Antoine/Atomizer
1
0
Fork 0
Code Issues Pull Requests Actions Packages Projects Releases Wiki Activity

feat: Complete Phase 2.5-2.7 - Intelligent LLM-Powered Workflow Analysis

Browse Source 
This commit implements three major architectural improvements to transform
Atomizer from static pattern matching to intelligent AI-powered analysis.

## Phase 2.5: Intelligent Codebase-Aware Gap Detection ✅

Created intelligent system that understands existing capabilities before
requesting examples:

**New Files:**
- optimization_engine/codebase_analyzer.py (379 lines)
  Scans Atomizer codebase for existing FEA/CAE capabilities

- optimization_engine/workflow_decomposer.py (507 lines, v0.2.0)
  Breaks user requests into atomic workflow steps
  Complete rewrite with multi-objective, constraints, subcase targeting

- optimization_engine/capability_matcher.py (312 lines)
  Matches workflow steps to existing code implementations

- optimization_engine/targeted_research_planner.py (259 lines)
  Creates focused research plans for only missing capabilities

**Results:**
- 80-90% coverage on complex optimization requests
- 87-93% confidence in capability matching
- Fixed expression reading misclassification (geometry vs result_extraction)

## Phase 2.6: Intelligent Step Classification ✅

Distinguishes engineering features from simple math operations:

**New Files:**
- optimization_engine/step_classifier.py (335 lines)

**Classification Types:**
1. Engineering Features - Complex FEA/CAE needing research
2. Inline Calculations - Simple math to auto-generate
3. Post-Processing Hooks - Middleware between FEA steps

## Phase 2.7: LLM-Powered Workflow Intelligence ✅

Replaces static regex patterns with Claude AI analysis:

**New Files:**
- optimization_engine/llm_workflow_analyzer.py (395 lines)
  Uses Claude API for intelligent request analysis
  Supports both Claude Code (dev) and API (production) modes

- .claude/skills/analyze-workflow.md
  Skill template for LLM workflow analysis integration

**Key Breakthrough:**
- Detects ALL intermediate steps (avg, min, normalization, etc.)
- Understands engineering context (CBUSH vs CBAR, directions, metrics)
- Distinguishes OP2 extraction from part expression reading
- Expected 95%+ accuracy with full nuance detection

## Test Coverage

**New Test Files:**
- tests/test_phase_2_5_intelligent_gap_detection.py (335 lines)
- tests/test_complex_multiobj_request.py (130 lines)
- tests/test_cbush_optimization.py (130 lines)
- tests/test_cbar_genetic_algorithm.py (150 lines)
- tests/test_step_classifier.py (140 lines)
- tests/test_llm_complex_request.py (387 lines)

All tests include:
- UTF-8 encoding for Windows console
- atomizer environment (not test_env)
- Comprehensive validation checks

## Documentation

**New Documentation:**
- docs/PHASE_2_5_INTELLIGENT_GAP_DETECTION.md (254 lines)
- docs/PHASE_2_7_LLM_INTEGRATION.md (227 lines)
- docs/SESSION_SUMMARY_PHASE_2_5_TO_2_7.md (252 lines)

**Updated:**
- README.md - Added Phase 2.5-2.7 completion status
- DEVELOPMENT_ROADMAP.md - Updated phase progress

## Critical Fixes

1. **Expression Reading Misclassification** (lines cited in session summary)
   - Updated codebase_analyzer.py pattern detection
   - Fixed workflow_decomposer.py domain classification
   - Added capability_matcher.py read_expression mapping

2. **Environment Standardization**
   - All code now uses 'atomizer' conda environment
   - Removed test_env references throughout

3. **Multi-Objective Support**
   - WorkflowDecomposer v0.2.0 handles multiple objectives
   - Constraint extraction and validation
   - Subcase and direction targeting

## Architecture Evolution

**Before (Static & Dumb):**
User Request → Regex Patterns → Hardcoded Rules → Missed Steps ❌

**After (LLM-Powered & Intelligent):**
User Request → Claude AI Analysis → Structured JSON →
├─ Engineering (research needed)
├─ Inline (auto-generate Python)
├─ Hooks (middleware scripts)
└─ Optimization (config) ✅

## LLM Integration Strategy

**Development Mode (Current):**
- Use Claude Code directly for interactive analysis
- No API consumption or costs
- Perfect for iterative development

**Production Mode (Future):**
- Optional Anthropic API integration
- Falls back to heuristics if no API key
- For standalone batch processing

## Next Steps

- Phase 2.8: Inline Code Generation
- Phase 2.9: Post-Processing Hook Generation
- Phase 3: MCP Integration for automated documentation research

🚀 Generated with Claude Code

Co-Authored-By: Claude <noreply@anthropic.com>
This commit is contained in:
Anto01
2025-11-16 13:35:41 -05:00
parent 986285d9cf
commit 0a7cca9c6a
94 changed files with 12761 additions and 10670 deletions
Download Patch File Download Diff File Expand all files Collapse all files

305
studies/README.md Normal file
View File

@@ -0,0 +1,305 @@
# Atomizer Studies Directory
This directory contains optimization studies for the Atomizer framework. Each study is a self-contained workspace for running NX optimization campaigns.
## Directory Structure
```
studies/
├── README.md # This file
├── _templates/ # Study templates for quick setup
│ ├── basic_stress_optimization/
│ ├── multi_objective/
│ └── constrained_optimization/
├── _archive/ # Completed/old studies
│ └── YYYY-MM-DD_study_name/
└── [active_studies]/ # Your active optimization studies
└── bracket_stress_minimization/ # Example study
```
## Study Folder Structure
Each study should follow this standardized structure:
```
study_name/
├── README.md # Study description, objectives, notes
├── optimization_config.json # Atomizer configuration file
│
├── model/ # FEA model files (NX or other solvers)
│ ├── model.prt # NX part file
│ ├── model.sim # NX Simcenter simulation file
│ ├── model.fem # FEM file
│ └── assembly.asm # NX assembly (if applicable)
│
├── optimization_results/ # Generated by Atomizer (DO NOT COMMIT)
│ ├── optimization.log # High-level optimization progress log
│ ├── trial_logs/ # Detailed iteration logs (one per trial)
│ │ ├── trial_000_YYYYMMDD_HHMMSS.log
│ │ ├── trial_001_YYYYMMDD_HHMMSS.log
│ │ └── ...
│ ├── history.json # Complete optimization history
│ ├── history.csv # CSV format for analysis
│ ├── optimization_summary.json # Best results summary
│ ├── study_*.db # Optuna database files
│ └── study_*_metadata.json # Study metadata for resumption
│
├── analysis/ # Post-optimization analysis
│ ├── plots/ # Generated visualizations
│ ├── reports/ # Generated PDF/HTML reports
│ └── sensitivity_analysis.md # Analysis notes
│
└── notes.md # Engineering notes, decisions, insights
```
## Creating a New Study
### Option 1: From Template
```bash
# Copy a template
cp -r studies/_templates/basic_stress_optimization studies/my_new_study
cd studies/my_new_study
# Edit the configuration
# - Update optimization_config.json
# - Place your .sim, .prt, .fem files in model/
# - Update README.md with study objectives
```
### Option 2: Manual Setup
```bash
# Create study directory
mkdir -p studies/my_study/{model,analysis/plots,analysis/reports}
# Create config file
# (see _templates/ for examples)
# Add your files
# - Place all FEA files (.prt, .sim, .fem) in model/
# - Edit optimization_config.json
```
## Running an Optimization
```bash
# Navigate to project root
cd /path/to/Atomizer
# Run optimization for a study
python run_study.py --study studies/my_study
# Or use the full path to config
python -c "from optimization_engine.runner import OptimizationRunner; ..."
```
## Configuration File Format
The `optimization_config.json` file defines the optimization setup:
```json
{
"design_variables": [
{
"name": "thickness",
"type": "continuous",
"bounds": [3.0, 8.0],
"units": "mm",
"initial_value": 5.0
}
],
"objectives": [
{
"name": "minimize_stress",
"description": "Minimize maximum von Mises stress",
"extractor": "stress_extractor",
"metric": "max_von_mises",
"direction": "minimize",
"weight": 1.0,
"units": "MPa"
}
],
"constraints": [
{
"name": "displacement_limit",
"description": "Maximum allowable displacement",
"extractor": "displacement_extractor",
"metric": "max_displacement",
"type": "upper_bound",
"limit": 1.0,
"units": "mm"
}
],
"optimization_settings": {
"n_trials": 50,
"sampler": "TPE",
"n_startup_trials": 20,
"tpe_n_ei_candidates": 24,
"tpe_multivariate": true
},
"model_info": {
"sim_file": "model/model.sim",
"note": "Brief description"
}
}
```
## Results Organization
All optimization results are stored in `optimization_results/` within each study folder.
### Optimization Log (optimization.log)
**High-level overview** of the entire optimization run:
- Optimization configuration (design variables, objectives, constraints)
- One compact line per trial showing design variables and results
- Easy to scan and monitor optimization progress
- Perfect for quick reviews and debugging
**Example format**:
```
[08:15:35] Trial 0 START | tip_thickness=20.450, support_angle=32.100
[08:15:42] Trial 0 COMPLETE | max_von_mises=245.320, max_displacement=0.856
[08:15:45] Trial 1 START | tip_thickness=18.230, support_angle=28.900
[08:15:51] Trial 1 COMPLETE | max_von_mises=268.450, max_displacement=0.923
```
### Trial Logs (trial_logs/)
**Detailed per-trial logs** showing complete iteration trace:
- Design variable values for the trial
- Complete optimization configuration
- Execution timeline (pre_solve, solve, post_solve, extraction)
- Extracted results (stress, displacement, etc.)
- Constraint evaluations
- Hook execution trace
- Solver output and warnings
**Example**: `trial_005_20251116_143022.log`
These logs are invaluable for:
- Debugging failed trials
- Understanding what happened in specific iterations
- Verifying solver behavior
- Tracking hook execution
### History Files
**Structured data** for analysis and visualization:
- **history.json**: Complete trial-by-trial results in JSON format
- **history.csv**: Same data in CSV for Excel/plotting
- **optimization_summary.json**: Best parameters and final results
### Optuna Database
**Study persistence** for resuming optimizations:
- **study_NAME.db**: SQLite database storing all trial data
- **study_NAME_metadata.json**: Study metadata and configuration hash
The database allows you to:
- Resume interrupted optimizations
- Add more trials to a completed study
- Query optimization history programmatically
## Best Practices
### Study Naming
- Use descriptive names: `bracket_stress_minimization` not `test1`
- Include objective: `wing_mass_displacement_tradeoff`
- Version if iterating: `bracket_v2_reduced_mesh`
### Documentation
- Always fill out README.md in each study folder
- Document design decisions in notes.md
- Keep analysis/ folder updated with plots and reports
### Version Control
Add to `.gitignore`:
```
studies/*/optimization_results/
studies/*/analysis/plots/
studies/*/__pycache__/
```
Commit to git:
```
studies/*/README.md
studies/*/optimization_config.json
studies/*/notes.md
studies/*/model/*.sim
studies/*/model/*.prt (optional - large CAD files)
studies/*/model/*.fem
```
### Archiving Completed Studies
When a study is complete:
```bash
# Archive the study
mv studies/completed_study studies/_archive/2025-11-16_completed_study
# Update _archive/README.md with study summary
```
## Study Templates
### Basic Stress Optimization
- Single objective: minimize stress
- Single design variable
- Simple mesh
- Good for learning/testing
### Multi-Objective Optimization
- Multiple competing objectives (stress, mass, displacement)
- Pareto front analysis
- Weighted sum approach
### Constrained Optimization
- Objectives with hard constraints
- Demonstrates constraint handling
- Pruned trials when constraints violated
## Troubleshooting
### Study won't resume
Check that `optimization_config.json` hasn't changed. The config hash is stored in metadata and verified on resume.
### Missing trial logs or optimization.log
Ensure logging plugins are enabled:
- `optimization_engine/plugins/pre_solve/detailed_logger.py` - Creates detailed trial logs
- `optimization_engine/plugins/pre_solve/optimization_logger.py` - Creates high-level optimization.log
- `optimization_engine/plugins/post_extraction/log_results.py` - Appends results to trial logs
- `optimization_engine/plugins/post_extraction/optimization_logger_results.py` - Appends results to optimization.log
### Results directory missing
The directory is created automatically on first run. Check file permissions.
## Advanced: Custom Hooks
Studies can include custom hooks in a `hooks/` folder:
```
my_study/
├── hooks/
│ ├── pre_solve/
│ │ └── custom_parameterization.py
│ └── post_extraction/
│ └── custom_objective.py
└── ...
```
These hooks are automatically loaded if present.
## Questions?
- See main README.md for Atomizer documentation
- See DEVELOPMENT_ROADMAP.md for planned features
- Check docs/ for detailed guides

86
studies/bracket_stress_minimization/README.md Normal file
View File

@@ -0,0 +1,86 @@
# Bracket Stress Minimization Study
## Overview
This study optimizes a structural bracket to minimize maximum von Mises stress while maintaining displacement constraints.
## Objective
Minimize maximum von Mises stress in the bracket under applied loading conditions.
## Design Variables
- **tip_thickness**: 15.0 - 25.0 mm
- Controls the thickness of the bracket tip
- Directly affects stress distribution and structural rigidity
- **support_angle**: 20.0 - 40.0 degrees
- Controls the angle of the support structure
- Affects load path and stress concentration
## Constraints
- **Maximum displacement** ≤ 1.0 mm
- Ensures the bracket maintains acceptable deformation under load
- Prevents excessive deflection that could affect functionality
## Model Information
All FEA files are located in [model/](model/):
- **Part**: [Bracket.prt](model/Bracket.prt)
- **Simulation**: [Bracket_sim1.sim](model/Bracket_sim1.sim)
- **FEM**: [Bracket_fem1.fem](model/Bracket_fem1.fem)
## Optimization Settings
- **Sampler**: TPE (Tree-structured Parzen Estimator)
- **Total trials**: 50
- **Startup trials**: 20 (random sampling for initial exploration)
- **TPE candidates**: 24
- **Multivariate**: Enabled
## Running the Optimization
From the project root:
```bash
python run_5trial_test.py # Quick 5-trial test
```
Or for the full optimization:
```python
from pathlib import Path
from optimization_engine.runner import OptimizationRunner
config_path = Path("studies/bracket_stress_minimization/optimization_config_stress_displacement.json")
runner = OptimizationRunner(
config_path=config_path,
model_updater=bracket_model_updater,
simulation_runner=bracket_simulation_runner,
result_extractors={...}
)
study = runner.run(study_name="bracket_study", n_trials=50)
```
## Results
Results are stored in [optimization_results/](optimization_results/):
- **trial_logs/**: Detailed logs for each trial iteration
- **history.json**: Complete trial-by-trial results
- **history.csv**: Results in CSV format for analysis
- **optimization_summary.json**: Best parameters and final results
- **study_*.db**: Optuna database for resuming optimizations
## Notes
- Uses NX Simcenter 2412 for FEA simulation
- Journal-based solver execution for automation
- Results extracted from OP2 files using pyNastran
- Stress values in MPa, displacement in mm
## Analysis
Post-optimization analysis plots and reports will be stored in [analysis/](analysis/).

BIN
studies/bracket_stress_minimization/model/Bracket.prt Normal file
View File

Binary file not shown.

BIN
studies/bracket_stress_minimization/model/Bracket_fem1.fem Normal file
View File

Binary file not shown.

BIN
studies/bracket_stress_minimization/model/Bracket_sim1.sim Normal file
View File

Binary file not shown.

52
studies/bracket_stress_minimization/optimization_config_stress_displacement.json Normal file
View File

@@ -0,0 +1,52 @@
{
"design_variables": [
{
"name": "tip_thickness",
"type": "continuous",
"bounds": [15.0, 25.0],
"units": "mm",
"initial_value": 20.0
},
{
"name": "support_angle",
"type": "continuous",
"bounds": [20.0, 40.0],
"units": "degrees",
"initial_value": 35.0
}
],
"objectives": [
{
"name": "minimize_max_stress",
"description": "Minimize maximum von Mises stress",
"extractor": "stress_extractor",
"metric": "max_von_mises",
"direction": "minimize",
"weight": 10.0,
"units": "MPa"
}
],
"constraints": [
{
"name": "max_displacement_limit",
"description": "Maximum allowable displacement",
"extractor": "displacement_extractor",
"metric": "max_displacement",
"type": "upper_bound",
"limit": 1.0,
"units": "mm"
}
],
"optimization_settings": {
"n_trials": 50,
"sampler": "TPE",
"n_startup_trials": 20,
"tpe_n_ei_candidates": 24,
"tpe_multivariate": true,
"comment": "20 random trials for exploration, then 30 TPE trials for exploitation"
},
"model_info": {
"sim_file": "C:\\Users\\antoi\\Documents\\Atomaste\\Atomizer\\examples\\bracket\\Bracket_sim1.sim",
"note": "Stress minimization with displacement constraint (mass not available in OP2)"
}
}