科研 Skill
仿真参数优化与敏感性分析
parameter-optimization
Explores and optimizes simulation parameters via design of experiments, sensitivity analysis, and optimizer selection: Latin Hypercube, quasi-random. Use when calibrating simulation parameters.
SKILL.md
Parameter Optimization
Goal
Provide a workflow to design experiments, rank parameter influence, and select optimization strategies for materials simulation calibration.
Requirements
- Python 3.10+
- No external dependencies (uses Python standard library only)
Inputs to Gather
Before running any scripts, collect from the user:
| Input | Description | Example |
|---|---|---|
| Parameter bounds | Min/max for each parameter with units | kappa: [0.1, 10.0] W/mK |
| Evaluation budget | Max number of simulations allowed | 50 runs |
| Noise level | Stochasticity of simulation outputs | low, medium, high |
| Constraints | Feasibility rules or forbidden regions | kappa + mobility < 5 |
Decision Guidance
Choosing a DOE Method
Is dimension <= 3 AND full coverage needed?
├── YES → Use factorial
└── NO → Is sensitivity analysis the goal?
├── YES → Use quasi-random (preferred; "sobol" is accepted but deprecated)
└── NO → Use lhs (Latin Hypercube)
| Method | Best For | Avoid When |
|---|---|---|
lhs | General exploration, moderate dimensions (3-20) | Need exact grid coverage |
quasi-random | Sensitivity analysis, uniform coverage (preferred) | Very high dimensions (>20) |
sobol | Deprecated alias of quasi-random (emits a warning) | New code (use quasi-random) |
factorial | Low dimension (<4), need all corners | High dimension (exponential growth) |
Factorial sizing: the factorial grid is
levelsevenly spaced values per parameter, producing exactlylevels ** paramssamples. Set the resolution explicitly with--levels(e.g.--params 2 --levels 4-> 16 samples). If you use--budgetinstead, the script back-computeslevels = round(budget ** (1/params))and warns whenever the realized sample count differs from the requested budget (e.g.--budget 20 --params 2realizes 16 samples). For an exact design, pass a perfect power (--budget 16) or, preferably,--levels.
Choosing an Optimizer
Is dimension <= 10 AND budget <= 100?
├── YES → Bayesian Optimization
└── NO → Is dimension <= 20?
├── YES → CMA-ES
└── NO → Random Search with screening
| Noise Level | Recommendation |
|---|---|
| Low | Gradient-based if derivatives available, else Bayesian Optimization |
| Medium | Bayesian Optimization with noise model |
| High | Evolutionary algorithms or robust Bayesian Optimization |
Script Outputs (JSON Fields)
| Script | Output Fields |
|---|---|
scripts/doe_generator.py | samples, method, coverage (count, dimension; plus levels and a top-level requested_budget/note for factorial) |
scripts/optimizer_selector.py | recommended, expected_evals, notes |
scripts/sensitivity_summary.py | ranking, notes |
scripts/surrogate_builder.py | model_type, metrics (mse, cv_error, output_variance), notes |
Workflow
- Generate DOE with
scripts/doe_generator.py - Run simulations at DOE sample points (user's responsibility)
- Summarize sensitivity with
scripts/sensitivity_summary.py - Choose optimizer using
scripts/optimizer_selector.py - (Optional) Fit surrogate with
scripts/surrogate_builder.py
CLI Examples
# Generate 20 LHS samples for 3 parameters
python3 scripts/doe_generator.py --params 3 --budget 20 --method lhs --json
# Full factorial with 4 levels per parameter (2 params -> 16 samples)
python3 scripts/doe_generator.py --params 2 --levels 4 --method factorial --json
# Rank parameters by sensitivity scores
python3 scripts/sensitivity_summary.py --scores 0.2,0.5,0.3 --names kappa,mobility,W --json
# Get optimizer recommendation for 3D problem with 50 eval budget
python3 scripts/optimizer_selector.py --dim 3 --budget 50 --noise low --json
# Build surrogate model from simulation data
python3 scripts/surrogate_builder.py --x 0,1,2 --y 10,12,15 --model rbf --json
Conversational Workflow Example
User: I need to calibrate thermal conductivity and diffusivity for my FEM simulation. I can run about 30 simulations.
Agent workflow:
- Identify 2 parameters →
--params 2 - Budget is 30 →
--budget 30 - Use LHS for general exploration:
python3 scripts/doe_generator.py --params 2 --budget 30 --method lhs --json - After user runs simulations and provides outputs, summarize sensitivity:
python3 scripts/sensitivity_summary.py --scores 0.7,0.3 --names conductivity,diffusivity --json - Recommend optimizer:
python3 scripts/optimizer_selector.py --dim 2 --budget 30 --noise low --json
Error Handling
| Error | Cause | Resolution |
|---|---|---|
params must be positive | Zero or negative dimension | Ask user for valid parameter count |
budget must be positive | Zero or negative budget | Ask user for realistic simulation budget |
argument --method: invalid choice: <value> (choose from lhs, sobol, quasi-random, factorial) | Invalid method (argparse) | Use decision guidance to pick a valid method |
could not convert string to float: <token> | Non-numeric value in --scores/--x/--y | Reformat as 0.1,0.2,0.3 |
scores must be a comma-separated list | Empty --scores input | Provide at least one numeric score |
Verification checklist
- Recorded the exact
doe_generator.pycoverage.countand confirmed it matches the intended design — forfactorial, verifiedcount == levels ** paramsand that nonote/requested_budgetmismatch warning was emitted (or that the realized count is acceptable). - Confirmed the chosen
--methodmatches the Decision Guidance for the actual dimension/goal, and thatquasi-randomwas used instead of the deprecatedsobolalias (noDeprecationWarningin output). - Recorded the
optimizer_selector.pyrecommendedstrategy andexpected_evals, and verifiedexpected_evals <= budgetso the plan is feasible within the stated evaluation budget. - Logged the
sensitivity_summary.pyrankingand checked whether the top sensitivity is< 0.1(the "All sensitivities are low" note); if so, did not over-interpret the ranking and revisited the output metric. - For surrogate fits, judged quality with
metrics.cv_error(leave-one-out), NOT in-samplemse— especially forrbf, wheremseis near zero by construction — and comparedcv_erroragainstmetrics.output_varianceto confirm the surrogate beats the constant-mean baseline. - Confirmed any reported
cv_erroris a finite number (notNaN), i.e. there were enough samples for leave-one-out (poly:n > degree+1;rbf:n >= 3).
Common pitfalls & rationalizations
| Tempting shortcut | Why it's wrong / what to do |
|---|---|
"RBF surrogate mse is ~0, so the model is excellent." | RBF is an exact interpolant — in-sample mse is near zero by construction and says nothing about generalization. Judge fit with metrics.cv_error and compare it to output_variance. |
"I asked for --budget 20 factorial, so I got 20 samples." | Factorial honors levels ** params, not the budget; --budget 20 --params 2 realizes 16 samples and emits a note/warning. Use --levels for an exact, intended design. |
"sobol gives me a true Sobol low-discrepancy sequence." | sobol is a deprecated alias that emits a DeprecationWarning and uses a simplified golden-ratio additive recurrence, not a true Sobol sequence. Use quasi-random; for production Sobol use scipy.stats.qmc. |
| "The optimizer recommendation is just advice — budget doesn't matter." | The recommendation is gated on dimension AND budget (BO only for dim<=10 AND budget<=100), and expected_evals is capped at the budget. Record both and confirm the plan fits the real budget. |
| "One sensitivity score is highest, so that parameter dominates." | The script only sorts the scores you pass in; it computes no sensitivity itself. If the top score is < 0.1 it flags that all sensitivities are low — get the scores from a real screening/Sobol analysis before trusting the ranking. |
| "It printed JSON without erroring, so the result is valid." | Exit success only means inputs parsed. Verify the design size, expected_evals <= budget, a finite cv_error, and that the surrogate beats output_variance before trusting any output. |
Security
Input Validation
sensitivity_summary.pyvalidates--namesagainst[a-zA-Z_][a-zA-Z0-9_ .-]*with a 200-char limit, preventing shell metacharacter injection via crafted parameter names- All numeric list inputs are validated as finite numbers (
NaN/Infrejected) - Comma-separated value lists are capped (10,000 for scores, 100,000 for surrogate data) to prevent resource exhaustion
doe_generator.pycaps dimension at 1,000 and budget at 1,000,000;optimizer_selector.pycaps dimension at 100,000 and budget at 10,000,000--methodis validated against a fixed allowlist (lhs,quasi-random/sobol,factorial);sobolis an accepted but deprecated alias ofquasi-random--noiseis validated against a fixed allowlist (low,medium,high)--model(surrogate type) is validated against a fixed allowlist (rbf,poly)--levels(factorial grid resolution) is validated as an integer in[2, 1000]
File Access
- Scripts read no external files; all inputs are provided via CLI arguments
- Scripts write only to stdout (JSON output); no files are created unless the agent explicitly uses the Write tool
Tool Restrictions
- Read: Used to inspect script source, references, and user data files
- Write: Used to save DOE sample plans, sensitivity rankings, or optimizer recommendations; writes are scoped to the user's working directory
- Grep/Glob: Used to locate relevant files and search references
- The skill's
allowed-toolsexcludesBashto prevent the agent from executing arbitrary commands when processing user-provided parameter names and constraints
Safety Measures
- No
eval(),exec(), or dynamic code generation - All subprocess calls use explicit argument lists (no
shell=True) - Reduced tool surface (no Bash) limits the agent to read/write operations only
- Parameter names are sanitized before use, preventing injection via crafted identifiers
Limitations
- Not for real-time optimization: Scripts provide recommendations, not live optimization loops
- Surrogate is lightweight:
surrogate_builder.pyfits a real 1-D least-squares polynomial (poly) or Gaussian RBF interpolant (rbf) using only the standard library and reports honest residualmse, leave-one-outcv_error, and the dataoutput_variance; for production use scipy/scikit-learn/GPyTorch. Forrbf, in-samplemseis near zero by construction (exact interpolation) — judge fit quality withcv_error - No automatic simulation execution: User must run simulations externally and provide results
References
references/doe_methods.md- Detailed DOE method comparisonreferences/optimizer_selection.md- Optimizer algorithm detailsreferences/sensitivity_guidelines.md- Sensitivity analysis interpretationreferences/surrogate_guidelines.md- Surrogate model selection
Version History
-
v1.2.3 (2026-10-19): Center and scale polynomial fits and expose rank-deficient or unvalidated fits.
-
v1.2.2 (2026-06-24): Added Verification checklist and Common pitfalls & rationalizations sections to drive evidence-based use of the DOE, optimizer, sensitivity, and surrogate scripts
-
v1.2.0 (2026-06-23): Real surrogate fits (
polyleast-squares,rbfinterpolation) with honestmse/cv_error/output_variance; explicit factorial--levelswith budget-mismatch warnings; BO dimension cutoff harmonized to dim<=10; corrected Security/Error-Handling/output-field docs to match script behavior -
v1.1.0 (2024-12-24): Enhanced documentation, decision guidance, conversational examples
-
v1.0.0: Initial release with core scripts
Numerical and evidence limits
Polynomial surrogate fits retain a centered/scaled basis and use SVD least squares. Rank failure is reported as unavailable fit metrics. Cross-validation with insufficient samples is null. Neither fit residual nor cross-validation establishes physical validity outside the sampled domain.
示例提示词
帮我用拉丁超立方抽样对这个仿真模型的参数做敏感性排序和优化。
所属插件
安装该插件会一并获得同领域的 23 个 Skill / MCP。
同插件的其他科研 Skill
- 制造解法验证与基准测试规划Plans verification and validation campaigns using manufactured solutions, canonical benchmarks. Use when an agent must prove a solver is correct before trusting results.
- 数值解空间时间收敛性分析Performs spatial/temporal convergence analysis for solution verification: observed order, Richardson extrapolation. Use when checking that a simulation result is grid/timestep independent.
- PDE数值微分格式选择Selects and applies numerical differentiation schemes for PDE/ODE discretization: finite-difference stencils. Use when choosing a spatial discretization scheme.
- 仿真结果FAIR归档打包Creates FAIR-minded reproducibility bundles for simulations: file inventories, hashes, units, engine versions. Use before publishing or archiving a simulation dataset.
- HPC作业运行诊断Diagnoses HPC runtime and scheduler problems for materials simulations: MPI/OpenMP/GPU layout, modules, scratch paths, walltime. Use when a cluster job fails, hangs, or underperforms.
- 线性方程组求解器选型Selects and configures linear solvers for Ax=b systems: direct (LU, Cholesky) versus iterative (CG, GMRES, BiCGSTAB) methods, sparsity. Use when choosing a linear solver for a simulation.
- 分子动力学分析方案规划Plans molecular dynamics post-processing for materials simulations: RDF, MSD/diffusion, VACF/VDOS, coordination numbers. Use when deciding which MD analysis to run and in what order.
- 数值模拟网格生成规划Plans and evaluates mesh generation for numerical simulations: grid resolution from physics scales, aspect ratio/skewness checks. Use when designing or auditing a simulation mesh.
- 非线性方程组求解器选型Selects and configures nonlinear solvers for root-finding, optimization, and least-squares problems: Newton, quasi-Newton, Broyden, Anderson acceleration. Use when choosing a nonlinear solver.
- 时间积分方法选择Selects and configures time-integration methods for ODE/PDE simulations: explicit Runge-Kutta, BDF, Rosenbrock. Use when choosing a time-stepping scheme for a solver.
- 偏微分方程数值稳定性分析Analyzes numerical stability for time-dependent PDE simulations: CFL/Fourier criteria, von Neumann analysis. Use when a simulation blows up or to check a scheme’s stability limits.
- 材料科学本体浏览查询Parses, navigates, and queries materials science ontology structures: class hierarchies, property domain/range lookup. Use when exploring an unfamiliar materials ontology.
- 材料术语本体映射Maps materials science terms, crystal structures, and sample descriptions to standardized ontology classes with confidence scores. Use when annotating data for ontology interoperability.
- 本体标注一致性校验Validates material sample annotations against ontology constraints: class/property existence, domain/range consistency. Use before publishing ontology-annotated materials data.
- 仿真计算性能剖析Identifies computational bottlenecks, parallel scaling, and memory requirements for materials simulations: parses timing logs for dominant phases. Use when a simulation is slow or scales poorly.
- 仿真输出数据后处理Extracts, analyzes, and summarizes simulation output data: spatial fields at timesteps, time-series trends, steady-state detection. Use when turning raw simulation output into summary statistics.
- 多代码仿真故障分诊Provides first-response triage for cross-code LAMMPS/VASP/QE/MOOSE simulation failures: classifies nonconvergence, NaN/Inf. Use when a simulation crashes and the cause is unclear.
- 多任务仿真批量编排Orchestrates multi-simulation campaigns: generates parameter sweeps (grid, linspace, Latin Hypercube), tracks batch job status. Use when running and managing many related simulation jobs.
- 仿真三阶段有效性校验Validates simulations across three stages: pre-flight config checks, runtime log monitoring for NaN/Inf and residual growth. Use to catch simulation errors before, during, and after a run.
- SLURM作业脚本生成Generates correct SLURM sbatch job scripts and sanity-checks HPC resource requests: nodes, MPI/OpenMP layout, memory, GPUs, walltime. Use when writing or debugging a SLURM submission script.
- 瞬态仿真时间步策略Plans and controls time-step policies for transient simulations: CFL and physics-based stability limits, adaptive stepping. Use when setting up a time-stepping strategy for a solver.
- 计算材料工作流引擎选型Maps computational materials tasks onto workflow engines (atomate2, jobflow, AiiDA, pyiron) or a one-off script. Use when deciding how to structure a reproducible campaign or DAG.
常见问题
仿真参数优化与敏感性分析能做什么?
Explores and optimizes simulation parameters via design of experiments, sensitivity analysis, and optimizer selection: Latin Hypercube, quasi-random. Use when calibrating simulation parameters。
如何安装仿真参数优化与敏感性分析?
在科搭 Codar 左侧点击「插件」,在「组件」页签搜索「仿真参数优化与敏感性分析」并安装;在 Claude Code 或 Codex 中可以发送本页的安装提示词,或下载 Skill 包后解压到 ~/.claude/skills 或 ~/.codex/skills。
仿真参数优化与敏感性分析需要哪些权限?
该科研 Skill没有声明额外权限;执行具体操作时仍由 Agent 按自身规则请求确认。
仿真参数优化与敏感性分析经过官方审核吗?
是。SpeedSkills 收录的每个版本都会记录内容摘要与权限声明,当前版本已通过安全审核。