Articles | Volume 20, issue 6
https://doi.org/10.5194/hess-20-2267-2016
https://doi.org/10.5194/hess-20-2267-2016
Research article
 | 
14 Jun 2016
Research article |  | 14 Jun 2016

Dissolved oxygen prediction using a possibility theory based fuzzy neural network

Usman T. Khan and Caterina Valeo

Abstract. A new fuzzy neural network method to predict minimum dissolved oxygen (DO) concentration in a highly urbanised riverine environment (in Calgary, Canada) is proposed. The method uses abiotic factors (non-living, physical and chemical attributes) as inputs to the model, since the physical mechanisms governing DO in the river are largely unknown. A new two-step method to construct fuzzy numbers using observations is proposed. Then an existing fuzzy neural network is modified to account for fuzzy number inputs and also uses possibility theory based intervals to train the network. Results demonstrate that the method is particularly well suited to predicting low DO events in the Bow River. Model performance is compared with a fuzzy neural network with crisp inputs, as well as with a traditional neural network. Model output and a defuzzification technique are used to estimate the risk of low DO so that water resource managers can implement strategies to prevent the occurrence of low DO.

Download
Short summary
This paper contains a new two-step method to construct fuzzy numbers using observational data. In addition an existing fuzzy neural network is modified to account for fuzzy number inputs. This is combined with possibility-theory based intervals to train the network. Furthermore, model output and a defuzzification technique is used to estimate the risk of low Dissolved Oxygen so that water resource managers can implement strategies to prevent the occurrence of low Dissolved Oxygen.