What the data for this category looks like
Data for this category comes from the hot metal and scrap steel entry system, L2 process control terminals, laboratory testing systems, manufacturing execution systems, energy medium monitoring platforms, and equipment status collection nodes. Daily documents include heat plans, batching lists, converter smelting records, composition test reports, continuous casting process parameter records, and slab quality judgment sheets. Data update cycles cover per-heat generation, real-time synchronization, and shift-based archiving. Fields include process parameters with units such as smelting temperature, casting speed, and composition content. Most document structures are structured forms and time-series logs.
Constraints on Deployment and Upgrade Posed by These Characteristics
The data sources for this category are scattered, and update cycles are frequent. During deployment, permissions for connecting multiple data sources must be configured to avoid missing any process parameter collection links. Real-time and per-heat level data updates require adjusting synchronization trigger logic during deployment to adapt to short-cycle data flow. Structured heat and continuous casting document structures require knowledge base chunking to match the business granularity of casting sequences or heats, to avoid mixing cross-batch data. Since production processes cannot be stopped arbitrarily, use a gray release strategy during the upgrade phase to switch knowledge base versions in batches, preventing abnormal impacts on continuous production.
Configuration Settings
The following configuration items are adapted for FastGPT v4.17.0.
| Configuration Item | Recommended Value | Rationale |
|---|---|---|
DB_MAX_LINK | 15–25 concurrent connections | Adapts to high-frequency per-heat data writes and queries, avoiding database connection exhaustion that disrupts production data synchronization |
MONGODB_URI | Cluster read-write separation connection string | Meets the process document query needs of multiple departments, ensuring stable reading and archiving of real-time data |
REDIS_URL | Enterprise intranet Redis cluster address, connection timeout set to 600 seconds | Caches frequently accessed composition reports and continuous casting process records, reducing query pressure on the core database |
STORAGE_S3_ENDPOINT | Enterprise internal object storage node, single file limit set to 1000 MB | Stores large-volume smelting logs and slab judgment sheets, adapting to batch document upload and retrieval |
FE_DOMAIN | Production intranet exclusive domain name | Restricts the Q&A interface to only be accessible within the enterprise production intranet, complying with industrial production security specifications |
FILE_TOKEN_KEY | Random 32-character string | Used to generate temporary access keys for documents, ensuring access security for smelting records |
Three Common Misconfigurations
- A workflow code running component returns status code 500 with a connection timeout prompt. Cause: The intranet access permission for
REDIS_URLis not configured, causing cached process data to be unavailable to the code component. - The latest FastGPT version update package cannot be pulled after deployment. Cause:
FE_DOMAINis configured as an intranet domain name, and no external network access permission is enabled, preventing pulling from the official update source. - The number of knowledge base retrieval results is insufficient to cover the process parameters for the current heat. Cause:
DB_MAX_LINKis set too low, and database connection exhaustion causes some data to fail to be retrieved.
How to Verify Successful Configuration
- Log in to the FastGPT backend database monitoring panel, check the current concurrent connection count, and confirm it does not reach the threshold set by
DB_MAX_LINK. - Upload a converter smelting record document, verify that it can be correctly indexed, and retrieve the corresponding composition parameters and casting speed through the Q&A interface.
- Attempt to access the Q&A interface within the production intranet, confirm that the
FE_DOMAINconfiguration takes effect, and that direct external network access is not possible. - Trigger a simulated heat data synchronization, verify that process data can be written to the knowledge base in real time and indexed completely.
The values provided here are common starting points and should be measured against the reader's own samples.