The study describes the simulation of the biological processes at Milan San Rocco wastewater treatment plant (WWTP). The biological section of the plant is constituted by a sequence of aerated and anoxic tanks, with influent entering the plant according to a step-feed mode, and with a hybrid fixed/suspended biomass process, called IFAS (Integrated Fixed-film Activated Sludge). The study reports the methodology to build the mathematical model of the biological section by applying the Activated Sludge Model No. 1 (ASM1) and the Activated Sludge Model No . 2d (ASM2d). The physical structure of the model has been developed using the platform WEST® 2016 DHI. The model was calibrated with data of the influent loads in the period between 6 and 10 June 2016, using the maximum growth rate of autotrophic biomass, as measured by means of laboratory analysis. The model was finally validated by using a second data set (10 June - 3 July 2016) with a time step of 30 minutes and, then, for the full year 2016, with a daily time step. The SST concentrations in the tank and in the recirculation flow results of the stationary and dynamic simulations show good congruence with the values measured by the probes, placed in-situ. In some cases the model overestimated nitrate concentrations compared to the measured values. The discrepancy can be explained: - partly because the model cannot adequately simulate the denitrification processes that take place in tanks with intermittent aeration, as they are not available as hourly data during the periods of activation and deactivation of air diffusers, which are automatically operated by the PID control system; - partly because the values measured by the probes (used for calibration and validation) can deviate from the actual values in the tank, as found in some cases by comparing the values between probe response and laboratory analysis on spot samples in the same time instant. The model has allowed the analysis of possible interventions for improvement, aimed at reducing the risk of management, energy consumption, the use of chemical reagents, and to obtain a better effluent quality.
Lo studio descrive la simulazione del processo biologico dell’impianto di Milano San Rocco, caratterizzato da una configurazione impiantistica non convenzionale, costituita da una sequenza di vasche aerate e anossiche con alimentazione a step-feed, in cui il processo a fanghi attivi è parzialmente integrato da un processo a biomassa adesa. Lo studio documenta le modalità di acquisizione dei dati e di riconciliazione degli stessi per l’ottenimento degli input o dei valori di confronto per la costruzione del modello matematico mediante l’Activated Sludge Model No. 1 (ASM1) e l’Activated Sludge Model No. 2d (ASM2d). La struttura fisica del modello è stata sviluppata utilizzando la piattaforma West® 2016 di DHI. Sono stati acquisiti i dati di processo dell’impianto, in particolare: le portate affluenti, i carichi inquinanti in ingresso ai processi biologici, le concentrazioni in ingresso e in uscita dei principali inquinanti, le concentrazioni di biomassa nelle vasche. Il modello è stato calibrato con i dati dei carichi influenti nel periodo compreso tra il 6 e 10 giugno 2016, utilizzando i tassi massimi di crescita della biomassa autotrofa individuati medianti analisi di laboratorio ed è stato infine validato utilizzando prima un secondo set di dati (10 giugno - 3 luglio 2016) con passo temporale di 30 minuti e, poi, sull’intero anno 2016, con passo temporale giornaliero. Le concentrazioni di SST in vasca e nella portata di ricircolo risultati delle simulazioni stazionarie e dinamiche presentano una buona congruenza con i valori misurati dalle sonde in continuo. Il modello sovrastima in alcuni casi le concentrazioni di azoto nitrico rispetto ai valori misurati. La discrepanza si può spiegare: - in parte nel fatto che il modello non può simulare adeguatamente i processi di denitrificazione che avviene nelle vasche con aerazione intermittente, in quanto non sono disponibili i dati orari dei periodi di attivazione e disattivazione dei diffusori d'aria, che avvengono in automatico in funzione del sistema di controllo GreenBass®; - in parte nel fatto che i valori misurati dalla sonda in continuo (usati per la calibrazione e la validazione) si possono discostare dai valori effettivi in vasca, come riscontrato in alcuni casi con analisi di laboratorio. L’utilizzo del modello costruito ha permesso l’analisi di possibili interventi di miglioramento, volti a ridurre i rischi di gestione, i consumi energetici, l’uso di reagenti chimici, e ad ottenere una migliore qualità dell’effluente.
Modellazione dei processi biologici del depuratore Milano San Rocco e analisi di possibili scenari di miglioramento
LENNA, SIMONE
2015/2016
Abstract
The study describes the simulation of the biological processes at Milan San Rocco wastewater treatment plant (WWTP). The biological section of the plant is constituted by a sequence of aerated and anoxic tanks, with influent entering the plant according to a step-feed mode, and with a hybrid fixed/suspended biomass process, called IFAS (Integrated Fixed-film Activated Sludge). The study reports the methodology to build the mathematical model of the biological section by applying the Activated Sludge Model No. 1 (ASM1) and the Activated Sludge Model No . 2d (ASM2d). The physical structure of the model has been developed using the platform WEST® 2016 DHI. The model was calibrated with data of the influent loads in the period between 6 and 10 June 2016, using the maximum growth rate of autotrophic biomass, as measured by means of laboratory analysis. The model was finally validated by using a second data set (10 June - 3 July 2016) with a time step of 30 minutes and, then, for the full year 2016, with a daily time step. The SST concentrations in the tank and in the recirculation flow results of the stationary and dynamic simulations show good congruence with the values measured by the probes, placed in-situ. In some cases the model overestimated nitrate concentrations compared to the measured values. The discrepancy can be explained: - partly because the model cannot adequately simulate the denitrification processes that take place in tanks with intermittent aeration, as they are not available as hourly data during the periods of activation and deactivation of air diffusers, which are automatically operated by the PID control system; - partly because the values measured by the probes (used for calibration and validation) can deviate from the actual values in the tank, as found in some cases by comparing the values between probe response and laboratory analysis on spot samples in the same time instant. The model has allowed the analysis of possible interventions for improvement, aimed at reducing the risk of management, energy consumption, the use of chemical reagents, and to obtain a better effluent quality.| File | Dimensione | Formato | |
|---|---|---|---|
|
2017_04_Lenna.pdf
accessibile in internet per tutti
Descrizione: Testo della tesi
Dimensione
7.33 MB
Formato
Adobe PDF
|
7.33 MB | Adobe PDF | Visualizza/Apri |
I documenti in POLITesi sono protetti da copyright e tutti i diritti sono riservati, salvo diversa indicazione.
https://hdl.handle.net/10589/133611