It is proposed a study for the experimental and numerical characterization of mechanical behavior of sandstone aimed at the implementation of a numerical model to simulate rock perforations in the oil and gas field. The use of numerical simulations is essential for developing the Dual Reciprocating Drilling (DRD) system, which is an innovative, bioinspired system of perforation already developed in the aerospace field. The extension of the application of the DRD requires the redesign of the system in order to satisfy different specifications. One of the main design variable is the geometry of the drill bit, and for its optimization experimental test or numerical simulations are necessary. The first step to develop a numerical campaign is the implementation of a numerical model able to simulate the mechanical behavior of rock and a reference material must be chosen. The selected reference material is Berea Sandstone which is commonly used in the field. The low availability of Berea Sandstone in Italy, imposes to use another material for the experimental tests, which is Pietra Serena. Berea Sandstone and Pietra Serena have got similar mechanical properties and so the study takes both the materials into consideration as long as numerical simulations are concerned. The main failure criteria for rocks, available in the software Abaqus, are the Mohr-Coulomb, Drucker-Prager and Concrete Damage Plasticity models. These material models are calibrated and implemented to investigate which one is able to simulate more closely the mechanical behavior. The finite element model is implemented with the conversion to SPH particles, which is fundamental for the future simulations of perforation. First of all, the behavior of the Mohr-Coulomb and Drucker-Prager criteria is investigated, taking into consideration experimental data of Berea Sandstone found in literature. Subsequently, an experimental campaign with Pietra Serena is performed for the calibration and the validation of the Mohr-Coulomb, Drucker-Prager and Concrete Damage Plasticity models. The unconfined compression test and the Brazilian test are performed for the calibration of the material parameters. The flexural test is performed for validation. The experimental tests performed are characterized by the use of only one test machine and simple specimens: consequently, the proposed procedure is interesting on the point of view of resources and time required. The numerical models are validated also simulating the uniaxial tension test, whose results for Pietra Serena are available in literature. The material model which shows to better replicate the experimental tests is the Mohr-Coulomb. Consequently, a final material model is proposed which is a Mohr-Coulomb model calibrated on the experimental tests which, instead, uses the friction angle obtained from data of Berea Sandstone. The friction angle is a material parameter which describes the different strength at different levels of confinement. The proposed method to obtain the friction angle from the experimental test data is reliable only for states of stress near to the tested conditions. The proposed method to determine the friction angle from the data of Berea Sandstone, instead, should be suitable for a large variety of stress conditions. Therefore, it is assumed that Pietra Serena has got the same friction angle of Berea Sandstone. The unconfined penetration test is also performed on the final proposed material model. The results are compared with the data of an experimental test previously performed by Politecnico di Milano. In conclusion, the proposed numerical model is able to replicate the unconfined perforation both in terms of predicted load and fracture pattern.
Viene presentato uno studio sulla caratterizzazione sperimentale e numerica del comportamento meccanico dell’arenaria per lo sviluppo di un modello numerico in grado di simulare la perforazione in roccia nell’ambito dell’industria oil and gas. L’uso delle simulazioni numeriche è fondamentale per sviluppare il sistema Dual Reciprocating Drilling (DRD), un sistema innovativo, bio-ispirato all’ovopositore dell’insetto Megarhyssa, già sviluppato in ambito aerospaziale. L’applicazione in ambito oil and gas necessita la riprogettazione del sistema DRD per soddisfare diverse specifiche. Una delle tante variabili di progetto è la geometria della punta e per la sua ottimizzazione posso essere seguiti due approcci: una campagna sperimentale o numerica. Il primo passo per lo sviluppo di una campagna numerica è l’implementazione di un modello numerico capace di simulare il comportamento meccanico della roccia, scegliendo come materiale di riferimento Berea Sandstone, che è il tipico materiale utilizzato nell’ambito. Per le prove sperimentali è stata utilizzata Pietra Serena, poiché Berea Sandstone non è facilmente reperibile in Italia. I due materiali hanno proprietà meccaniche simili e sono considerati entrambi in questo studio. I principali criteri di rottura disponibili nel software Abaqus sono Mohr-Coulomb, Drucker-Prager e Concrete Damage Plasticity. Questi tre modelli sono calibrati e implementati per valutare quale sia il migliore nel simulare il comportamento meccanico. È stato utilizzato il metodo degli elementi finiti con la conversione a particelle SPH, fondamentale per le future simulazioni di perforazione. Inizialmente, è stato studiato il comportamento dei modelli di Mohr-Coulomb e Drucker-Prager, considerando i dati sperimentali di Berea Sandstone trovati in letteratura. Successivamente è stata effettuata una campagna sperimentale con Pietra Serena, per la calibrazione e la validazione dei modelli di Mohr-Coulomb, Drucker-Prager e Concrete Damage Plasticity. Sono stati utilizzati il test di compressione non confinata e il Brazilian test per la calibrazione dei parametri di caratterizzazione del materiale. Si è utilizzato il test di flessione per validare i modelli numerici. Le prove sperimentali sono state effettuate con un solo macchinario e i provini sono di geometria semplice: quindi la procedura proposta è interessante dal punto di vista delle risorse e del tempo necessari. I modelli numerici sono validati inoltre dalla simulazione di test di tensione monoassiale, comparando i risultati numerici con i risultati sperimentali trovati in letteratura. Il modello numerico che meglio replica i test sperimentali è Mohr-Coulomb: di conseguenza si propone un modello di materiale, il quale consiste in un modello Mohr-Coulomb calibrato sui test sperimentali, ma che utilizza il friction angle ottenuto dai dati di Berea Sandstone. Il friction angle è un parametro del materiale che descrive la diversa resistenza per diversi livelli di confinamento. Il metodo proposto per determinare il friction angle dai dati sperimentali è affidabile solo per stati di sforzo vicini a quelli testati. Il metodo proposto per determinare il friction angle dai dati di Berea Sandstone, invece porta a risultati affidabili per una gamma più ampia di stati di sforzo. Si ipotizza quindi, che il friction angle di Pietra Serena sia uguale a quello di Berea Sandstone. Il test di penetrazione non confinato è simulato con il materiale proposto, comparando i risultati con test sperimentali già effettuati dal Politecnico di Milano. Si dimostra che il modello numerico proposto è in grado di replicare la perforazione in termini di carichi previsti e tipo di frattura.
Experimental and numerical characterization of mechanical behaviour of sandstone
SCAZZOSI, RICCARDO;MAIOLANI, FIORENZO
2015/2016
Abstract
It is proposed a study for the experimental and numerical characterization of mechanical behavior of sandstone aimed at the implementation of a numerical model to simulate rock perforations in the oil and gas field. The use of numerical simulations is essential for developing the Dual Reciprocating Drilling (DRD) system, which is an innovative, bioinspired system of perforation already developed in the aerospace field. The extension of the application of the DRD requires the redesign of the system in order to satisfy different specifications. One of the main design variable is the geometry of the drill bit, and for its optimization experimental test or numerical simulations are necessary. The first step to develop a numerical campaign is the implementation of a numerical model able to simulate the mechanical behavior of rock and a reference material must be chosen. The selected reference material is Berea Sandstone which is commonly used in the field. The low availability of Berea Sandstone in Italy, imposes to use another material for the experimental tests, which is Pietra Serena. Berea Sandstone and Pietra Serena have got similar mechanical properties and so the study takes both the materials into consideration as long as numerical simulations are concerned. The main failure criteria for rocks, available in the software Abaqus, are the Mohr-Coulomb, Drucker-Prager and Concrete Damage Plasticity models. These material models are calibrated and implemented to investigate which one is able to simulate more closely the mechanical behavior. The finite element model is implemented with the conversion to SPH particles, which is fundamental for the future simulations of perforation. First of all, the behavior of the Mohr-Coulomb and Drucker-Prager criteria is investigated, taking into consideration experimental data of Berea Sandstone found in literature. Subsequently, an experimental campaign with Pietra Serena is performed for the calibration and the validation of the Mohr-Coulomb, Drucker-Prager and Concrete Damage Plasticity models. The unconfined compression test and the Brazilian test are performed for the calibration of the material parameters. The flexural test is performed for validation. The experimental tests performed are characterized by the use of only one test machine and simple specimens: consequently, the proposed procedure is interesting on the point of view of resources and time required. The numerical models are validated also simulating the uniaxial tension test, whose results for Pietra Serena are available in literature. The material model which shows to better replicate the experimental tests is the Mohr-Coulomb. Consequently, a final material model is proposed which is a Mohr-Coulomb model calibrated on the experimental tests which, instead, uses the friction angle obtained from data of Berea Sandstone. The friction angle is a material parameter which describes the different strength at different levels of confinement. The proposed method to obtain the friction angle from the experimental test data is reliable only for states of stress near to the tested conditions. The proposed method to determine the friction angle from the data of Berea Sandstone, instead, should be suitable for a large variety of stress conditions. Therefore, it is assumed that Pietra Serena has got the same friction angle of Berea Sandstone. The unconfined penetration test is also performed on the final proposed material model. The results are compared with the data of an experimental test previously performed by Politecnico di Milano. In conclusion, the proposed numerical model is able to replicate the unconfined perforation both in terms of predicted load and fracture pattern.File | Dimensione | Formato | |
---|---|---|---|
2016_12_Maiolani_Scazzosi.pdf
non accessibile
Descrizione: Testo della tesi
Dimensione
14.47 MB
Formato
Adobe PDF
|
14.47 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/10589/131308