Improved Water Quality Prediction with Hybrid Wavelet-Genetic Programming Model and Shannon Entropy
- PDF / 2,133,243 Bytes
- 22 Pages / 595.276 x 790.866 pts Page_size
- 38 Downloads / 158 Views
Original Paper
Improved Water Quality Prediction with Hybrid Wavelet-Genetic Programming Model and Shannon Entropy Hamideh Jafari,1 Taher Rajaee,1 and Ozgur Kisi
2,3
Received 31 January 2020; accepted 7 May 2020
Prediction of biochemical oxygen demand (BOD) as the main pollution indicators of organic pollution in freshwater resources is necessary. In the present work, a hybrid wavelet-genetic programming (WGP) method was implemented to improve prediction of BOD. The Shannon entropy was used to identify the optimal input combinations of WGP. In addition, an investigation was done to find which functions of wavelet and decomposition levels have better results in conjunction with genetic programming (GP). For comparison of WGP efficiency, five machine learning methods consisting of WANN (wavelet-artificial neural network), ANN (artificial neural network), GP, DT (decision tree) and BN (Bayesian network) were considered. Experiments on wavelet-processed data revealed that the best results were obtained when the models WGP and WANN were calibrated at three levels of decomposition using the Dmey mother wavelet function. The WGP model created rational forecasts for the peak BOD values. The results show that the use of Shannon entropy is suitable for determining the optimal composition of inputs to machine learning methods. Comparison of the results indicate that the WGP model is superior to the GP, ANN, DT, BN and WANN models based on data from the Varian Hotel and Dam Input stations. KEY WORDS: Wavelet-genetic programming, Shannon entropy, Bayesian network, Decision tree, Karaj River, BOD.
INTRODUCTION Surface water is more likely to be polluted by various sources of water. The delivery of external contaminants alters a stable links between physical, chemical and biological processes alternately contributing to the deterioration of water quality. Hence, with entry of the exterior materials, environmental interactions begin simultaneously (Alimoradi et al. 2018; Ramezani et al. 2019). Indexes of water pollution in a river can be measured by taking 1
Department of Civil Engineering, University of Qom, Qom, Iran. Department of Civil Engineering, Ilia State University, Tbilisi, Georgia. 3 To whom correspondence should be addressed; e-mail: [email protected]
2
the biochemical oxygen demand (BOD) into account. The BOD is the key parameter required for water quality assessment as well as the design of water resource management strategies. The determination of BOD is laborious and subjected to measurement errors resulting from sampling procedure and/or the existence of toxic substances in the specimen. Therefore, the determination of BOD under laboratory conditions can differ substantially from those in the field. The identification of a pattern in time series reflecting temporal variability of BOD is important for the prediction of future trends. Numerical simulations of surface water quality are very complicated and have difficult mathematics and often make it impossible to achieve at simple concepts that are n
Data Loading...
