Application of Machine Learning Methods on Predicting Irrigation Water Quality

應用機器學習方法於灌溉水質預測

10.6937/TWC.202003/PP_68(1).0001

林裕彬(YU-PIN LIN)；連宛渝(WAN-YU LIEN)；陳欣鈺(HSIN-YU CHEN)；何俊宏(JYUN-HONG HE)；周承復(CHENG-FU CHOU)

SVM ； RF ； Irrigation water pollution ； Prediction ； Water quality ； 支援向量機 ； 隨機森林 ； 灌溉水污染 ； 水質 ； 預測

台灣水利

68卷1期（2020 / 03 / 01）

1 - 14

英文

The pollution of irrigation water leads to the pollution of farmlands directly or indirectly, which will further cast impacts on crop quality. Therefore, accurate predictions of future pollution events are essential for management of irrigation water. The aim of our study is to predict the potential occurrence of future abrupt pollution events by historical and real time monitoring water quality data. The 12 basic water quality monitoring stations and 2 heavy metal monitoring stations are selected in this study. We then use SVM and RF methods to predict whether the water quality might exceed normal standard in the near future. Our result shows that both of the methods received high credibility in predicting the standard-exceeding conditions of irrigation water. In addition, our study takes water level as well as precipitation factors into the models for a better precision in predicting of major standard-exceeding concentration of heavy metal, copper, in the irrigation water of study area. The result indicates that the prediction ability increased after water level factor was added, but not in the case of precipitation factor. Additionally, by making water quality data resemble the actual conditions, data segmentation should be conducted based on time series while analyzing the data instead of random selection. The accuracy of SVM model can be increased to 99.7% and 85.18% in the validation and test data set. By predicting potential occurring time of pollution events via historical as well as water monitoring data, it is possible to take necessary preventions to lower the risks of crops being polluted, which is a major issue in agricultural production nowadays.

灌溉水質受到污染將直接或間接導致農地遭受污染，進而使得作物品質受到影響，然而，灌溉水質污染事件往往無法預測可能的發生時間，無法及時採取可能的措施降低對農地及農產品可能的危害。為能藉由歷史資料及即時預測未來是否可能發生污染事件導致灌溉水質不符合水質標準，本研究選用桃園地區之12個一般水質測站及2做重金屬測站之歷史及即時監測資料，利用SVM及RF模式，預測未來是水質是否可能超過標準，研究結果顯示，兩種方法預測灌溉水質超標的可信度均高；此外主要超標之重金屬濃度為銅，因此於模式中納入水位及降雨量因素，討論是否可更加精確預測銅離子濃度是否可能超標的情形，研究結果顯示，加入水位因子後預測能力有提升的情形，但加入雨量則無。而由於水質資料具有連續性，因此在分析時應依照時間序列分割資料，而非以隨機選取資料的方式進行模擬，研究結果顯示，當以時間序列分割資料進行模擬時，SVM模式在驗證及測試資料集之精確度可提升至99.7%及85.18%，顯示此種選取資料之方法可獲得符合實際情形之預測結果。由於灌溉水質直接影響作物品質，使得作物受到污染的機率增加，進而危及糧食安全，因此藉由歷史與即時水質監測資料，預測可能的污染發生時間，採取可能的預防措施，降低作物可能遭受污染的風險，實為一重要課題。

1. Department of Irrigation and Engineering, Council of Agriculture, Executive Yuan. Retrieved from: https://doie.coa.gov.tw/about.php(2019/8/6), 2019.
2. Alizadeh, M. J.,Kavianpour, M. R.(2015).Development of wavelet-ANN models to predict water quality parameters in Hilo Bay, Pacific Ocean.Marine pollution bulletin,98(1-2),171-178.
3. Arnon, T. A.,Ezra, S.,Fishbain, B.(2019).Water characterization and early contamination detection in highly varying stochastic background water, based on Machine Learning methodology for processing realtime UV-Spectrophotometry.Water research,155,333-342.
4. Aryafar, A.,Gholami, R.,Rooki, R.,Ardejani, F. D.(2012).Heavy metal pollution assessment using support vector machine in the Shur River, Sarcheshmeh copper mine, Iran.Environmental earth sciences,67(4),1191-1199.
5. Avila, R.,Horn, B.,Moriarty, E.,Hodson, R.,Moltchanova, E.(2018).Evaluating statistical model performance in water quality prediction.Journal of environmental management,206,910-919.
6. Barzegar, R.,Moghaddam, A. A.,Adamowski, J.,Ozga-Zielinski, B.(2018).Multi-step water quality forecasting using a boosting ensemble multiwavelet extreme learning machine model.Stochastic environmental research and risk assessment,32(3),799-813.
7. Bhuyan, M. S.,Bakar, M. A.(2017).Seasonal variation of heavy metals in water and sediments in the Halda River, Chittagong, Bangladesh.Environmental Science and Pollution Research,24(35),27587-27600.
8. Breiman, L.(2001).Random forests.Machine learning,45,5-32.
9. Chang, P. W.,Hung, C. J.,Wang, I. S.,Tan, Y. C.(2003).Remediation of Cd-contaminated Farming and Fertility Enhancement.Journal of Taiwan Agricultural Engineering,49,23-34.
10. Chen, T. T.,Lee, S. J.(2015).A weighted LS-SVM based learning system for time series forecasting.Information Sciences,299,99-116.
11. Chen, Y. H.(2018).Department of Occupational Safety and Health. Chung Hwa University of Medical Technology.
12. Chung, M. X.(2003).Feng Chia University.
13. Da, Y.,Wang, X.,Zhao, Y.,Jiang, M.,Ye, M.(2015).Water quality prediction model based on relevance vector machine regression.Acta Scientiae Circumstantiae,35,3730-3735.
14. Domingos, P.M.(2012).A few useful things to know about machine learning.Commun. acm,55,78-87.
15. Dongwen, C.(2013).Application of extreme learning machine to total phosphorus and total nitrogen forecast in lakes and reservoirs.Water Resour Protect,29,61-66.
16. Environmental Protection Administration, Executive Yuan(2014).Environmental Protection Administration, Executive Yuan. 2014a. Control and Investigation Program for National Agricultural Land with High Potential of Heavy Metal Pollution (III)..
17. Environmental Protection Administration, Executive Yuan(2010).Environmental Protection Administration, Executive Yuan. 2010. Control and Investigation Program for National Agricultural Land with High Potential of Heavy Metal..
18. Environmental Protection Administration, Executive Yuan(2016).Environmental Protection Administration, Executive Yuan. 2016b. Control and Investigation Program for National Agricultural Land with High Potential of Heavy Metal Pollution (V)..
19. Environmental Protection Administration, Executive Yuan(2016).Environmental Protection Administration, Executive Yuan. 2016a. Control and Investigation Program for National Agricultural Land with High Potential of Heavy Metal Pollution (IV)..
20. Environmental Protection Administration, Executive Yuan(2014).Environmental Protection Administration, Executive Yuan. 2014b. Investigation of Contaminated Farmland Adjacent to Changhua's East-West Second and Third Irrigation System Canals..
21. Environmental Protection Administration, Executive Yuan(2012).Environmental Protection Administration, Executive Yuan. 2012. Control and Investigation Program for National Agricultural Land with High Potential of Heavy Metal Pollution(II)..
22. Ghamisi, P.,Maggiori, E.,Li, S.,Souza, R.,Tarablaka, Y.,Moser, G.,Serpico, S. B.(2018).New frontiers in spectral-spatial hyperspectral image classification: The latest advances based on mathematical morphology, Markov random fields, segmentation, sparse representation, and deep learning.IEEE Geoscience and Remote Sensing Magazine,6(3),10-43.
23. Goutte, C.,Gaussier, E.(2005).A probabilistic interpretation of precision, recall and F-score, with implication for evaluation.European Conference on Information Retrieval,Berlin, Heidelberg:
24. Guo, Q. H.,Kelly, M.,Graham, C. H.(2005).Support vector machines for predicting distribution of Sudden Oak Death in California.Ecological modelling,182(1),75-90.
25. Guo, Y.,Wang, G.,Zhang, X.,Deng, W.(2014).An improved hybrid ARIMA and support vector machine model for water quality prediction.International Conference on Rough Sets and Knowledge Technology
26. Han, H. G.,Chen, Q. L.,Qiao, J. F.(2011).An efficient self-organizing RBF neural network for water quality prediction.Neural Networks,24,717-725.
27. Hawkins, C. P.,Olson, J. R.,Hill, R. A.(2010).The reference condition: predicting benchmarks for ecological and water-quality assessments.Journal of the North American Benthological Society,29(1),312-343.
28. Hollister, J. W.,Milstead, W. B.,Kreakie, B. J.(2016).Modeling lake trophic state: a random forest approach.Ecosphere,7(3),e01321.
29. Huang, B.,Xie, C. L.,Tay, R.(2010).Support vector machines for urban growth modeling.Geoinformatica,14,83-99.
30. Hung, M. S.(2013).Department of Bioenvironmental Systems Engineering, National Taiwan University.
31. Krenkel, P.(2012).Water quality management.Elsevier.
32. Kuo, C. W.(2014).Department of Hydraulic and Ocean Engineering, National Cheng Kung University.
33. Liaw, A.,Wiener, M.(2002).Classification and regression by randomForest.R news,2,18-22.
34. Lin, K. H.(2016).Department of Hydraulic and Ocean Engineering, National Cheng Kung University.
35. Lin, S. H.(2008).Department of Civil Engineering, National Taiwan University.
36. Lin, Y. P.,Chang, T. K.,Fan, C. H.,Anthony, J.,Petway, J.,Lien, W. Y.,Liang, C. P.,Ho, Y. F.(2017).Applications of information and communication technology for improvements of water and soil monitoring and assessments in agricultural areas—A case study in the taoyuan irrigation district.Environments,4,6.
37. Lin, Y. P.,Lien, W. Y.,Yu, C. K.,Wu, H. H.,Li, Y. H.(2017).Irrigation Water Quality Monitoring and Management System.Taiwan water Conservancy,65,10-18.
38. Liu, M.,Lu, J.(2014).Support vector machine―an alternative to artificial neuron network for water quality forecasting in an agricultural nonpoint source polluted river?.Environ Sci Pollut R,21,11036-11053.
39. Liu, S. Y.,Tai, H. J.,Ding, Q. S.,Li, D. L.,Xu, L. Q.,Wei, Y. G.(2013).A hybrid approach of support vector regression with genetic algorithm optimization for aquaculture water quality prediction.Mathematical and Computer Modelling,58(3-4),458-465.
40. Liu, X. M.(2015).Department of Bioenvironmental Systems Engineering, National Taiwan University.
41. Mahapatra, S. S.,Nanda, S. K.,Panigrahy, B. K.(2011).A Cascaded Fuzzy Inference System for Indian river water quality prediction.Advances in Engineering Software,42(10),787-796.
42. Najah, A.,El-Shafie, A.,Karim, O.A.,Jaafar, O.,ElShafie, A.H.(2011).An application of different artificial intelligences techniques for water quality prediction.International Journal of Physical Sciences,6,5298-5308.
43. Rohmer, J.,Brisset, N.(2017).Short-term forecasting of saltwater occurrence at La Comté River (French Guiana) using a kernel-based support vector machine.Environmental Earth Sciences,76(6),246.
44. Salehi, E.,Abdi, J.,Aliei, M. H.(2016).Assessment of Cu (II) adsorption from water on modified membrane adsorbents using LS-SVM intelligent approach.Journal of Saudi Chemical Society,20(2),213-219.
45. Singh, K. P.,Basant, N.,Gupta, S.(2011).Support vector machines in water quality management.Analytica chimica acta,703(2),152-162.
46. Sokolova, M.,Japkowicz, N.,Szpakowicz, S.(2006).Beyond accuracy, F-score and ROC: a family of discriminant measures for performance evaluation.Australasian joint conference on artificial intelligence,Berlin, Heidelberg:
47. Tesoriero, A. J.,Gronberg, J. A.,Juckem, P. F.,Miller, M. P.,Austin, B. P.(2017).Predicting redox-sensitive contaminant concentrations in groundwater using random forest classification.Water Resources Research,53(8),7316-7331.
48. Vapnik, V.,Golowich, S. E.,Smola, A. J.(1997).Support vector method for function approximation, regression estimation and signal processing.Advances in neural information processing systems
49. Wang, P.,Lu, B.,Zhang, H.,Zhang, W.,Sun, Y.,JI, Y.(2014).Water demand prediction model based on random forests model and its application.Water Resources Protection,30,34-37.
50. Wang, X.,Jiake, L. V.,Deti, X.(2010).A hybrid approach of support vector machine with particle swarm optimization for water quality prediction.The 5th International Conference on Computer Science & Education,Hefei, China:
51. Wang, Y. Y.(1996).Department of Agricultural Engineering, National Taiwan University.
52. Yahya, A.,Saeed, A.,Ahmed, A. N.,Binti Othman, F,Ibrahim, R. K.,Afan, H. A.,Elshafie, A.(2019).Water Quality Prediction Model Based Support Vector Machine Model for Ungauged River Catchment under Dual Scenarios.Water,11(6),1231.
53. Yang, X. G.,Zhang, H. J.,Zhou, H. L.(2014).A Hybrid Methodology for Salinity Time Series Forecasting Based on Wavelet Transform and NARX Neural Networks.Arabian Journal for Science and Engineering,39(10),6895-6905.
54. Yao, P. H.,Shyu, G. S.,Cheng, B. Y.,Chang, Y. H.,Chang, T. K.(2013).The Water Footprints of Rice in Taiwan.Journal of Taiwan Agricultural Engineering,59,1-12.
55. Zhang, Y.,Gao, Q.(2016).Water quality evaluation of Chaohu Lake based on random forest method.Chinese Journal of Environmental Engineering,10,992-998.