Purpose: As artificial intelligence (AI) systems are actively deployed in the defense field, the need for the test and evaluation of AI weapon systems is increasing. This paper proposes test and evaluation methods for AI systems. Methods: In this paper, improvements to test and evaluation methods and procedures for AI weapon systems are proposed. In addition, based on case studies, a test and evaluation metric suitable for classification models is suggested. Results: A method for testing and evaluating AI weapon systems by unifying terminology and using the confusion matrix as a performance evaluation method is presented. Conclusion: This paper forms a consensus for the necessity of improving as well as implementing guides to institutionalize AI weapon system test and evaluation.
The ball bearings used to support the shaft of LNG pumps are nonstandard bearings. The bearings operate in the cryogenic environment(-162°C) and they are a type of self-lubricated ball bearing using solid lubricant and have larger internal clearance grade than general standard bearings. The defect in ball bearings is the most common failure cause of LNG pumps. In order to ensure the stable operation of LNG pumps, it is important to optimize the design of bearings with the performance evaluation test and reliability test that ensure fatigue life is also needed. In this paper, the tribo-test apparatus was developed for the performance test and reliability test. This test apparatus is designed to evaluate the performance by measuring friction torque and cage orbit of test bearings. Prior to the perform the ball bearing test, the verification test of the test apparatus was done by operating for about 4 hours with the LN2(-189°C) submerged chamber. The stability of the rotor system(motor to shaft) was monitored with the motor input current. The relation between motor input current and torque was confirmed by the test. After the verification of the test apparatus, the performance test of a ball bearing was performed. As we verified, the rotor system was stable and friction torque and cage whirling orbit was measured properly. From these results, we can expect that this apparatus would be helpful in developing the process of ball bearings used in LNG pumps.
Oil-free turbochargers (TCs) will increase the power and efficiency of internal combustion engines, both sparking ignition and compression ignition, without engine oil lubricant feeding or scheduled maintenance. Using gas foil bearings (GFBs) in passenger vehicle TCs enables compact, lightweight, oil-free systems, along with accurate shaft motion. This paper presents extensive test measurements on GFBs for oil-free TCs, including static load-deflection measurements of test GFBs, rotordynamic performance measurements of a compressed air driven oil-free TC unit supported on test GFBs, and bench test measurements of the oil-free TC driven by a passenger vehicle diesel engine. Two configurations of GFBs, one original and the other modified with three shims, are subjected to a series of experimental tests. For the shimmed GFB, three metal shims are inserted under the bump-strip layers, in contact with the bearing housing. The installation of shims creates mechanical preloads that enhance a hydrodynamic wedge in the assembly radial clearance to generate more film pressure. Simple static load-deflection tests estimate the assembly radial clearance of the shimmed GFB, which is smaller than that of the original GFB. Model predictions agree well with test data. The discrepancy between the model predictions and test data is attributed to fabrication inaccuracy in the top foil and bump strip layers. Test GFBs are installed into a TC test rig driven by compressed air for rotordynamic performance measurements. The test TC rotor, 335 g in weight and 117 mm long, is coated with a commercially available, wear-resistant solid lubricant, Amorphous M, to prevent severe wear during start-up and shutdown in the absence of an air film. A pair of optical proximity probes positioned orthogonally at the compressor end record lateral rotor motions. Rotordynamic test results show that the shimmed GFB significantly diminishes the large amplitude of subsynchronous rotor motions arising in the unmodified GFB. Predicted synchronous rotor amplitudes and rigid body mode natural frequencies agree reasonably well with recorded test data. Finally, the oil-free TC is installed into a passenger vehicle diesel engine test bench. The TC rotor speed is controlled by the vehicle engine. Speed-up tests show dominant synchronous motion (1X) of the rotor. Whirl frequencies of the relatively small subsynchronous motions are associated with the rigid body natural mode of the TC rotor-GFB system as well as (forced) excitation from the four-cylinder diesel engine. The bench test measurements demonstrate a significant reduction in the amplitude of subsynchronous motions for the shimmed GFB, thus verifying the preliminary test results in the TC test rig driven by compressed air.
Engineered design of modern efficient turbomachinery based on accurate model predictions is of importance as operating speed and rate power increase. Industrial applications use hydrodynamic fluid film bearings as rotor support elements due to their advantages over rolling element bearings in operating speed, system stability (rotordynamic and thermal), and maintenance life. Recently, microturbomachinery (< 250 kW) implement gas foil bearings (GFBs) as its rotor supports due to its compact design without lubricant supply systems and enhanced stability characteristics. To meet the needs from manufacturers, the turbomachinery development procedure includes a rotordynamic design and a gas foil journal bearing (GFJB) analysis in general. The present research focuses on the role of gas foil thrust bearings (GFJBs) supporting axial load (static and dynamic) in an oil-free turbo blower with a 75 kW (100 HP) rate power at 30,000 rpm. The turbo blower provides a compressed air with a pressure ratio of 1.6 at a mass flow rate of 0.92 kg/s, using a centrifugal impeller installed at the rotor end. Two GFJBs with a diameter of 66mm and a length of 50 mm and one pair of GFTB with an outer diameter of 144 mm and an inner diameter of 74 mm support the rotor with an axial length of 493 mm and a weight of 12.7 kg. A finite element rotordynamic model prediction using predicted linearized GFJB force coefficients designs the rotor-GFB system with stability at the rotor speed of 30,000 rpm. Model predictions of the GFTB show axial load carrying performance. Experimental tests on the designed turbo blower; however, demonstrate unexpected large amplitudes of subsynchronous rotor lateral motions. Post-inspection reveals minor rubbing signs on the GFJB top foils and significant wear on the GFTB top foil. Therefore, GFTB is redesigned to have the larger outer diameter of 166 mm for the enhanced load capacity, i.e., 145%, increase in its loading area. The modification improves the rotor-GFB system performance with dominant synchronous motions up to the rate speed of 30,000 rpm. In addition, the paper studies the effect of GFTB tilting angles on the system performance. Insertion of shims between the GFTB brackets changes the bearing tilting angles. Model predictions show the decrease in the thrust load capacity by as large as 86% by increase in the tilting angle to 0.0006 rad (0.03438 deg). Experimental test data verify the computational model predictions.
Turbocharger has a main purpose on recycling of the exhaust gas from the engine cylinder. On the basis of the facility characteristics, the turbocharger supported on floating ring bearings has some problems such as the large volume, oil supplement for lubrication and high power loss due to high operating torque. The air foil bearing has been studied as the bearing element to be able to alternate the floating ring bearing without the problems of the floating ring bearing. In this study, the air foil bearing has 2 parts; journal and thrust bearings, and the test facility consists of the engine, exhaust and intake parts. In addiction, the specification of the turbocharger follows a small turbocharger for SUV engine. The engine speed is varied from 750 (idle rpm) to 2,500 rpm and then, the rotating speed of the turbocharger rotor is accelerated from 0 to 100,000 rpm. From those experiments, the comparison between the performances of the air foil bearing and floating ring bearing is conducted and the results show that the air foil bearing has less power loss, maximum 770 watt, than the floating ring bearing, maximum 5,110 watt. This result verifies that the air foil bearing is more efficient and able to output more power under the same condition of the input power.
'%3e%3cg id='b'%3e%3cpath id='a' stroke='%23000' stroke-width='2' d='M-6-25H6m-12 3H6m-12 3H6'/%3e%3cuse xlink:href='%23a' y='44'/%3e%3c/g%3e%3cpath stroke='%23fff' d='M0 17v10'/%3e%3ccircle r='12' fill='%23cd2e3a'/%3e%3cpath fill='%230047a0' d='M0-12A6 6 0 0 0 0 0a6 6 0 0 1 0 12 12 12 0 0 1 0-24z'/%3e%3c/g%3e%3cg transform='rotate(-123.69)'%3e%3cuse xlink:href='%23b'/%3e%3cpath stroke='%23fff' d='M0-23.5v3M0 17v3.5m0 3v3'/%3e%3c/g%3e%3c/svg%3e)