run-fluent-autoclave
Run, diagnose, reproduce, and validate ANSYS Fluent CFD simulations of forced-convection autoclaves using Fluent MCP or PyFluent. Use for autoclave or hot-air vessel projects involving SCDOC/STEP geometry, annular velocity inlets, pressure outlets, Spalart-Allmaras turbulence, energy transport, calorimeter temperature boundaries, automatic meshing, longitudinal velocity contours, streamlines, mass-balance checks, or the Bohne paper boundary-condition case.
DeepseekModel
官方收录技能
质量 优秀 · 90
v1.0.0
获取
https://deepseekmodel.com/api/download.php?id=cai-aa-cae-agent-hub-skill-ansys-run-fluent-autoclave-skill-md&format=skill
下载 .skill
标准格式,含 system_prompt 与 model_config,导入任意 Agent 框架即可使用
.skill 文件中 system_prompt 字段的实际内容。
name run-fluent-autoclave description Run, diagnose, reproduce, and validate ANSYS Fluent CFD simulations of forced-convection autoclaves using Fluent MCP or PyFluent. Use for autoclave or hot-air vessel projects involving SCDOC/STEP geometry, annular velocity inlets, pressure outlets, Spalart-Allmaras turbulence, energy transport, calorimeter temperature boundaries, automatic meshing, longitudinal velocity contours, streamlines, mass-balance checks, or the Bohne paper boundary-condition case. Run Fluent Autoclave Reproduce the validated open autoclave workflow while preserving geometry-specific judgment and numerical evidence. Workflow Detect Fluent before launching it. Confirm fluent.exe , PyFluent, version, and job directory. Inspect the geometry and existing project files. Reuse a valid STEP/fluid domain instead of rebuilding geometry. Read references/paper-case.md before applying the Bohne-style boundary conditions. Read references/mcp-workflow.md before controlling Fluent MCP or handling a Chinese path. Copy the bundled scripts into the job directory and adjust only the geometry-dependent constants. Generate and check the mesh. Reject unmatched boundary faces, non-manifold faces, invalid volumes, or a misidentified full-face inlet. Configure Fluent in small validated chunks. Print each model, material, and boundary state after setting it. Run one time step first. Continue only if the mesh is valid and residuals remain finite. Run the remaining steps asynchronously when possible, monitor stdout/stderr, and save case/data before post-processing. Extract mass flow, inlet/outlet average and maximum speeds, global maximum speed and location, pressure drop, temperature range, and the final residuals. Generate the vertical longitudinal mid-plane velocity contour and streamlines with outlet left and inlet/head right. Run scripts/validate_results.py on the final JSON. Do not call the result complete if conservation or provenance checks fail. Required physical setup Use the annular clearance at the ellipsoidal head as the velocity inlet; never use the whole end face without verifying the geometry. Use a pressure outlet at the opposite duct. Keep vessel walls adiabatic and no-slip. Split exposed calorimeter faces into a separate fixed-temperature wall zone. Use the material properties and boundary values in references/paper-case.md when reproducing that case. Treat the reference image peak velocity as an outcome, not a prescribed outlet value. Reliability rules Preserve prior cases and results; write the new run under distinct names. Prefer named zones over raw zone IDs after the mesh is loaded. Verify the global maximum location. A maximum at the outlet is suspicious unless the geometry physically contracts there. Report local outlet reverse-flow area separately from net mass conservation. State geometry-scale differences before comparing peak speeds to a paper. Do not report wall heat-transfer coefficients as design values without a near-wall mesh and y-plus assessment. Keep the final Fluent MCP session loaded when the user is likely to continue interactively. Bundled scripts scripts/mesh_autoclave_paper_case.py : Boolean fluid-domain construction, boundary splitting, and refined tetrahedral mesh. scripts/prepare_fluent_paper_mesh.py : Deterministic Gmsh-to-Fluent ASCII mesh conversion with named zones. scripts/plot_fluent_paper_results.py : Main-view velocity, streamline, temperature, and pressure figures from extracted plane data. scripts/validate_results.py : Acceptance checks for the final result JSON. The three simulation scripts are validated against the supplied autoclave topology. Inspect and adapt their STEP volume ordering, coordinate tests, and output paths for a different geometry.
Agent 识别该技能的关键词,点击任意一个即可复制。
该技能未提供触发词。
下载的 .skill 包内含以下字段。
| 字段 | 说明 |
|---|---|
| format | 格式标识(skill/v1) |
| skill_id | 技能唯一 ID |
| name | 技能名称 |
| version | 版本号 |
| description | 技能描述 |
| category | 所属分类(数组) |
| trigger_words | 触发词列表 |
| tags | 标签列表 |
| source | 来源标识 |
| source_url | 来源链接(本页地址) |
| exported_at | 导出时间(每次下载生成) |
| system_prompt | 系统提示词正文 |
| model_config | 模型参数:provider / model / temperature / max_tokens / top_p |
| examples | 示例 |
| install_guide | 各平台导入说明(Coze / Dify / Claude / 自定义框架) |