What this category of data looks like
Data for this category comes from electromagnetic design documents, copper wire and insulation material batch ledgers, winding equipment operation logs, dynamic balancing machine test data, electrical test bench records, and production work orders from ERP/MES systems. Daily documents include winding process cards, insulation treatment records, dynamic balance reports, withstand voltage insulation test records, and type and factory test reports. Fields include number of turns, wire diameter, insulation class, unbalance amount, withstand voltage value, temperature rise parameters, and others. The update frequency is after each product or batch production, and syncs whenever a product type is changed.
Constraints for Source Citation and Traceability
Data for this category is scattered across device logs, system ledgers, and multiple types of specialized documents. Traceability links require precise cross-data-source association. The update frequency per unit or batch requires traceability matching to bind to specific production batches and corresponding process records. Specific fields and units in specialized documents require citation recognition to match dedicated parameters such as winding turns, unbalance amount, and withstand voltage value. The business characteristics of multiple constraints require traceability to cover full-process data for winding, insulation, and testing, to avoid one-sidedness from a single source.
Configuration Settings
This document is compatible with FastGPT v4.17.0.
| Configuration Item | Recommended Setting | Rationale |
|---|---|---|
quoteMaxToken | 800–1200 token | Matches the content length of long documents such as winding process cards and test reports, to avoid truncating critical parameters |
responseData | Include complete citation sources | Meets traceability requirements for multi-stage inspection records |
detail | Display original fields | Facilitates verification of specialized parameters such as winding turns and withstand voltage values |
| Single recall limit | Top 5 entries | Covers associated documents for the three core processes of winding, insulation, and testing |
| Context splicing length | Calibrated via actual testing | Adapts to total length limits for multi-source data |
Three Common Configuration Errors
- Phenomenon: After calling the API, the citation list is returned first, and the answer body is not displayed directly. Cause: The return priority configuration of
responseDatawas not adjusted. The default setting prioritizes loading citation source data. - Phenomenon: After associating a local knowledge base, the citation list displays correctly but the answer body does not include knowledge base content. Cause: The body embedding configuration for the
detailparameter was not enabled. Only source data was recalled but not spliced into the answer. - Phenomenon: After local deployment, knowledge base content cannot be called, and the first entry in the citation list matches correctly. Cause: A reasonable context splicing length was not configured, causing knowledge base content to exceed the truncation threshold and not be included.
How to Confirm Configuration is Successful
- Initiate a query about winding turns, and check if the citation source of the corresponding process card is included in the returned results.
- Review the structure of the response data, and confirm that the association logic between citation sources and the answer body meets configuration requirements.
- Simulate an inspection and audit query for a single electric motor, and verify that the recalled documents cover the winding, insulation, and testing links.
- Check if the parameters returned by the API match the preset
quoteMaxTokenvalue, and confirm that they meet business needs.
The values provided are common starting points and should be measured against the reader's own samples.