A New Model on Triaxial Compressive Strength of Sandstone Based on Deep-Well Logging Data

The inversion of rock mechanical properties based on well-logging data is a common method to drillbit selection in petroleum engineering, the basic principle of which is to use deep-well logging data and experimental regression model to calculate rock mechanical properties. The profile of rock mechanical properties can be established. According to the profile, the type of drillbit can be selected in different formations, which will improve the economic benefit of bit use indicators. Compressive strength of rock is one of the most important parameters. Conventional method to establish regression models of acoustic velocity and uniaxial compressive strength is on the basis of rock core sonic experiments and uniaxial compressive experiments. However, with the increase of deep drilling operation, triaxial compressive strength prediction model will play a more vital role in drillbit optimization selection in deep wells and improving ROP (Rate of Penetration). Triaxial compressive strength is an important parameter to evaluate the formation strength property. In this paper, to the sandstone commonly encountered in oil drilling, combined the rock mechanical experimental results with the changing porosity experimental results under confining pressure, a new interpretation model of triaxial compressive strength based on well logging data is proposed. Firstly, three types of sandstone samples are collected. Then measured components, uniaxial compressive strength, triaxial compressive strength and porosity. Experimental results show that: confining pressure has a significant influence on sandstone strength, with the increase of confining pressure, triaxial compressive strength increase; furthermore, the triaxial compressive strength of different porosity sandstones have different change rules when confining pressures changes; sandstone porosity decreases with the increase of confining pressure and it obeys logarithmic function. According to the results, regression models of sandstone, in-situ uniaxial compressive strength model and uniaxial compressive strength model under normal pressure are generated. Meanwhile, using well logging data in the acoustic velocity, density and gamma value can derive in-situ rock formations uniaxial compressive strength and in-situ porosity. Eventually, a new model established in this paper, containing in porosity, uniaxial compressive strength, lateral pressure and power exponent of regression, can predict triaxial compressive strength which will guide drillbit selection.