Modelling biological organic matter and nutrient removal processes from wastewater using respirometric and titrimetric techniques

Abstract

The increase in water demand in the 20th century brought the necessity of water treatment before being restored to the environment. The removal of organic matter (COD), nitrogen and phosphorus from wastewater is required to avoid the phenomenon of eutrophication. Eutrophication is defined as the over-enrichment of a water body with nutrients, resulting in excessive growth of organisms and depletion of oxygen concentration. A model for each of the main biological nutrient removal processes is developed, calibrated and validated in this thesis: biological COD removal, niological nitrogen oxidation (Nitrification), Enhanced Biological Phosphorus Removal (EBPR)

These models were developed so that they could be calibrated with common WWTP measurements: oxygen and pH. Therefore, the models are calibrated using respirometric and titrimetric techniques. On the one hand, respirometry is the measurement and interpretation of biological oxygen consumption rate. The oxygen uptake rate (OUR) profile gives information of the process rate. On the other hand, titrimetric techniques are based on the measurement of the proton production rate (HPR) through the amount of acid and base dosed to the system to maintain the pH constant.
Moreover, mathematical tools should be used to ensure the reliability of the parameter estimation values. Confidence intervals are as important as the parameter estimation value itself. The parameter estimation error assessment methodology used in this thesis is based on the Fisher Information Matrix (FIM), which is regarded as an indicator of the amount of information contained in the experimental data.

A new model for the description of the biological COD removal process was developed. The major achievement of this model was the inclusion of the simultaneous growth and storage on external substrate process. Moreover, it was extended to be calibrated with titrimetric measurements. This means that the model described the evolution of the CO2 profile in the reactor and the proton production (consumption) in each process. The model was successfully calibrated using respirometric/titrimetric measurements and helpful information of the system was obtained.
A new model for the description of the nitrification process was developed. This model was based on previous two-step nitrification models of the literature and extended with titrimetry. Moreover, the gas phase has been modelled as a serial of different CSTR for a better description of the CO2 profile. This model also included the description and modelling of the acceleration phase, typical from respirometric batch experiments with ammonium as substrate. The model contained more than forty parameters which needed to be estimated. Some of them were assumed from the literature. Some of them were calculated developing specific experiments. This thesis focused on the characterisation of the substrate limitation on the nitrification process and oxygen and inorganic carbon limitations were deeply analysed. Finally, the rest of the parameters were estimated with on-line common measurements: oxygen (OUR) and pH (HPR).

Finally, the EBPR process was also modelled. The model used was a modification of the widespread ASM2 model. The main modification introduced is the inclusion of glycogen as a component and, therefore, the inclusion of all the processes where glycogen is involved. This model was successfully used to study the differences between acetate and propionate as sole carbon source, to test new control strategies for the improvement of the start-up efficiency of a EBPR process in a Sequencing Batch Reactor and to study the effect of the coexistence of the electron donor (organic matter) and the electron acceptor (oxygen).

Full text:

Summary in English and Spanish
Chapter 1-3
Chapter 4
Chapter 5
Chapter 6B
Chapter 6E