What the data for this category looks like
Documents and materials for new energy development mainly come from project approval documents, environmental impact assessment reports, grid connection application materials, construction logs, acceptance documents, and similar sources. Updates follow project phase cycles, with dedicated documents generated for each phase from project initiation to completion. Most documents have a fixed hierarchical chapter structure, including fields such as project number, installed capacity, and land area. Some documents are official PDFs with digital signatures or scanned copies containing complex equipment list tables. A single document can be tens of pages long.
What constraints these characteristics impose on the document parsing and chunking step
Documents include PDFs with digital signatures and scanned copies, so parsing tools must support content extraction from non-standard PDF formats. Single documents are long and contain structured technical parameters, so chunking must avoid breaking field relationships and chapter integrity. Complex tables and fixed field units such as MW and hectares require preserved table structure and consistent field units after parsing. Phase-based updated documents need to support incremental parsing and chunking to avoid reprocessing already parsed content.
Configuration Settings
| Configuration Item | Recommended Value | Rationale for This Value |
|---|---|---|
chunkSize | 800–1200 characters | New energy development documents include multiple sets of technical parameters and structured chapters. This range balances chunk integrity and retrieval granularity, as calibrated through actual testing |
chunkSplitter | Split by title + paragraph | Documents have a fixed hierarchical chapter structure. Splitting by title preserves the logical integrity of chapters and adapts to the hierarchical classification needs of project documents |
customPdfParse | Enabled | Official new energy documents often use PDFs with digital signatures and scanned formats. Enabling this option optimizes metadata and text extraction effects for non-standard PDFs |
UPLOAD_FILE_MAX_SIZE | 1000 MB | Single large environmental impact assessment or grid connection application documents can reach hundreds of MB. This configuration adapts to large file upload and parsing requirements |
PARSE_FILE_TIMEOUT_SECONDS | 600 seconds | Long documents take longer to parse. This duration covers the complete parsing process for most large project documents |
chunkSettingMode | Manual configuration | Different types of new energy documents (such as environmental impact assessment reports, construction logs) have varying chunking requirements. Manual configuration allows flexible parameter adjustment |
Three Common Mistakes
- Phenomenon: After uploading a signed new energy project PDF, the parsing result has no valid text content. Cause: The
customPdfParseconfiguration is not enabled. The default parsing logic cannot recognize encrypted metadata in signed PDFs. - Phenomenon: Calling the Doc2x tool returns a file read error, with an error prompt indicating only specified file formats are supported. Cause: The
UPLOAD_FILE_MAX_SIZEconfiguration is not set to adapt to large files, or the uploaded file was not converted to a standard parsable format. - Phenomenon: Parsed chunk results lose individual technical parameters (such as installed capacity in MW). Cause: The
chunkSizesetting is too small, splitting complete paragraphs containing a single parameter into different chunks and breaking field relationships.
How to Verify Proper Configuration
For FastGPT v4.17.0, the configuration effect can be verified through the following actions:
- Upload a signed new energy project PDF document, and check if the parsing result contains complete text and field content.
- Upload a project environmental impact assessment report larger than 500 MB, and confirm the upload and parsing process does not trigger file size limit or timeout errors.
- Adjust the
chunkSizeparameter, and check if the chunk results retain complete chapter titles and technical parameter paragraphs. - Test calling the Doc2x tool, and confirm it can normally read the uploaded new energy document files without format restriction errors.
The values provided are common starting points and should be measured against the reader's own samples.
Question material comes from public community discussions. Configuration values are common starting points and should be measured against your own samples. Verified on 2026-09-28.