Abstract: In one embodiment, a method for auditing the results of a machine learning model includes: retrieving a set of state estimates for original time series data values from a database under audit; reversing the state estimation computation for each of the state estimates to produce reconstituted time series data values for each of the state estimates; retrieving the original time series data values from the database under audit; comparing the original time series data values pairwise with the reconstituted time series data values to determine whether the original time series and reconstituted time series match; and generating a signal that the database under audit (i) has not been modified where the original time series and reconstituted time series match, and (ii) has been modified where the original time series and reconstituted time series do not match.

Abstract: The invention provides fatigue testing of a material specimen while the specimen is disposed in a high pressure fluid environment. A specimen is placed between receivers in an end cap of a vessel and a piston that is moveable within the vessel. Pressurized fluid is provided to compression and tension chambers defined between the piston and the vessel. When the pressure in the compression chamber is greater than the pressure in the tension chamber, the specimen is subjected to a compression force. When the pressure in the tension chamber is greater than the pressure in the compression chamber, the specimen is subjected to a tension force. While the specimen is subjected to either force, it is also surrounded by the pressurized fluid in the tension chamber. In some examples, the specimen is surrounded by hydrogen.

Abstract: Provided are several examples of apparatuses and methods for applying loads to material specimens in pressurized fluid environments. The apparatuses use the fluid pressure as the force for applying the loads to the specimens.

Abstract: A method for determining fracture toughness KIC of materials ranging from metallic alloys, brittle ceramics and their composites, and weldments. A cylindrical specimen having a helical V-groove with a 45° pitch is subjected to pure torsion. This loading configuration creates a uniform tensile-stress crack-opening mode, and a transverse plane-strain state along the helical groove. The full length of the spiral groove is equivalent to the thickness of a conventional compact-type specimen. KIC values are determined from the fracture torque and crack length measured from the test specimen using a 3-D finite element program (TOR3D-KIC) developed for the purpose. In addition, a mixed mode (combined tensile and shear stress mode) fracture toughness value can be determined by varying the pitch of the helical groove. Since the key information needed for determining the KIC value is condensed in the vicinity of the crack tip, the specimen can be significantly miniaturized without the loss of generality.

Abstract: A method for determining fracture toughness KIC of materials ranging from metallic alloys, brittle ceramics and their composites, and weldments. A cylindrical specimen having a helical V-groove with a 45° pitch is subjected to pure torsion. This loading configuration creates a uniform tensile-stress crack-opening mode, and a transverse plane-strain state along the helical groove. The full length of the spiral groove is equivalent to the thickness of a conventional compact-type specimen. KIC values are determined from the fracture torque and crack length measured from the test specimen using a 3-D finite element program (TOR3D-KIC) developed for the purpose. In addition, a mixed mode (combined tensile and shear stress mode) fracture toughness value can be determined by varying the pitch of the helical groove. Since the key information needed for determining the KIC value is condensed in the vicinity of the crack tip, the specimen can be significantly miniaturized without the loss of generality.

Abstract: Apparatus for tensile testing plate-type ceramic specimens having dogbone- or T-shaped end sections without introducing bending stresses in the specimens during the application of a dynamic tensile loading on the specimens is described. A pair of elongated pull rods disposed in a side-by-side relationship are used to grip the shoulders on each T-shaped end section. The pull rods are pivotally attached to a piston-displaceable, disk-shaped member so as to be longitudinally movable with respect to one another effecting the self-alignment thereof with the shoulders on the T-shaped end sections of the specimen to compensate for shoulders being located in different longitudinal positions.

Abstract: The present invention is directed to a self-aligning grip housing assembly that can transmit an uniaxial load to a tensil specimen without introducing bending stresses into the specimen. Disposed inside said grip housing assembly are a multiplicity of supporting pistons connected to a common source of pressurized oil that carry equal shares of the load applied to the specimen irregardless whether there is initial misalignment between the specimen load column assembly and housing axis.