SUMMARY

The increasing urban growth, the unsustainable use of the natural resources and the society awareness of the environmental impact, highlight the necessity to develop and implement advanced technologies aimed to prevent, mitigate, correct and offset the pollution problems derived from anthropogenic origin. Currently, one key environmental problem is the wastewater treatment to be applied depending on its discharging place and taking into account its possible reuse before its discharge.

With regard to legislation, the Directive 2000/60/EC of the European Parliament and the Council of Europe provides a framework for community action in the water policy field. Specifically for wastewater treatment, the Directive 91/271/EC establishes the discharging limits of pollutants from wastewater treatment plants (WWTPs), and defines the concept of “sensitive areas”, which denotes the importance of removing nutrients from wastewater, in particular nitrogen and phosphorus, due to their negative impact on the aquatic ecosystems.

Several studies have shown the possibility of optimizing WWTP operation by means of processes that allow removing nitrogen via nitrite from wastewater streams with high ammonium concentration, such as the supernatant from anaerobic sludge digestion. The scientific community has also tackled the improvement of the instrumentation and process assessment related to nitrification and denitrification via nitrite. In this respect, different methodologies and strategies have been developed to monitor, control and automate in real time the operation of these processes in order to: fulfill the discharge requirements established in the legislation; minimize the negative impact to the environmental; and reduce operational costs in WWTP operation.

The main objective of this PhD thesis has been to study, at laboratory scale, the nitrogen removal process via nitrite in order to reduce the ammonium concentration in the supernatant from anaerobic sludge digestion.

Initially, several starting-up strategies for the so-called SHARON reactor have been studied in order to achieve a proper partial nitritation process. After achieving a stable process operation, activated sludge was withdrawn from the partial nitritation SHARON reactor to study the effect of several environmental and operating factors on the ammonia-oxidizing bacteria (AOB) and on the effluent quality of the SHARON reactor, respectively. Respirometric batch experiments were carried out in order to determine and quantify the effect of these environmental and operating factors on the performance and stability of the SHARON process. These results are included and discussed in this PhD thesis.

Subsequently, several starting-up strategies for the SHARON reactor have been studied in order to achieve a suitable nitrification and denitrification process via nitrite, using organic matter as external carbon source. On the basis of the information provided by pH, redox potential and dissolved oxygen sensors installed in the reactor, and their correlations found with the nitrogenous compounds concentrations throughout the stages of the start-up and process operating periods, a monitoring strategy was developed in order to establish the criteria of optimum process operation. This information was used to develop a fuzzy-logic-based control system based on pH and redox potential sensors, which main aim was to optimise the nitrification and denitrification process via nitrite in a SHARON reactor.

The developed controller includes two independent control algorithms to optimize both aerobic and anoxic stage durations of each operation cycle, and organic matter dosage. The first algorithm commands in real time the optimal aerobic and anoxic stage durations using the pH evolution profile in each stage, whereas the second algorithm controls the organic matter dosage at the beginning of the anoxic stage in each operation cycle based on the information provided by pH and redox potential sensors.

In this PhD thesis the development, implementation and validation of the control system in a SHARON reactor fed with synthetic supernatant whose composition is similar to the supernatant from anaerobic sludge digestion, and with supernatant from the anaerobic sludge digestion of the Carraixet WWTP (Valencia) are described.

The results from the control system performance after its implementation and validation have shown that it is possible to control automatically in real time the nitrogen removal via nitrite using robust and low-cost sensors. The main novelty of the control system lies on its flexibility and capacity to real-time adaptation to variations in both wastewater influent characteristics and process state.