Parameter estimation using pressure and temperature transient data from wireline formation tester interval pressure transient tests (IPTT)

Reservoir behavior may be considered as isothermal because changes in temperature are generally small. In the past, use of transient temperature measurements had limitations due to the simplified assumption of isothermal fluid flow. Moreover, owing to poor resolution of temperature devices, small differences in the bottomhole (sandface) temperatures could not be detected. On the other hand, recent enhancement in the resolution of the downhole temperature devices allowed researchers to analyze and interpret transient temperature measurements to estimate near wellbore and reservoir characteristics. Therefore, the use of bottomhole temperature analysis has been expanded due to the development in resolution and accuracy in downhole gauges and the information that temperatures retain. Traditional pressure transient analysis has been used to obtain reservoir characteristics for a long time for conventional well testing and interval pressure transient test (IPTT). Recently, several works showed that near wellbore parameters can be acquired from transient temperature measurements. Previous works on transient temperature analysis performed based on slightly compressible fluid for fully-penetrated vertical or horizontal well in a single or multilayer homogeneous reservoir. Nonetheless, none of the studies analyzed temperature transient measurements acquired from the interval pressure transient test in partially-penetrated well to use in parameter estimation. Temperature data is obtained in IPTT; however, it is never being accounted in interpretation. The main target of this study is to estimate formation and near wellbore characteristics, especially horizontal and vertical permeability in the damaged zone and undamaged zone in a reservoir where damage in the formation is modeled as composite, by analyzing transient temperature measurements from the interval pressure transient test (IPTT). Furthermore, the effect of mud filtrate invasion on transient temperature measurements and parameter estimation is investigated. In this study, one of the commercial multi-purpose nonisothermal CMG-STARS simulator based on the fully-implicit numerical procedure is used. In the course of modeling, well block pressure and temperature in the simulator, some oil physical properties such as density, isothermal compressibility, and formation volume factor are allowed to vary with pressure and temperature. On the other hand, oil, water, and solid matrix specific heat capacities and isobaric thermal expansion coefficients are taken into consideration as constant with temperature. In the model, viscosity is treated as constant, at the initial conditions. To model conduction with oil, water, and rock system, a constant value of thermal conductivity is used. J-T expansion coefficient is accounted implicitly in the simulator through the variation enthalpy with pressure. Dual-packer tool configuration is used in modeling to compute transient temperature data to apply straight-line and nonlinear regression analysis including the effects of damage in the formation and mud filtrate invasion along with anisotropy. Probe configuration is modeled to generate transient temperature data to study nonlinear regression analysis including the effects of damage in the formation together with anisotropy. In this study, a commercial software CMOST AI for assisted history matching is used as an optimization module. A stochastic-based optimization technique that minimizes the difference between the observed data and the simulation responses as quantified with an objective function is used. To begin with, a synthetic model configured with a dual-packer tool is used to compute transient temperature data with constant oil production in a partially-penetrated well for a thin reservoir. Absolute temperature derivative showed that temperature propagates much slower than the pressure in the formation; thus, radial flow regime exhibits at very late times both in the damaged and undamaged zone. Horizontal permeability could be estimated precisely in the case of zero damage. In the presence of damage, horizontal permeability in the damaged zone may be estimated well if damaged zone radius is long enough for the reason that J-T expansion masks intermediate-time IARF for small damaged zone radius. In addition, another synthetic model constructed with the dual-packer tool is used to generate transient temperature data with constant production of oil in a partially-penetrated well for a thick reservoir. Nonlinear regression analysis is applied to only transient temperatures computed from this model. Horizontal permeability in the damaged zone is estimated correctly in both isotropic and anisotropic cases. Moreover, it is assessed precisely under shortened test period and mud filtrate invasion effect. The estimation of horizontal permeability in the undamaged zone is meaningful in case of no damage effect or small damaged zone radius in both isotropic and anisotropic cases. On the other hand, deep damaged zone radius and short test time result in deviation at the true value of horizontal permeability in the undamaged zone. Unfortunately, the estimations of vertical permeability in both damaged and undamaged zone are not close enough to the correct values due to the sensitivity of this parameter to temperature transient. Lastly, a synthetic model created with sink probe tool is used to compute transient temperatures with constant production of oil in a partially-penetrated well for a thick reservoir. Similarly, nonlinear regression analysis is applied to only transient temperatures generated from this model. In the case of isotropy, horizontal permeability in the damaged zone is estimated accurately, and horizontal permeability in the undamaged zone is estimated exactly only in absence of damage. When the reservoir is homogeneous and anisotropic, none of the parameters is estimated precisely. Moreover, the impact of various rock, fluid, and thermal properties on transient drawdown temperature is investigated. It is observed that reservoir thickness, porosity, and rock heat capacity do not have a significant effect on temperatures from IPTT. On the other hand, penetration interval, horizontal permeability, and viscosity may create extensive temperature changes. Also, the presence of skin and anisotropy have a substantial impact on transient temperatures from IPTT.