Global ETD Search
Search theses and dissertations gathered from participating repositories worldwide. Every result links back to the library that holds it. No account is needed.
Results
Showing 1 to 20 of 175 for “"model errors"”.
-
Categorization and Mitigation of AI Model Errors
… for the field of deep learning. Today, neural models dominate the vast majority of benchmark leaderboards, get deployed in the real world and even achieve superhuman capabilities for some tasks. However, they are still brittle in many ways, can fail unexpectedly and their reliability and …
-
Quantifying the impacts of resolution-dependent model errors on tropospheric ozone simulation
… variability. Although current Chemical Transport Models (CTMs) are capable of capturing the general characteristics of ozone, there are still biases in the models. One factor that contributes to the biases is the model resolution. This thesis quantifies the impacts of resolution- dependent model …
-
An Offshore Wind Resource Assessment and a Look into Model Errors in Wind Forecasts
… these prolers and comparing them to forecast models such as the Canadian GEM model in Chapter 2. The second part of this thesis will look at offshore wind resource assessment. Wind energy is a clean and viable alternative to burning fossil fuels for energy and is being expanded all over the …
-
Understanding the Impact of Model Errors on the Inverse Modeling of MOPITT CO Observations
… CO is a useful tracer of long-range transport in models. However, estimates of the regional sources of CO are uncertain. Inverse modeling has become a widely used approach for better quantifying the sources, but a fundamental assumption in these inversions, which is typically not valid, is that …
-
Mitigating Model Errors in Chemical Data Assimilation: Application of New Data Assimilation and Machine Learning Approaches
… which relies on imperfect atmospheric models. This thesis explores the potential of mitigating the impact of model errors on chemical data assimilation from three approaches: improving conventional data assimilation systems, reducing model errors, and the use of a new machine learning …
-
Can model errors in atmospheric variables within the low-level inflow region of mesoscale convective systems be used to anticipate displacement errors in initiation location?
… and sustenance of MCSs. Thus, the ability of models to depict specific humidity as well as equivalent potential temperature (𝜃𝑒) flowing into the region where convection initiates is likely to play a role in how accurately they forecast MCSs. The Weather Research and Forecasting (WRF) model …
-
Local ensemble transform Kalman filter with realistic observations
… for assimilating real data. The first issue is model errors. We assimilated observations generated from the NCEP/NCAR reanalysis fields into the SPEEDY model. The performance of the LETKF without accounting for model errors is seriously degraded compared with that in the perfect model scenario. …
-
Ensemble tests of a regional climate model using a perturbed cumulus parameterization scheme
… condition problem into the regional climate model and to extend integration into the seasonal period. The seasonal integration has its own characteristics: not only the assumption of perfect model is no longer proper in the regional climate integration, but also sampling initial conditions is …
-
State Variable and Parameter Estimation
… then developed which effectively compensates for errors in the mathematical description of the process. This adaptive filter finds the mean and covariance of ‘fictitious inputs' and uses these parameters to compensate for the model errors. The results show the application of the above topics to …
-
Measurement covariance-constrained estimation for poorly modeled dynamic systems
… dynamic systems which accounts for system model errors in a much more rigorous manner than Kalman filter-smoother type methods. The Kalman filter-smoother type methods, which currently dominate post-experiment estimation practice, treat model errors via “process noise", which essentially …
-
Tractable Learning and Inference in High-Treewidth Graphical Models
Probabilistic graphical models, by making conditional independence assumptions, can represent complex joint distributions in a factorized form. However, in large problems graphical models often run into two issues. First, in non-treelike graphs, computational issues frustrate exact inference. There …
-
Characterization of orographic cloud and precipitation features over southern Baffin Island and surrounding area
… and accurately represent these features in models. The importance of cold season precipitation to regional moisture cycling and our limited understanding of orographic cloud and precipitation processes in the Arctic provide the motivation for this research. Using high-resolution datasets …
-
Efficient trajectory and policy optimization using dynamics models
… and compact form of prior knowledge is dynamics models; they summarize our knowledge of the robot in the mechanical design and prior interactions with the robot through system identification. Unfortunately, often utilizing dynamics models to their full potential is not straightforward: (1) they …
-
Sensitivity analysis of impedance-based fault location methods
… line impedance values, which arise from circuit model errors, have an adverse impact on fault location estimates of the impedance-based methods. Measurement errors in current and voltage transformers can also lead to inaccuracy in estimation. Further, all methods use simplistic models to …
-
Adjoint based solution and uncertainty quantification techniques for variational inverse problems
… system. Control parameters of the computational model are optimized such that the simulation results fit the physical measurements.The solution procedure is computationally expensive since it involves running the simulation computer model (the emph{forward model}) and the associated emph {adjoint …
-
Verification of the Rapid Refresh (RAP) and High-Resolution Rapid Refresh (HRRR) Models within Landfalling Tropical Cyclones toward the Improvement of Rainband Tornado Forecasting
… use high-resolution, rapidly-updating models such as the Rapid Refresh (RAP) and High-resolution Rapid Refresh (HRRR), and an ingredients-based approach similar to that used for forecasting continental mid-latitude tornadoes. Though RAP and HRRR model errors have been identified in …
-
A Probabilistic Deformation Demand Model and Fragility Estimates for Asymmetric Offshore Jacket Platforms
… the development of a probabilistic demand model that accounts for the relevant uncertainties and model errors. In this study, a probabilistic demand model is developed to assess the deformation demand on asymmetric offshore jacket platforms subject to wave and current loadings. The …
-
Mitigating Model Error in CO Emission Estimation
… quantify CO emissions using a top-down or inversemodeling approach. Recent studies have shown, however, that there are significant uncertaintiesin emissions estimated using this approach, due to the assumption that thechemical transport model used in the inversion is perfect or unbiased.In this …
-
Characterizing the Adventitious Model Error as a Random Effect in Item-Response-Theory Models
… are usually concerned about two types of errors: sampling error and model error. The sampling error is caused by the discrepancy between the observed sample and the population from which the sample is drawn from (i.e., operational population). The model error refers to the discrepancy …
Page 1 of 9