Machine learning techniques for predictive modeling and testing— with applications in power systems
Bibliographic record
Abstract
This thesis addresses several research topics in the field of system identification. The first theme is related to nonparametric specification testing applied to nonlinear block-oriented models. It is the common approach in system identification to employ parametric models and to design the corresponding parametric system identification algorithm. However, the question arises how well the assumed parametric structure represents the underlying characteristics of the model. The nonparametric specification testing method is developed for this purpose, where the nonparametric regression estimation theory is utilized. The proposed methodology is demonstrated in the context of the Hammerstein block-oriented structure. Analogous nonparametric testing techniques can be also extended to other block-oriented systems. The second task of this thesis is concerned with the problem of selecting smoothing parameters in nonparametric kernel regression estimators applied to identification of the Hammerstein system. Commonly, some form of traditional cross-validation technique has been applied in this context. The correlation nature of the output signal of the Hammerstein system makes this classical cross-validation methods sub-optimal. In the thesis, several re-sampling alternatives are proposed that reflect the statistical dependence of the observed data. The third goal of this thesis is devoted to adapting the high-dimensional machine learning methods for stability analysis in large-scale power systems. The transient stability problem is characterized by high-dimensional features in the given power system. The previous studies on this problem have applied classical linear regression analysis. In the thesis, this is extended to nonlinear regression algorithms capable of choosing lower-dimensional solutions. The developed methodology is based on sparse machine learning techniques that play crucial role in the modern statistical learning. The resulting models for predicting transient stability achieve the superior performance compared to the known solutions in the field. The final topic of the thesis is about parametric testing of load models appearing in dynamic security assessment of power systems. Using the formal statistical techniques, it is shown which load model out of possible alternatives should be selected. The real data load testing problem is examined using observations from Manitoba Hydro radial load system. The presented studies confirm the importance of the frequency component in load modeling.
Fetched live from OpenAlex and de-inverted. Abstracts are not stored in this database: the inverted indexes are 8.6 GB of the frame’s 9.3 GB of text, and the host has 13 GB free.
How this classification was reachedexpand
Full frame machine prediction
Teacher imitationNot calibrated prevalence, not ground truth. Human validation pending. The Gemma side is a direct model label for every work in the frame, read from the title-only record. The Codex side is a classifier learned from the 10,348 direct Codex labels and calibrated to design-weighted sample rates; fields without enough sample support carry no Codex call. Candidate is the union of the two sides; consensus is their intersection. These outputs are machine_predicted_unvalidated and are not human labels.
Distilled classifier scores by category (both heads)
| Category | Codex | Gemma |
|---|---|---|
| Metaresearch | 0.004 | 0.012 |
| Meta-epidemiology (narrow) | 0.001 | 0.001 |
| Meta-epidemiology (broad) | 0.002 | 0.001 |
| Bibliometrics | 0.002 | 0.004 |
| Science and technology studies | 0.000 | 0.002 |
| Scholarly communication | 0.002 | 0.002 |
| Open science | 0.002 | 0.002 |
| Research integrity | 0.002 | 0.004 |
| Insufficient payload (model declined to judge) | 0.004 | 0.001 |
Machine scores (provisional)
The two teacher heads of the student model, read on this work. A score orders the frame for review; it never asserts a category, and the validation status ships verbatim with every row.
Baseline scores from an immature model (maturity gate not passed, 7 training rounds). Scores rank; they never assert a category.
score_only:v0-immature-baseline · verbatim from the scoring run: score_only means the number may rank works, and no category label ships from itClassification
machine, unvalidatedMachine predicted; a candidate call from one source (direct Gemma or distilled Codex), not a consensus.
How this classification was reached, model by model and score by score, is at the end of the page under "How this classification was reached".