A PINN-based digital parameter identification of the pipe conveying fluidsystems
Bibliographic record
Abstract
Digital twins are being developed for the monitoring, maintenance planning and operation optimization of industrial systems such as hydraulic infrastructures.Most hydraulic turbines are equipped with many different sensors, such as accelerometers and pressure sensors, to monitor the health condition of the equipment.Machine learning method has been more widely used with the increase of data availability and requirement.However, in most cases, adequate data are not always available, while the limited available data are often noisy and scattered.This study introduces a new method based on physics-informed neural networks (PINN), which takes advantage of the prior physical knowledge of the system in addition to the sensor data and compensates for the sparse data condition to perform parameter identification.A pipe conveying fluid system is considered as the case study to develop a digital twin for, because it is simple to build and exhibits a diverse range of complex behaviors which can be analogous to this of a hydraulic turbine, such as water-added mass, flutter, flow-induced excitation or hydrodynamic damping.Both experimental data and prior knowledge, such as partial differential equations, are used during the training to optimize the weights and biases of the PINN model.Physical prior knowledge is softly encoded by penalizing the residuals of the partial differential equation and the initial and boundary conditions in the loss function.The loss is also composed of the experimental data and is minimized during the training to fit the data as well as the physical model.The physical part of PINN significantly reduces the need for data, limits the impact of over-fitting, and allows for better prediction of the deflection of the pipe conveying system.This method can also solve the inverse problem by optimizing parameters in the partial differential equation, such as the flow rate in the pipe conveying system.In conclusion, the developed PINN-based digital twin not only provides information on the system over time but also infers its hidden characteristics and detects different phenomena.
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 distilled prediction
Teacher imitationNot calibrated prevalence, not ground truth. Human validation pending. Learned from the 10,348 direct Codex labels and 10,348 direct Gemma labels. Candidate is the union of thresholded teacher heads; consensus is their intersection. These outputs are machine_predicted_unvalidated and are not human labels or direct frontier model labels.
Codex and Gemma teacher scores by category
| Category | Codex | Gemma |
|---|---|---|
| Metaresearch | 0.000 | 0.000 |
| Meta-epidemiology (narrow) | 0.000 | 0.000 |
| Meta-epidemiology (broad) | 0.000 | 0.000 |
| Bibliometrics | 0.000 | 0.000 |
| Science and technology studies | 0.000 | 0.000 |
| Scholarly communication | 0.000 | 0.000 |
| Open science | 0.000 | 0.000 |
| Research integrity | 0.000 | 0.000 |
| Insufficient payload (model declined to judge) | 0.000 | 0.000 |
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 teacher head, 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".