏,刘志雄,等. 洪湖地区养殖鱼塘春夏季水温变化及预报研究[J]. 中国农学通报, 2013, 29(29): 61-68.
[10]韦耀国,赵海镜,杨国华,等. 神经网络在南水北调中线冬季水温预测中的应用[J]. 水利水电技术(中英文), 2021, 52(S2): 335-338.
[11]施珮,袁永明,匡亮,等. 基于EMD-IGA-SELM的池塘养殖水温预测方法[J]. 农业机械学报, 2018, 49(11): 312-319.
[12]陈英义,程倩倩,成艳君,等. 基于GA-BP神经网络的池塘养殖水温短期预测系统[J]. 农业机械学报, 2017, 48(8): 172-178.
[13]柳海涛,孙双科,郑铁刚,等. 水电站下游鱼类产卵场水温的人工神经网络预报模型[J].农业工程学报, 2018, 34(4): 185-191.
[14]朱森林,吴时强. 高斯过程回归模型在河流水温模拟中的应用[J]. 华中科技大学学报(自然科学版), 2018, 46(10): 122-126.
[15]程海燕,脱友才,许维忠,等. 气候变暖对稳定分层型供水水库水温影响研究[J]. 人民长江, 2022,53(10)：22-30.
[16]杨开林. 冰期明渠水温模型[J]. 水利学报, 2022, 53(1): 20-30.
[17]邹琳,刘伟,蒋定国. 基于数值方法的大型水库水温结构影响因素分析[J]. 中国农村水利水电, 2022(9): 113-119.
[18]杨青青,曾月,邓艳君, 等. 稻虾共作模式虾沟水温预报模型研究[J]. 江苏农业科学, 2021, 49(5): 194-199.
[19]PAWLAK Z. Vagueness and uncertainty: A rough set perspective[J]. Computational Intelligence, 1995, 11(2): 227-232.
[20]HU Q H, YU D, LIU J F, et al. Neighborhood rough set based heterogeneous feature subset selection[J]. Information Sciences, 2008, 178(18): 3577-3594.
[21]SUN B Z, MA W M, XIAO X. Three-way group decision making based on multigranulation fuzzy decision-theoretic rough set over two universes[J]. International Journal of Approximate Reasoning, 2017, 81: 87-102.
[22]DONG L J, CHEN D G, WANG N L, et al. Key energy-consumption feature selection of thermal power systems based on robust attribute reduction with rough sets[J]. Information Sciences, 2020, 532: 61-71.
[23]ZHANG P F, LI T R, WANG G Q, et al. Multi-source information fusion based on rough set theory: A review[J]. Information Fusion, 2021, 68: 85-117.
[24]ZHANG X Y, YAO H, LU Z Y, et al. Class-specific information measures and attribute reducts for hierarchy and systematicness[J]. Information Sciences, 2021, 563: 196-225.
[25]CHEN H M, LI T R, FAN X, et al. Feature selection for imbalanced data based on neighborhood rough sets[J]. Information Sciences, 2019, 483: 1-20.
[26]DAI J H, WANG W T, XU Q. An uncertainty measure for incomplete decision tables and its applications[J]. IEEE Transactions on Cybernetics, 2012, 43(4): 1277-1289.
[27]WAN J H, CHEN H M, ZHONG Y, et al. A novel hybrid feature selection method considering feature interaction in neighborhood rough set[J]. Knowledge-Based Systems, 2021, 227: 107167.
[28]WANG C Z, HUANG Y, SHAO M W, et al. Feature selection based on neighborhood self-information[J]. IEEE Transactions on Cybernetics, 2019, 50(9): 4031-4042.
[29]SUGANTHAN P N, KATUWAL R. On the origins of randomization-based feedforward neural networks[J]. Applied Soft Computing, 2021, 105: 107239.
[30]李亚茹,张宇来,王佳晨. 面向超参数估计的贝叶斯优化方法综述[J].计算机科学,2022,49(S1):86-92.
[31]张慧春,张萌,边黎明,等. 基于YOLO v5的植物叶绿素含量估测与可视化技术[J]. 农业机械学报, 2022, 53(4): 313-321.
[32]YUAN Z, CHEN H M, ZHANG P F, et al. A novel unsupervised approach to heterogeneous feature selection based on fuzzy mutual information[J]. IEEE Transactions on Fuzzy Systems, 2021,39(9) 3395-3409.
[33]施珮,袁永明,匡亮,等. 基于GA-SELM算法的工厂化水产养殖水温预测方法研究[J]. 传感技术学报, 2018, 31(10): 1592-1597，1612.
[34]任妮,鲍彤,刘杨,等. 基于粒子群优化算法和长短时记忆神经网络的蟹塘溶解氧预测[J]. 江苏农业学报, 2021, 37(2): 426-434.
[35]李祥铜,曹亮,李湘丽,等. 基于WTD-LSTM的对虾养殖水温组合预测模型[J]. 广东农业科学, 2021, 48(2): 153-160.
[36]袁红春,王敏,刘慧,等. 基于特征交互与卷积网络的渔场预测模型[J]. 江苏农业学报, 2021, 37(6): 1501-1509.

相似文献/References:

[1]张垚,葛均筑,周广生,等.气象因子对饲用油菜越冬性及产量和品质的影响[J].江苏农业学报,2020,(06):1419.[doi:doi:10.3969/j.issn.1000-4440.2020.06.010]
　ZHANG Yao,GE Jun-zhu,ZHOU Guang-sheng,et al.Effects of meteorological factors on overwintering ability, yield and quality of forage rape[J].,2020,(03):1419.[doi:doi:10.3969/j.issn.1000-4440.2020.06.010]

备注/Memo

备注/Memo:
收稿日期：2022-07-21 基金项目：广西壮族自治区自然科学基金项目(2020GXNSFAA238046)；广西壮族自治区气象局气象科研计划重点项目(桂气科2020Z03)；广西民族大学软件工程重点实验室项目(2022-18XJSY-03) 作者简介：胡晶晶(1996-)，女，江苏淮安人，硕士研究生，主要从事粗糙集理论、数据挖掘与知识发现研究。(E-mail)jingjinghu_star304@163.com 通讯作者：罗永明，(E-mail) mingyongluo858@163.com
更新日期/Last Update: 2023-07-11