Files
differential-equations/roadmap/features/01-bs3-method.md
Connor Johnstone e3788bf607 Added the roadmap
2025-10-23 16:47:48 -04:00

192 lines
5.7 KiB
Markdown

# Feature: BS3 (Bogacki-Shampine 3/2) Method
## Overview
The Bogacki-Shampine 3/2 method is a 3rd order explicit Runge-Kutta method with an embedded 2nd order method for error estimation. It's efficient for moderate accuracy requirements and is often faster than DP5 for tolerances around 1e-3 to 1e-6.
**Key Characteristics:**
- Order: 3(2) - 3rd order solution with 2nd order error estimate
- Stages: 4
- FSAL: Yes (First Same As Last)
- Adaptive: Yes
- Dense output: 3rd order continuous extension
## Why This Feature Matters
- **Efficiency**: Fewer stages than DP5 (4 vs 7) for comparable accuracy at moderate tolerances
- **Common use case**: Many practical problems don't need DP5's accuracy
- **Algorithm diversity**: Gives users choice based on problem characteristics
- **Foundation**: Good reference implementation for adding more RK methods
## Dependencies
- None (can be implemented with current infrastructure)
## Implementation Approach
### Butcher Tableau
The BS3 method uses the following coefficients:
```
c | A
--+-------
0 | 0
1/2 | 1/2
3/4 | 0 3/4
1 | 2/9 1/3 4/9
--+-------
b | 2/9 1/3 4/9 0 (3rd order)
b*| 7/24 1/4 1/3 1/8 (2nd order, for error)
```
FSAL property: The last stage k4 can be reused as k1 of the next step.
### Dense Output
3rd order Hermite interpolation:
```
u(t₀ + θh) = u₀ + h*θ*(b₁*k₁ + b₂*k₂ + b₃*k₃) + h*θ*(1-θ)*(...additional terms)
```
Coefficients from Bogacki & Shampine 1989 paper.
### Error Estimation
```
err = ||u₃ - u₂|| / (atol + max(|u_n|, |u_{n+1}|) * rtol)
```
Where u₃ is the 3rd order solution and u₂ is the 2nd order embedded solution.
## Implementation Tasks
### Core Algorithm
- [ ] Define `BS3` struct implementing `Integrator<D>` trait
- [ ] Add tableau constants (A, b, b_error, c)
- [ ] Add tolerance fields (a_tol, r_tol)
- [ ] Add builder methods for setting tolerances
- [ ] Implement `step()` method
- [ ] Compute k1 = f(t, y)
- [ ] Compute k2 = f(t + c[1]*h, y + h*a[0,0]*k1)
- [ ] Compute k3 = f(t + c[2]*h, y + h*(a[1,0]*k1 + a[1,1]*k2))
- [ ] Compute k4 = f(t + c[3]*h, y + h*(a[2,0]*k1 + a[2,1]*k2 + a[2,2]*k3))
- [ ] Compute 3rd order solution: y_next = y + h*(b[0]*k1 + b[1]*k2 + b[2]*k3 + b[3]*k4)
- [ ] Compute error estimate: err = h*(b[0]-b*[0])*k1 + ... (for all ki)
- [ ] Store dense output coefficients [k1, k2, k3, k4]
- [ ] Return (y_next, Some(error_norm), Some(dense_coeffs))
- [ ] Implement `interpolate()` method
- [ ] Calculate θ = (t - t_start) / (t_end - t_start)
- [ ] Evaluate 3rd order interpolation polynomial
- [ ] Return interpolated state
- [ ] Implement constants
- [ ] `ORDER = 3`
- [ ] `STAGES = 4`
- [ ] `ADAPTIVE = true`
- [ ] `DENSE = true`
### Integration with Problem
- [ ] Export BS3 in prelude
- [ ] Add to `integrator/mod.rs` module exports
### Testing
- [ ] **Convergence test**: Linear problem (y' = λy)
- [ ] Run with decreasing tolerances
- [ ] Verify 3rd order convergence rate
- [ ] Compare to analytical solution
- [ ] **Accuracy test**: Exponential decay
- [ ] y' = -y, y(0) = 1
- [ ] Verify error < tolerance at t=5
- [ ] Check intermediate points via interpolation
- [ ] **FSAL test**: Verify function evaluation count
- [ ] Count evaluations for multi-step integration
- [ ] Should be ~4n for n accepted steps (plus rejections)
- [ ] **Dense output test**:
- [ ] Interpolate at multiple points
- [ ] Verify 3rd order accuracy of interpolation
- [ ] Compare to fine-step reference solution
- [ ] **Comparison test**: Run same problem with DP5 and BS3
- [ ] Verify both achieve required tolerance
- [ ] BS3 should use fewer steps at moderate tolerances
### Benchmarking
- [ ] Add benchmark in `benches/`
- [ ] Simple 1D problem
- [ ] 6D orbital mechanics problem
- [ ] Compare to DP5 performance
### Documentation
- [ ] Add docstring to BS3 struct
- [ ] Explain when to use BS3 vs DP5
- [ ] Note FSAL property
- [ ] Reference original paper
- [ ] Add usage example
- [ ] Show tolerance selection
- [ ] Demonstrate interpolation
## Testing Requirements
### Convergence Test Details
Standard test problem: y' = -5y, y(0) = 1, exact solution: y(t) = e^(-5t)
Run from t=0 to t=1 with tolerances: [1e-3, 1e-4, 1e-5, 1e-6, 1e-7]
Expected: Error tolerance^3 (3rd order convergence)
### Stiffness Note
BS3 is an explicit method and will struggle with stiff problems. Include a test that demonstrates this limitation (e.g., Van der Pol oscillator with large μ should require many steps).
## References
1. **Original Paper**:
- Bogacki, P. and Shampine, L.F. (1989), "A 3(2) pair of Runge-Kutta formulas",
Applied Mathematics Letters, Vol. 2, No. 4, pp. 321-325
- DOI: 10.1016/0893-9659(89)90079-7
2. **Dense Output**:
- Same paper, Section 3
3. **Julia Implementation**:
- `OrdinaryDiffEq.jl/lib/OrdinaryDiffEqLowOrderRK/src/low_order_rk_perform_step.jl`
- Look for `perform_step!` for `BS3` cache
4. **Textbook Reference**:
- Hairer, Nørsett, Wanner (2008), "Solving Ordinary Differential Equations I: Nonstiff Problems"
- Chapter II.4 on embedded methods
## Complexity Estimate
**Effort**: Small (2-4 hours)
- Straightforward explicit RK implementation
- Similar structure to existing DP5
- Main work is getting tableau coefficients correct and testing
**Risk**: Low
- Well-understood algorithm
- No new infrastructure needed
- Easy to validate against reference solutions
## Success Criteria
- [ ] Passes convergence test with 3rd order rate
- [ ] Passes all accuracy tests within specified tolerances
- [ ] FSAL optimization verified via function evaluation count
- [ ] Dense output achieves 3rd order interpolation accuracy
- [ ] Performance comparable to Julia implementation for similar problems
- [ ] Documentation complete with examples