DEVELOPMENT OF COST-EFFECTIVE FAULT LOCALIZATION PLATFORM FOR UNDERGROUND CABLE SYSTEM
Bibliographic record
Abstract
The mining industry is highly dependent on electricity or other forms of energy converted into electricity. Here, one of the common forms of electricity distribution is through underground cables, but the electricity systems regularly confront contingencies due to cable breakdowns or defects. Depending on fault types, detective tools, and cable position, it may take several hours to days to detect the exact fault location with existing methods and tools, which affects the entire mining operation with financial losses and safety issues. Besides, industrial fault localization platforms are very expensive and seldom tailored for the mining industry. To locate such faults, the online fault localization platform has drawn wide attention but is limited in functionalities and applications. This research aims to develop a cost-effective double-ended online fault localization platform while proposing solutions for data synchronization and accurate traveling wave arrival time detection problems. At first, extensive market research has been done on available platforms to access those platforms’ functionalities and costings. Then, a hardware setup has been developed by combining appropriate sensors, a data acquisition unit, a computational platform, and other necessary components. Furthermore, to establish communication between remote platforms, a Python-based software program has been developed. Besides, query-based server management has been introduced to handle and manage huge amounts of data. Combining the hardware and software, the overall cost of a single platform is CAD $ 2,850.00, which is at least ten times less than the least expensive market option. The chosen double-ended traveling wave-based online fault localization method requires accurate synchronized time to properly compute the traveling wave arrival time difference. Thus, coordinated universal time alignment of acquired time series data is needed. Global Positioning System (GPS) based universal time synchronization is one of the popular ways. There are several other techniques, but all of these have some practical difficulties and dependencies. To solve this problem, a cost-effective novel zero-crossing point-based data synchronization approach has been proposed. This approach doesn’t rely on GPS receivers or any other existing methods; rather, the measurement is synchronized by calibrating the zero-crossing points of the sinusoid measurements before the fault. In this way, appropriate synchronization has been realized. The accuracy of traveling wave-based fault localization highly depends on how accurately its arrival times are detected from ends of cable. Accurate detections are achieved when signals sampling rates are high. Usually, a data acquisition system with a higher sampling rate may address the issue, but it increases cost. On the other hand, available interpolation-based up-sampling techniques have several constraints. Machine learning model can solve this problem if trained by real inputs and outputs with desired sampling rates for different types of faults. In this research, a machine learning-based up-sampling model has been presented, which improves the accuracy of traveling wave arrival time detection. Besides, it is cost-effective and outperforms interpolation techniques. Therefore, the proposed fault localization platform combines all the above solutions, which makes the platform very advance, reliable, and cost-effective.
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 | 0.000 | 0.001 |
| Meta-epidemiology (narrow) | 0.001 | 0.000 |
| Meta-epidemiology (broad) | 0.001 | 0.000 |
| Bibliometrics | 0.001 | 0.000 |
| Science and technology studies | 0.000 | 0.000 |
| Scholarly communication | 0.001 | 0.001 |
| Open science | 0.002 | 0.001 |
| Research integrity | 0.000 | 0.001 |
| Insufficient payload (model declined to judge) | 0.007 | 0.002 |
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".