Data profile for this category
Core data for this category originates from process, equipment, quality control, and production departments of crystalline silicon cell manufacturers. Daily documents include process cards, diffusion and coating parameters, printing and sintering records, EL inspection images, IV grading data, and yield reports. Data sources cover silicon wafer incoming materials, MES systems, production equipment such as diffusion furnaces, EL and IV testing equipment, and laboratory data. Updates occur per production batch, per individual inspection step, and daily summaries. Documents are organized by production process nodes, and include fields such as specific parameters, inspection results, and grading data.
Constraints imposed on workflow orchestration
High-frequency production data updates require workflows to support triggering by batch or individual inspection node, to avoid using expired historical slices. Mixed input of multiple document types requires configuring nodes adapted to different document structures, separating parameter documents from visual inspection images. Multi-source data from cross-departmental teams must be aggregated via a unified variables node to ensure review calls use only associated documents from the same production batch. Production process continuity requires workflows to execute strictly in the order of production nodes, preventing cross-batch data mixing.
Configuration settings
This guide is written for FastGPT v4.17.0.
| Configuration Item | Recommended Setting | Rationale |
|---|---|---|
variables | Bind three core fields: batch number, diffusion sheet resistance, and EL crack rate | Photovoltaic cell production uses batches as the core management unit. Binding core fields enables accurate matching of associated documents from the same batch |
| Original file preview association | Bind the storage path of the original file before chunking | Direct access to the original document before importing into the knowledge base is required to ensure complete original data traceability during cross-review |
nodeId | Assign unique identifiers according to the three processes of diffusion, coating, and printing | Distinguish review nodes for different production processes to avoid confusion of process data during workflow execution |
| HTTP request timeout | 600 seconds | The average response time for production equipment to return inspection data is 300-500 seconds, with reasonable buffer time reserved |
| File link variable | Associate the storage paths of EL inspection images and IV grading data | Visual inspection data from production links requires quick preview to improve cross-review efficiency |
Three common configuration errors
- The workflow displays "file preview failed" during runtime, and the original file before chunking cannot be viewed. Cause: The binding path for original file preview association is not configured, and only chunked fragment data from the knowledge base is called.
- The workflow returns "third-party node call failed" during testing. Cause: The binding fields of the
variablesnode are not configured correctly, resulting in parameters that do not meet the node's verification rules. - The file link variable in the workflow returns a null value. Cause: The storage paths of EL inspection images and IV grading data are not correctly bound to the variable node, causing the workflow to fail to obtain the corresponding file paths.
How to confirm configuration is complete
- Trigger the workflow and pass in production data for a specified batch, then check whether the
variablesnode correctly binds the batch number, diffusion sheet resistance, and EL crack rate fields. - Click the file link variable in the workflow to confirm you can jump to the original storage path of the EL inspection image or IV grading data.
- Run the HTTP request node, confirm the response time meets the preset buffer requirements, and that returned data conforms to production document format specifications.
- View the workflow's node logs, confirm
nodeIdis correctly assigned per the diffusion, coating, and printing processes, with no node confusion occurring.
The values provided here are common starting points and should be measured against the reader's own samples.