Loading...

Predicting the Optimal Operation Pattern of Municipal Wastewater Treatment Plant Using Artificial Intelligence Approaches

Hakimi, Mahdi | 2022

59 Viewed
  1. Type of Document: M.Sc. Thesis
  2. Language: Farsi
  3. Document No: 56058 (09)
  4. University: Sharif University of Technology
  5. Department: Civil Engineering
  6. Advisor(s): Torkian, Ayoub
  7. Abstract:
  8. With the growth of the population and the significant expansion of industries in the last century, many environmental problems have plagued developed and developing countries. One of these environmental problems is water pollution. Observing the effects of water pollution over time, sanitary and industrial wastewater treatment was proposed as a reliable solution. With technology development, wastewater treatment requirements have become stricter. The increase in energy consumption and wastewater treatment costs due to population growth and industrialization on the one hand and strict regulations, on the other hand, have forced those involved in this field to employ a variety of technical and economic optimization methods. One of the solutions that has attracted researchers in this field is the use of artificial intelligence to optimize treatment processes. This research tries to find an optimal model for managing the wastewater treatment plant using machine learning algorithms. For this purpose, long-term quantitative and qualitative data of influent, effluent, and aeration tanks of the Southern Tehran Wastewater Treatment Plant were collected. Using two models, one based on the mutual information feature selection method and LSTM and the other based on CNN-LSTM, a hybrid prediction model, were developed to predict the quality characteristics of effluent. The results of this research showed that the model that utilizes the mutual information and LSTM provides better performance than the CNN-LSTM hybrid prediction model. Moreover, the most sensitive variables were identified for each model through sensitivity analysis. In the end, this research discusses that it is possible to achieve more optimal operation in wastewater treatment plants by having an efficient prediction model
  9. Keywords:
  10. Waste Treatment ; Artificial Intelligence ; Machine Learning ; Neural Network ; Mutual Information ; Prediction Models ; Long Short Term Memory (LSTM) ; Convolutional Neural Network

 Digital Object List

 Bookmark