What this category's data includes
Core data is generated for each individual power battery pack, energy storage cabinet, or BMS system. Daily documents cover grouping rules, BOM lists, welding and torque records, BMS flashing logs, insulation and voltage withstand reports, EOL test reports, and system configuration lists. Data sources include cell capacity grading databases, production MES systems, welding and torque device logs, BMS software version records, and EOL test bench data. Data is generated once per completed PACK or energy storage cabinet, and updated synchronously each time the BMS software version changes. Document fields include cell grade, welding torque value, insulation resistance value, BMS flashing version number, and other fields. Some documents must match the parameter standards of the corresponding production batch.
Constraints on tool calling and plugins workflows
Single-batch data for this category has strong correlations and frequent updates. Tool calling must accurately match the production documents of individual equipment. This requires limiting the tool calling context range to only the process specifications of the current production batch. Data sources are scattered across multiple systems. Tool calling must support pulling parameters across data sources, to avoid call failures caused by reliance on a single interface. Document fields for individual equipment follow fixed formats. Tool calling output must strictly match field units. For example, torque values must retain two decimal places, and insulation resistance must use megohm units. Additionally, process specifications and technical standards are updated frequently. Tool calling must link to the latest version of documents in real time, to avoid calling outdated standard content.
How to set configurations
| Configuration Item | Recommended Value | Rationale |
|---|---|---|
| Workflow/API tool selection | auto | Automatically select process documents and equipment data matching the current production batch |
| Single tool call timeout | 120 seconds | Aligns with typical response times for production device logs and MES data |
| Tool call context scope | Only link to documents from the current production batch | Restrict data scope to avoid redundant information interfering with answer accuracy |
| Output format validation | Enforce matching preset field units | Ensure returned torque, insulation resistance, and other data comply with process standards |
| Maximum retry attempts | 3 | Balance call success rate and wait time, suitable for multi-data-source pull scenarios |
Three common errors
- Tool calls return empty responses, with no corresponding process data pulled. The cause is failure to configure document association rules for the current production batch. Tool calls do not limit the context range, so redundant or mismatched documents are retrieved.
- Tool calls return an Invalid JSON error, indicating invalid control characters are present. The cause is failure to enable output format validation. Device logs returned by the tool contain line breaks or invisible characters that are not automatically cleaned.
- Tool call nodes cannot add connection lines, and no new connection circles appear in the interface. The cause is failure to upgrade to FastGPT v4.17.0. Older versions have compatibility issues with the tool node connection function.
How to confirm proper configuration
- Initiate a Q&A for the process specifications of a single PACK, and check if the data returned by the tool call includes fields such as the current batch's cell grade and welding torque value.
- After triggering a tool call, view the call log to confirm the timeout setting matches the response time of actual production data.
- Test the tool call output format, and verify that returned values match preset unit requirements.
- Check if the tool node connection function works normally, and confirm the currently running FastGPT version meets functional requirements.
The values listed here are common starting points, and should be evaluated against the reader's own samples.