Engineering Papers⌕ Search

SEARCH · Engineering Papers

Results for “CONDITIONING (LEARNING)”

Search indexed NASA NTRS and DOE OSTI research on propulsion, heat transfer, battery materials and energy systems. Follow report and document links to the original sources.

Quote a phrase for an exact phrase match. Source license links do not imply unrestricted reuse.

At least 19 records
All sourcesNASA NTRSDOE OSTI
NASA NTRS2001

Training Data Optimized and Conditioned to Learn Characteristic Patterns of Vibrating Blisks and Fan Blades

At the NASA Glenn Research Center, we have been training artificial neural networks to interpret the characteristic patterns (see the leftmost image) generated from electronic holograms of vibrating structures. These patterns not only visualize the vibration properties of structures, but small changes in the patterns can indicate structural changes, cracking, or damage. Neural networks detect these small changes well. Our objective has been to adapt the neural-network, electronic-holography combination for inspecting components in Glenn's Spin Rig.

Decker, Arthur J.↗
NASA NTRS2004

Analysis of Droplet Size during the Ice Accumulation Phase Of Flight Testing

There are numerous hazards associated with air travel. One of the most serious dangers to the pilot and passengers safety is the result of flying into conditions which are conducive to the formation of ice on the surface of an aircraft. Being a pilot myself I am very aware of the dangers that Icing can pose and the effects it can have on an airplane. A couple of the missions of the Icing branch is to make flying safer with more research to increase our knowledge of how ice effects the aerodynamics of an airfoil, and to increase are knowledge of the weather for better forecasting. The Icing Branch uses three different tools to determine the aerodynamic affects that icing has on a wing. The Icing research tunnel is an efficient way to test various airfoils in a controlled setting. To make sure the data received from the wind tunnel is accurate the Icing branch conducts real flight tests with the DHC-6 Twin Otter. This makes sure that the methods used in the wind tunnel accurately model what happens on the actual aircraft. These two tools are also compared to the LEWICE code which is a program that models the ice shape that would be formed on an airfoil in the particular weather conditions that are input by the user. One benefit of LEWICE is that it is a lot cheaper to run than the wind tunnel or flight tests which make it a nice tool for engineers designing aircraft that don t have the money to spend on icing research. Using all three of these tools is a way to cross check the data received from one and check it against the other two. industries, but it is also looked at by weather analysts who are trying to improve forecasting methods. The best way to avoid the troubles of icing encounters is to not go into it in the first place. By looking over the flight data the analyst can determine which conditions will most likely lead to an icing encounter and then this information will aid forecasters when briefing the pilots on the weather conditions. am looking at the weather data from certain flights and analyzing the type of precipitation that the plane is flying through. During flight tests there is a probe on the bottom of the aircraft that gathers information on the size and shape of the particles that it is flying through. The data can then be viewed on a computer. After grouping the weather into certain groups we can then pick certain groups which we think should be analyzed farther. The goal is to remove all the ice particles because they do not contribute to the icing on an aircraft. We use a 2D analyzer which measures the droplet size and categorizes the drops into bins of certain sizes. We can then look at what the characteristics of the weather that we were flying through such as the temperature and dew point and compare this with the size of the drops that the 2D analyzer measured. We can then look at what type and shape of ice that formed on the wing during this time period. Having this data will help us to reproduce these conditions using LEWICE and the wind tunnel. Having consistency among the tests will make things more accurate. With respect to weather forecasting we will be able to learn which conditions can lead to icing. Better accuracy in weather reporting will lead to fewer run-ins with icing which will also lead to fewer accidents.

Miller, Eric James↗
NASA NTRSDate not supplied

Sampling Functions from Gaussian Processes and Structured Covariance Gaussian Networks

When learning aerodynamic models from data, it is critical to incorporate estimates of model uncertainty. This motivates the design of probabilistic aerodynamic databases which can be sampled to generate physically and statistically plausible aerodynamic models. In this talk we discuss how to sample deterministic functions from two different kinds of probabilistic models and demonstrate their use. First, Gaussian Process Regressors (GPRs) are a widely used probabilistic kernel-based model which can be thought of as Gaussian distributions over functions. GPRs are generally trained by maximizing the marginal likelihood of seeing the training data over the kernel parameter space. Sample functions are easily generated by drawing points from the Gaussian distribution at desired input points. However, when the points are not known ahead of time, the classical sampling approach is not possible since successive function samples will generate different function realizations. We present an approach for sampling consistent function evaluations from a GPR over multiple samples. Second, we describe a neural network architecture which learns a conditional Gaussian distribution by maximizing the marginal likelihood at each point in the input space. We then discuss and compare several options for generating sample functions which match this distribution. Finally, we demonstrate the use of these probabilistic aerodynamic models in an atmospheric reentry simulation.

Gaussian process regression↗
NASA NTRS1969

Periodic shock with added clock.

Periodic shock in rats maintained on variable-interval schedule of reinforcement interspersed with sequence of three different stimulus conditions

CONDITIONING (LEARNING)↗
NASA NTRS1995

Constraint-Based Scheduling System

Report describes continuing development of software for constraint-based scheduling system implemented eventually on massively parallel computer. Based on machine learning as means of improving scheduling. Designed to learn when to change search strategy by analyzing search progress and learning general conditions under which resource bottleneck occurs.

Zweben, Monte↗
NASA NTRS2018

GEONEX: Progressive Conditional Generative Adversarial Training Using Transfer learning

Obtaining accurate segmentation on large scale images is an open problem in deep learning. The main problem is the amount of labeled data that exists for large scale images. Traditionally, the common solution to this problem is to crop the large images into smaller images to increase the amount of available data and train a Conditional Generative Adversarial Network (CGAN). CGANs are currently the state of the art in image to image translation and provide better accuracy than the traditional method of training a encoder based conv-net architecture to minimize the loss at each pixel. This method can produce noisy and discontinuous images with inaccurate results. We seek to solve this problem by utilizing the concepts of transfer learning and progressive training to create a CGAN that can segment large scale images with a limited amount of labeled data. In transfer learning we recognize that many learned features are applicable to many classes from multiple domains. This introduces the concept of feature reusability, which is the basis for finetuning. Progressive training got its start in training models on the same images at different resolutions. In this work we instead train a GAN on increasing image scales by transferring the weights from the smaller scales to the larger scales. The learned features at the smaller scales are continually reused and applied to larger scales to create a CGAN that can perform accurate segmentation on large scale images. We apply this method to detect building footprints on very high-resolution overhead imagery (e.g Digital Globe and high resolution airborne platforms).

GEONEX↗
NASA NTRS1991

Rhesus monkey (Macaca mulatta) complex learning skills reassessed

An automated computerized testing facility is employed to study basic learning and transfer in rhesus monkeys including discrimination learning set and mediational learning. The data show higher performance levels than those predicted from other tests that involved compromised learning with analogous conditions. Advanced transfer-index ratios and positive transfer of learning are identified, and indications of mediational learning strategies are noted. It is suggested that these data are evidence of the effectiveness of the present experimental apparatus for enhancing learning in nonhuman primates.

Washburn, David A.↗