Journal of Information Processing Systems

Korea Information Processing Society (한국정보처리학회)
권호 표지
DOI QR코드

DOI QR Code

CNN-LSTM Coupled Model for Prediction of Waterworks Operation Data

  • Received : 2018.10.01
  • Accepted : 2018.11.27
  • Published : 2018.12.31
Download PDF
⟨ Previous Next ⟩

Abstract

In this paper, we propose an improved model to provide users with a better long-term prediction of waterworks operation data. The existing prediction models have been studied in various types of models such as multiple linear regression model while considering time, days and seasonal characteristics. But the existing model shows the rate of prediction for demand fluctuation and long-term prediction is insufficient. Particularly in the deep running model, the long-short-term memory (LSTM) model has been applied to predict data of water purification plant because its time series prediction is highly reliable. However, it is necessary to reflect the correlation among various related factors, and a supplementary model is needed to improve the long-term predictability. In this paper, convolutional neural network (CNN) model is introduced to select various input variables that have a necessary correlation and to improve long term prediction rate, thus increasing the prediction rate through the LSTM predictive value and the combined structure. In addition, a multiple linear regression model is applied to compile the predicted data of CNN and LSTM, which then confirms the data as the final predicted outcome.

Keywords

E1JBB0_2018_v14n6_1508_f0001.png 이미지

Fig. 1. Conceptual construction of proposal model.

E1JBB0_2018_v14n6_1508_f0002.png 이미지

Fig. 2. Process of calculating and verifying significance of multiple linear regression model.

E1JBB0_2018_v14n6_1508_f0003.png 이미지

Fig. 3. Pattern of operational data sensor. (a) Flow rate and (b) water level.

E1JBB0_2018_v14n6_1508_f0004.png 이미지

Fig. 4. The RMSE per sensor in each model prediction.

E1JBB0_2018_v14n6_1508_f0005.png 이미지

Fig. 5. Comparison chart of RMSE between models.

Table 1. Performance comparison between proposed method and existing method

E1JBB0_2018_v14n6_1508_t0001.png 이미지

References

  1. J. R. Kim, G. W. Shin, S. T. Hong, and G. H. Yoo, "A study on detecting algorithm for outlier data in water supply," in Proceedings of the Korean Institute of Communications and Information Sciences Summer Conference, 2016, pp. 327-328, 2016.
  2. J. H. Seok, J. J. Kim, J. Y. Lee, J. J. Lee, Y. J. Song, G. W. Shin, "Abnormal data refinement and error percentage correction methods for effective short-term hourly water demand forecasting," International Journal of Control, Automation and Systems, vol. 12, no. 6, pp 1245-1256, 2014. https://doi.org/10.1007/s12555-014-0001-z
  3. G. H. Yoo, J. R. Kim, and G. W. Shin, "A study on the cleansing of water data using LSTM algorithm," in Proceedings of the Korea Institute of Information and Communication Engineering Fall Conference, 2017, pp. 501-503, 2017.
  4. H. S. Kim, "Data cleansing algorithm for reducing outlier of water supply information," Ph.D. dissertation, Pai Chai University, Daejeon, Korea, 2018.
  5. E. J. Lee "Basic and applied research of CNN and RNN," Broadcasting and Media Magazine, vol. 22, no. 1, pp. 87-95, 2017.
  6. Y. H. Kim, Y. K. Hwang, T. G. Kang, and K. M. Jung, "LSTM language model based korean sentence generation," The Journal of Korean Institute of Communications and Information Sciences, vol. 41, no. 5, pp. 592-601, 2016. https://doi.org/10.7840/kics.2016.41.5.592
  7. G. D. Sim and J. H. Seok, "A study on the prediction interest rate using LSTM systems," in Proceedings of the Korean Institute of Communications and Information Sciences Winter Conference, 2018, pp. 308-309.
  8. C. Kim and J. W. Kang, "LSTM based anomaly detection on semiconductor manufacturing data," in Proceedings of the Korea Software Conference 2017, Busan, Korea, 2017, pp. 760-762.