A SOFTWARE SENSOR FOR IN-SITU MONITORING OF THE 5-DAY BIOCHEMICAL OXYGEN DEMAND

Abstract

Due to the time-consuming procedure for determining the 5-day biochemical oxygen demand (BOD5), the present study developed two software sensors based on artificial intelligence techniques to estimate this indicator instantaneously. For this purpose, feed-forward and radial basis function neural networks (FFANN and RBFANN, respectively) were tuned to estimate the maximum values of BOD5 (BOD5(max)) as a function of average, maximum and minimum dissolved oxygen in the Sefidrood River. Also, Levenberg-Marquardt (LM), resilient back propagation (RP), and scaled conjugate gradient (SCG) algorithms were used to optimize the FFANN parameters. The results demonstrated that the performance of LM algorithm in tuning the FFANN was better than others, in verification step. Besides, the performance of both FFANN and RBFANN models for prediction of the BOD5(max) was approximately the same.

Keywords

• FFANN
• RBFANN
• Dissolved oxygen
• Calibration
• BOD5