What data for this category looks like
Data for this category comes from raw material batch ledgers, production MES systems, slurry and coating testing equipment, workshop dew point monitoring devices, formation and capacity cabinets, and laboratory testing equipment. Daily documents include batching records, coating and rolling parameter tables, winding/laminating process records, liquid injection ledgers, formation and capacity data, and OCV and safety test reports. Data is updated in real time per battery batch and individual cell, with fields including slurry solid content, coating areal density, rolling gap, dew point temperature, internal resistance value, and more, along with corresponding production batch and detection timestamps.
What constraints these characteristics impose on deployment and upgrade
Data in this category is high-frequency and real-time, with concentrated batch volumes and multi-dimensional process parameters in its format. This imposes multiple constraints on the deployment and upgrade process. High-frequency real-time detection data requires database connections to adapt to high-concurrency read and write operations to avoid question-and-answer response delays. Multi-source heterogeneous process documents must undergo format standardization adaptation in advance to ensure accurate retrieval after access. Electrochemical manufacturing processes have strict requirements for data accuracy. Field verification rules must be configured during deployment to prevent incorrect parameters from entering the knowledge base. During the upgrade process, the storage structure of original production batch data must be retained to avoid interrupting existing process Q&A services.
How to set configurations
| Config Item | Suggested Value | Rationale |
|---|---|---|
DB_MAX_LINK | 200–300 connections | Adapts to high-frequency process data read and write requests generated in real time per battery batch, preventing service interruptions caused by connection pool exhaustion |
MONGODB_URI | Points to a dedicated database instance storing process documents and detection data | Isolates business data from other system data to ensure the accuracy of process Q&A retrieval |
REDIS_URL | Configured as the access address of an internally deployed Redis cache instance | Accelerates high-frequency retrieval of process parameters and reduces access pressure on the primary database |
STORAGE_S3_ENDPOINT | Bound to the access endpoint of the enterprise's internal object storage | Uniformly stores attachment content for process documents and test reports |
ROOT_KEY | Generate a random string that complies with enterprise permission specifications | Ensures the security of access permissions for background configurations |
Three common mistakes
- Phenomenon: A
connection timed outoraccess deniederror is returned when calling the database, and process specification data cannot be loaded. Cause: The intranet or public network IP of the deployment node was not added to the whitelist of the database instance, and the access parameters forMONGODB_URIwere not configured correctly. - Phenomenon: After upgrading to FastGPT v4.17.0, a prompt indicating a missing configuration file appears when starting the service, and the original process Q&A configuration cannot be loaded. Cause: The configuration item migration was not completed according to the upgrade guide, and the assumption that the new version no longer reads the
config.jsonfile at all led to necessary configurations not being transferred to the new configuration system. - Phenomenon: After deployment is complete, the retrieved process parameter results do not match the actual production documents, with missing fields or incorrect units. Cause: Format standardization processing of process documents was not completed in advance, and corresponding data verification rules were not configured, resulting in incomplete parameter information imported into the knowledge base.
How to confirm the configuration is correct
- Run the database connection test script to verify that the configurations for
MONGODB_URIandREDIS_URLcan establish normal connections, and adjust the maximum connection count based on test results. - Upload a real coating and rolling parameter document to the knowledge base, and run a retrieval test to confirm that accurate retrieval results for corresponding process parameters can be obtained.
- Check the service monitoring panel to confirm that the database connection count does not exceed the configured value of
DB_MAX_LINK, and there are no frequent connection timeout errors. - Verify the object storage access configuration to confirm that attachment content for process documents can be uploaded and downloaded normally.
The values provided are common starting points and should be measured against the reader's own samples.