Natural gas is one of the most exploited sources of energy worldwide. Since its composition is highly variable, its characteristics can change both in space and time, meaning that the energy carried by the gas has to be constantly monitored in order to evaluate the quality of the transiting mixture. In the Italian network this is typically performed by gas chromatographs, which provide the output parameters of interest (real volumetric higher heating value, real relative mass density and molar fraction of carbon dioxide) directly by measuring the composition through a chemical process. However, they are characterized by high costs and the interest of substituting them with innovative technologies, such as quality analyzers, is constantly growing. These devices measure specific physical properties and compute the gas quality parameters without knowing its composition. This work is focused on the study of a new prototype of these quality analyzers, which measures the molar fraction of carbon dioxide, the speed of sound and the dynamic viscosity. This instrument has been developed by a company that has contacted Politecnico di Milano in order to check its proper working at severe operating conditions. The quality analyzer is tested inside a climatic chamber, where the temperature can be arbitrarily changed, and both cold and dry hot tests are performed in order to check the instrument's effectiveness and accuracy, basing on the indications provided by the reference Standard EN 60068. The device is tested both with gas of certified bottles and with gas coming from the Italian network, which features an unknown composition. For this reason, a gas chromatograph is positioned outside of the climatic chamber and connected in parallel to the analyzer. The output parameters provided by the instrument are then compared with the ones obtained by the gas chromatograph through the Standard ISO 6976, which are used as the reference ones, and they have to lie within specific limits defined by the Network code. The tests are carried out both at the headquarters of the company itself and in the Laboratory of Micro-Cogeneration of Politecnico di Milano. The main objectives of the experimental campaign are checking the instrument's proper working and its dependence on the operative conditions, suggesting and applying modifications to the instrument aimed to improve the mathematical model with whom the physical properties are computed, corrected to a reference state and finally used to calculate the output parameters. The results of the experimental campaign show a strong dependence of the dynamic viscosity computation with respect to variations of both atmospheric pressure and ambient temperature, a relevant dependence of the carbon dioxide molar fraction measurement with respect to pressure, no significant dependence of speed of sound measurement with respect to the operating conditions and finally systematic errors due to calibration parameters evaluated at different operating conditions or with different gases from the ones utilized in the experiments. Each of the mentioned issues is isolated, singularly studied and solved thanks to corrections applied basing on the observations of the experimental campaign. In particular the pressure dependencies of the carbon dioxide molar fraction and of the time constant with whom dynamic viscosity is computed are corrected thanks to empirically determined sensitivity coefficients (0.00084 [1/mbar] and -0.01054 [s /mbar], respectively). The calibration parameters of the instrument are re-calculated, leading to a reduction of the relative error on dynamic viscosity from 0.462% to 0.07%, which implies a reduction of the higher heating value relative error from 0.77% to about 0.05%. A way to correct the effect of temperature on the dynamic viscosity measurement is also suggested, basing on the results of the environmental tests. Finally, the model with whom dynamic viscosity is calculated is modified, improved and successfully validated, decreasing the value of the root of the sum of the minimum squared errors between the pressure discharge data and the values provided by the model of at least 75%.
Il gas naturale è una delle fonti energetiche più sfruttate a livello mondiale. Poiché la sua composizione è altamente variabile, le sue caratteristiche sono anch'esse variabili sia nello spazio che nel tempo e l'energia trasportata dal gas deve essere costantemente monitorata al fine di valutare la qualità della miscela in transito. Nella rete italiana questa funzione viene tipicamente svolta dai gascromatografi, i quali forniscono i parametri in uscita di interesse (potere calorifico superiore reale volumetrico, densità massica relativa reale e frazione molare di anidride carbonica) misurando direttamente la composizione tramite un processo chimico. Tuttavia, essi sono caratterizzati da costi elevati e l'interesse a sostituirli con tecnologie innovative, come gli analizzatori di qualità, è in costante crescita. Questi dispositivi misurano specifiche proprietà fisiche e calcolano i parametri di qualità del gas senza conoscerne la composizione. Questo lavoro è focalizzato sullo studio di un nuovo prototipo di questi strumenti, che misura la frazione molare di anidride carbonica, la velocità del suono e la viscosità dinamica. Esso è stato sviluppato da un'azienda che ha contattato il Politecnico di Milano per verificare il corretto funzionamento dello strumento in condizioni operative gravose. L'analizzatore di qualità viene testato all'interno di una camera climatica, dove la temperatura può essere modificata arbitrariamente, e vengono eseguiti sia test a freddo che a caldo secco per verificarne l'efficacia e l'accuratezza, in base alle indicazioni fornite dalla normativa di riferimento EN 60068. Il dispositivo viene testato sia con gas di bombole certificate che con gas proveniente dalla rete gas italiana, il quale presenta una composizione ignota. Per questo motivo, un gascromatografo è posizionato all'esterno della camera climatica e collegato in parallelo all'analizzatore. I parametri in uscita forniti dallo strumento vengono poi confrontati con quelli ottenuti dal gascromatografo tramite la normativa ISO 6976, che vengono usati come riferimento, e devono trovarsi entro limiti specifici definiti dal Codice di rete. I test vengono eseguiti sia presso la sede dell'azienda stessa che nel Laboratorio di Micro-Cogenerazione del Politecnico di Milano. I principali obiettivi della campagna sperimentale sono il controllo del corretto funzionamento dello strumento e della sua dipendenza dalle condizioni operative, suggerendo e applicando modifiche allo strumento stesso volte a migliorare il modello matematico con cui le proprietà fisiche vengono calcolate, corrette ad uno stato di riferimento ed infine utilizzate per calcolare i parametri in uscita. I risultati della campagna sperimentale mostrano una forte dipendenza del calcolo della viscosità dinamica rispetto alle variazioni sia della pressione atmosferica che della temperatura ambiente, una dipendenza rilevante della misura della frazione molare di anidride carbonica rispetto alla pressione, nessuna dipendenza significativa della misura di velocità del suono rispetto alle condizioni operative ed infine degli errori sistematici dovuti a parametri di calibrazione valutati in condizioni operative diverse o con gas diversi da quelli utilizzati negli esperimenti. Ciascuno dei problemi citati viene isolato, studiato singolarmente e risolto grazie a correzioni applicate basandosi sulle osservazioni della campagna sperimentale. In particolare, le dipendenze rispetto alla pressione atmosferica della frazione molare di anidride carbonica e della costante di tempo con cui la viscosità dinamica viene calcolata vengono corrette grazie a coefficienti di sensitività determinati empiricamente (0.00084 [1/mbar] e -0.01054 [s /mbar], rispettivamente). I parametri di calibrazione dello strumento vengono ricalcolati, portando ad ottenere una riduzione dell'errore relativo sulla viscosità dinamica da 0.462% a 0.07%, che implica una riduzione dell'errore relativo sul potere calorifico superiore da 0.77% a circa 0.05%. Viene proposta anche una maniera per correggere l'effetto della temperatura sulla misura della viscosità dinamica, basandosi sui risultati dei test ambientali. Infine, il modello con cui viene calcolata la viscosità dinamica viene modificato, migliorato e validato con successo, diminuendo il valore della radice della somma degli errori minimi quadrati fra i dati della scarica di pressione e i valori forniti dal modello di almeno il 75%.
Theoretical development and environmental testing of a new natural gas quality analyzer based on dynamic viscosity measurements
NATALE, DAVIDE
2018/2019
Abstract
Natural gas is one of the most exploited sources of energy worldwide. Since its composition is highly variable, its characteristics can change both in space and time, meaning that the energy carried by the gas has to be constantly monitored in order to evaluate the quality of the transiting mixture. In the Italian network this is typically performed by gas chromatographs, which provide the output parameters of interest (real volumetric higher heating value, real relative mass density and molar fraction of carbon dioxide) directly by measuring the composition through a chemical process. However, they are characterized by high costs and the interest of substituting them with innovative technologies, such as quality analyzers, is constantly growing. These devices measure specific physical properties and compute the gas quality parameters without knowing its composition. This work is focused on the study of a new prototype of these quality analyzers, which measures the molar fraction of carbon dioxide, the speed of sound and the dynamic viscosity. This instrument has been developed by a company that has contacted Politecnico di Milano in order to check its proper working at severe operating conditions. The quality analyzer is tested inside a climatic chamber, where the temperature can be arbitrarily changed, and both cold and dry hot tests are performed in order to check the instrument's effectiveness and accuracy, basing on the indications provided by the reference Standard EN 60068. The device is tested both with gas of certified bottles and with gas coming from the Italian network, which features an unknown composition. For this reason, a gas chromatograph is positioned outside of the climatic chamber and connected in parallel to the analyzer. The output parameters provided by the instrument are then compared with the ones obtained by the gas chromatograph through the Standard ISO 6976, which are used as the reference ones, and they have to lie within specific limits defined by the Network code. The tests are carried out both at the headquarters of the company itself and in the Laboratory of Micro-Cogeneration of Politecnico di Milano. The main objectives of the experimental campaign are checking the instrument's proper working and its dependence on the operative conditions, suggesting and applying modifications to the instrument aimed to improve the mathematical model with whom the physical properties are computed, corrected to a reference state and finally used to calculate the output parameters. The results of the experimental campaign show a strong dependence of the dynamic viscosity computation with respect to variations of both atmospheric pressure and ambient temperature, a relevant dependence of the carbon dioxide molar fraction measurement with respect to pressure, no significant dependence of speed of sound measurement with respect to the operating conditions and finally systematic errors due to calibration parameters evaluated at different operating conditions or with different gases from the ones utilized in the experiments. Each of the mentioned issues is isolated, singularly studied and solved thanks to corrections applied basing on the observations of the experimental campaign. In particular the pressure dependencies of the carbon dioxide molar fraction and of the time constant with whom dynamic viscosity is computed are corrected thanks to empirically determined sensitivity coefficients (0.00084 [1/mbar] and -0.01054 [s /mbar], respectively). The calibration parameters of the instrument are re-calculated, leading to a reduction of the relative error on dynamic viscosity from 0.462% to 0.07%, which implies a reduction of the higher heating value relative error from 0.77% to about 0.05%. A way to correct the effect of temperature on the dynamic viscosity measurement is also suggested, basing on the results of the environmental tests. Finally, the model with whom dynamic viscosity is calculated is modified, improved and successfully validated, decreasing the value of the root of the sum of the minimum squared errors between the pressure discharge data and the values provided by the model of at least 75%.File | Dimensione | Formato | |
---|---|---|---|
Theoretical development and environmental testing of a new natural gas quality analyzer based on dynamic viscosity measurements.pdf
non accessibile
Descrizione: Testo della tesi
Dimensione
23.35 MB
Formato
Adobe PDF
|
23.35 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/10589/148844