Improved Pearson Correlation Coefficient-Based Graph Neural Network for Dynamic Soft Sensor of Polypropylene Industries
Bibliographic record
Abstract
Polypropylene is an important product in the chemical industry and also a raw material for packaging bags, masks, and building boards. The melt index (MI) is a key indicator for evaluating the quality and efficiency of the polypropylene production process. Accurate measurement of the MI is beneficial to increase the polypropylene yield and save energy. The polypropylene production process is characterized by strong nonlinearity, obvious dynamic features, and complex structure, so the current soft sensor methods cannot carry out real-time and accurate soft sensor of the MI. In order to fully mine the complex relationship between variables of temporal data and extract the characteristics of time-series space in chemical production processes, this paper proposes a novel dynamic soft sensor method using an improved Pearson correlation coefficient-based graph neural network (GNN) (Pearson-GNN) method. The adjacency matrix is updated through the correlation coefficient of the data, which is integrated into the graph convolution and time sequence convolution modules of GNN to improve the accuracy of the soft sensing. Finally, the performance of the proposed Pearson-GNN is verified in time-series data soft-sensing task on the public air quality data set and actual polypropylene production processes. Compared with the diffusion concurrent recurrent neural networks (DCRNN), the multivariate time-series forecasting with graph neural networks (MTGNN), the spatiotemporal graph convolutional networks (STGCN), and the GNN based on a fully dynamic adjacency matrix without Pearson correlation updates (Fully GNN), the experimental results show that proposed Pearson-GNN is superior to other methods in terms of the mean absolute percentage error, the root-mean-square error, and mean absolute error.
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 | 0.002 |
| Meta-epidemiology (narrow) | 0.001 | 0.000 |
| Meta-epidemiology (broad) | 0.000 | 0.001 |
| Bibliometrics | 0.001 | 0.001 |
| 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".