Fault diagnosis with high accuracy and timeliness for semi‐batch crystallization process based on deep learning with multiple pattern representation
Bibliographic record
Abstract
Abstract The research on chemical process fault diagnosis has made significant progress, but there is still a big gap in its application to complex practical industrial processes. As for the fault diagnosis of batch crystallization processes, the recently‐proposed dynamic time warping–convolutional neural network (DTW‐CNN) model has achieved a great improvement in the fault diagnosis. However, its fault diagnosis rate (FDR) and timeliness of fault diagnosis are still low, and thus, it needs to improve further before being applied to the practical application. In this paper, a multiple pattern representation–convolutional neural network (MPR‐CNN) model is proposed and applied for the fault diagnosis of a semi‐batch crystallization process. The MPR‐CNN model enables the manual extraction of features with four pattern representation algorithms in the data pre‐processing stage, and generates a three‐dimensional matrix which is used as the training sample and input to the CNN for the formal feature extraction and weight learning. An excellent classification performance, with an average FDR of 97.5%, is achieved. This model is also applied for the fault diagnosis of process data within a shorter period of time after the occurrence of faults. The results indicate that the model could make timely fault diagnosis with a highly stable and accurate performance after the occurrence of a fault.
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 | 0.001 | 0.001 |
| Meta-epidemiology (narrow) | 0.001 | 0.000 |
| Meta-epidemiology (broad) | 0.001 | 0.001 |
| Bibliometrics | 0.001 | 0.000 |
| Science and technology studies | 0.000 | 0.000 |
| Scholarly communication | 0.001 | 0.001 |
| Open science | 0.001 | 0.001 |
| Research integrity | 0.001 | 0.001 |
| Insufficient payload (model declined to judge) | 0.001 | 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 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".