[{"data":1,"prerenderedAt":-1},["ShallowReactive",2],{"skill:01M72WNWN0FT7K9JS0H2ES1P27":3,"skill-same:01M72WNWN0FT7K9JS0H2ES1P27":46},{"skillNo":4,"name":5,"nameZh":6,"description":7,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":6,"sourceHtmlUrl":16,"categories":17,"keywords":18,"publishedAt":19,"linkOnly":20,"markdownContent":21,"markdownBody":22,"frontMatter":23,"refName":24,"commitSha":25,"skillDirectory":26,"forksCount":27,"repositoryHtmlUrl":28,"repositoryHomepageUrl":29,"ownerHtmlUrl":30,"ownerBlogUrl":6,"ownerDescription":6,"updatedAt":31,"install":32,"resources":45},"01M72WNWN0FT7K9JS0H2ES1P27","imc-tuning-rules",null,"Calculate PI\u002FPID controller gains using Internal Model Control (IMC) tuning rules for first-order systems.","agentscope-ai\u002FPawBench","agentscope-ai","https:\u002F\u002Favatars.githubusercontent.com\u002Fu\u002F211762292?v=4","01M6CRT6543BZGGVK3P85GC08Q","AgentScope-AI","01M72WJAXCYDAXK74090327ER7",127,"Apache-2.0","https:\u002F\u002Fgithub.com\u002Fagentscope-ai\u002FPawBench\u002Fblob\u002Fmain\u002Fdata\u002Fpawbench-v1.0\u002Fassets\u002FT131_skillsbench_hvac-control\u002Fskills\u002Fimc-tuning-rules\u002FSKILL.md",[],[],"2026-11-10 22:39:42",false,"---\nname: imc-tuning-rules\ndescription: Calculate PI\u002FPID controller gains using Internal Model Control (IMC) tuning rules for first-order systems.\n---\n\n# IMC Tuning Rules for PI\u002FPID Controllers\n\n## Overview\n\nInternal Model Control (IMC) is a systematic method for tuning PI\u002FPID controllers based on a process model. Once you've identified system parameters (K and tau), IMC provides controller gains.\n\n## Why IMC?\n\n- **Model-based**: Uses identified process parameters directly\n- **Single tuning parameter**: Just choose the closed-loop speed (lambda)\n- **Guaranteed stability**: For first-order systems, always stable if model is accurate\n- **Predictable response**: Closed-loop time constant equals lambda\n\n## IMC Tuning for First-Order Systems\n\nFor a first-order process with gain K and time constant tau:\n\n```\nProcess: G(s) = K \u002F (tau*s + 1)\n```\n\nThe IMC-tuned PI controller gains are:\n\n```\nKp = tau \u002F (K * lambda)\nKi = Kp \u002F tau = 1 \u002F (K * lambda)\nKd = 0  (derivative not needed for first-order systems)\n```\n\nWhere:\n- `Kp` = Proportional gain\n- `Ki` = Integral gain (units: 1\u002Ftime)\n- `Kd` = Derivative gain (zero for first-order)\n- `lambda` = Desired closed-loop time constant (tuning parameter)\n\n## Choosing Lambda (λ)\n\nLambda controls the trade-off between speed and robustness:\n\n| Lambda | Behavior |\n|--------|----------|\n| `lambda = 0.1 * tau` | Very aggressive, fast but sensitive to model error |\n| `lambda = 0.5 * tau` | Aggressive, good for accurate models |\n| `lambda = 1.0 * tau` | Moderate, balanced speed and robustness |\n| `lambda = 2.0 * tau` | Conservative, robust to model uncertainty |\n\n**Default recommendation**: Start with `lambda = tau`\n\nFor noisy systems or uncertain models, use larger lambda.\nFor precise models and fast response needs, use smaller lambda.\n\n## Implementation\n\n```python\ndef calculate_imc_gains(K, tau, lambda_factor=1.0):\n    \"\"\"\n    Calculate IMC-tuned PI gains for a first-order system.\n\n    Args:\n        K: Process gain\n        tau: Time constant\n        lambda_factor: Multiplier for lambda (default 1.0 = lambda equals tau)\n\n    Returns:\n        dict with Kp, Ki, Kd, lambda\n    \"\"\"\n    lambda_cl = lambda_factor * tau\n\n    Kp = tau \u002F (K * lambda_cl)\n    Ki = Kp \u002F tau\n    Kd = 0.0\n\n    return {\n        \"Kp\": Kp,\n        \"Ki\": Ki,\n        \"Kd\": Kd,\n        \"lambda\": lambda_cl\n    }\n```\n\n## PI Controller Implementation\n\n```python\nclass PIController:\n    def __init__(self, Kp, Ki, setpoint):\n        self.Kp = Kp\n        self.Ki = Ki\n        self.setpoint = setpoint\n        self.integral = 0.0\n\n    def compute(self, measurement, dt):\n        \"\"\"Compute control output.\"\"\"\n        error = self.setpoint - measurement\n\n        # Integral term\n        self.integral += error * dt\n\n        # PI control law\n        output = self.Kp * error + self.Ki * self.integral\n\n        # Clamp to valid range\n        output = max(output_min, min(output_max, output))\n\n        return output\n```\n\n## Expected Closed-Loop Behavior\n\nWith IMC tuning, the closed-loop response is approximately:\n\n```\ny(t) = y_setpoint * (1 - exp(-t \u002F lambda))\n```\n\nKey properties:\n- **Rise time**: ~2.2 * lambda to reach 90% of setpoint\n- **Settling time**: ~4 * lambda to reach 98% of setpoint\n- **Overshoot**: Minimal for first-order systems\n- **Steady-state error**: Zero (integral action eliminates offset)\n\n## Tips\n\n1. **Start conservative**: Use `lambda = tau` initially\n2. **Decrease lambda carefully**: Smaller lambda = larger Kp = faster but riskier\n3. **Watch for oscillation**: If output oscillates, increase lambda\n4. **Anti-windup**: Prevent integral wind-up when output saturates\n","# IMC Tuning Rules for PI\u002FPID Controllers\n\n## Overview\n\nInternal Model Control (IMC) is a systematic method for tuning PI\u002FPID controllers based on a process model. Once you've identified system parameters (K and tau), IMC provides controller gains.\n\n## Why IMC?\n\n- **Model-based**: Uses identified process parameters directly\n- **Single tuning parameter**: Just choose the closed-loop speed (lambda)\n- **Guaranteed stability**: For first-order systems, always stable if model is accurate\n- **Predictable response**: Closed-loop time constant equals lambda\n\n## IMC Tuning for First-Order Systems\n\nFor a first-order process with gain K and time constant tau:\n\n```\nProcess: G(s) = K \u002F (tau*s + 1)\n```\n\nThe IMC-tuned PI controller gains are:\n\n```\nKp = tau \u002F (K * lambda)\nKi = Kp \u002F tau = 1 \u002F (K * lambda)\nKd = 0  (derivative not needed for first-order systems)\n```\n\nWhere:\n- `Kp` = Proportional gain\n- `Ki` = Integral gain (units: 1\u002Ftime)\n- `Kd` = Derivative gain (zero for first-order)\n- `lambda` = Desired closed-loop time constant (tuning parameter)\n\n## Choosing Lambda (λ)\n\nLambda controls the trade-off between speed and robustness:\n\n| Lambda | Behavior |\n|--------|----------|\n| `lambda = 0.1 * tau` | Very aggressive, fast but sensitive to model error |\n| `lambda = 0.5 * tau` | Aggressive, good for accurate models |\n| `lambda = 1.0 * tau` | Moderate, balanced speed and robustness |\n| `lambda = 2.0 * tau` | Conservative, robust to model uncertainty |\n\n**Default recommendation**: Start with `lambda = tau`\n\nFor noisy systems or uncertain models, use larger lambda.\nFor precise models and fast response needs, use smaller lambda.\n\n## Implementation\n\n```python\ndef calculate_imc_gains(K, tau, lambda_factor=1.0):\n    \"\"\"\n    Calculate IMC-tuned PI gains for a first-order system.\n\n    Args:\n        K: Process gain\n        tau: Time constant\n        lambda_factor: Multiplier for lambda (default 1.0 = lambda equals tau)\n\n    Returns:\n        dict with Kp, Ki, Kd, lambda\n    \"\"\"\n    lambda_cl = lambda_factor * tau\n\n    Kp = tau \u002F (K * lambda_cl)\n    Ki = Kp \u002F tau\n    Kd = 0.0\n\n    return {\n        \"Kp\": Kp,\n        \"Ki\": Ki,\n        \"Kd\": Kd,\n        \"lambda\": lambda_cl\n    }\n```\n\n## PI Controller Implementation\n\n```python\nclass PIController:\n    def __init__(self, Kp, Ki, setpoint):\n        self.Kp = Kp\n        self.Ki = Ki\n        self.setpoint = setpoint\n        self.integral = 0.0\n\n    def compute(self, measurement, dt):\n        \"\"\"Compute control output.\"\"\"\n        error = self.setpoint - measurement\n\n        # Integral term\n        self.integral += error * dt\n\n        # PI control law\n        output = self.Kp * error + self.Ki * self.integral\n\n        # Clamp to valid range\n        output = max(output_min, min(output_max, output))\n\n        return output\n```\n\n## Expected Closed-Loop Behavior\n\nWith IMC tuning, the closed-loop response is approximately:\n\n```\ny(t) = y_setpoint * (1 - exp(-t \u002F lambda))\n```\n\nKey properties:\n- **Rise time**: ~2.2 * lambda to reach 90% of setpoint\n- **Settling time**: ~4 * lambda to reach 98% of setpoint\n- **Overshoot**: Minimal for first-order systems\n- **Steady-state error**: Zero (integral action eliminates offset)\n\n## Tips\n\n1. **Start conservative**: Use `lambda = tau` initially\n2. **Decrease lambda carefully**: Smaller lambda = larger Kp = faster but riskier\n3. **Watch for oscillation**: If output oscillates, increase lambda\n4. **Anti-windup**: Prevent integral wind-up when output saturates\n",{"name":5,"description":7},"main","0f794a8bb6c27aa9ee4091b2691fa30e4ed9cc8f","data\u002Fpawbench-v1.0\u002Fassets\u002FT131_skillsbench_hvac-control\u002Fskills\u002Fimc-tuning-rules",24,"https:\u002F\u002Fgithub.com\u002Fagentscope-ai\u002FPawBench","https:\u002F\u002Fagentscope-ai.github.io\u002FPawBench\u002F","https:\u002F\u002Fgithub.com\u002Fagentscope-ai","2026-08-03 20:09:53",{"source":8,"skillName":5,"commands":33,"prompt":43,"zipAvailable":44},[34,37,40],{"tool":35,"command":36},"npx","npx skills add agentscope-ai\u002FPawBench --skill imc-tuning-rules",{"tool":38,"command":39},"bunx","bunx skills add agentscope-ai\u002FPawBench --skill imc-tuning-rules",{"tool":41,"command":42},"pnpm","pnpm dlx skills add agentscope-ai\u002FPawBench --skill imc-tuning-rules","安装技能，技能地址：https:\u002F\u002Fgithub.com\u002Fagentscope-ai\u002FPawBench\u002Fblob\u002Fmain\u002Fdata\u002Fpawbench-v1.0\u002Fassets\u002FT131_skillsbench_hvac-control\u002Fskills\u002Fimc-tuning-rules\u002FSKILL.md",true,[],[47,53,60,67,74,81],{"skillNo":48,"name":49,"nameZh":6,"description":6,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":6,"sourceHtmlUrl":50,"categories":51,"keywords":52,"publishedAt":19},"01M72WQCJKBJZB6SC8MMYK3C8R","data-reconciliation","https:\u002F\u002Fgithub.com\u002Fagentscope-ai\u002FPawBench\u002Fblob\u002Fmain\u002Fdata\u002Fpawbench-v1.0\u002Fassets\u002FT140_skillsbench_xlsx-recover-data\u002Fskills\u002Fdata-reconciliation\u002FSKILL.md",[],[],{"skillNo":54,"name":55,"nameZh":6,"description":56,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":6,"sourceHtmlUrl":57,"categories":58,"keywords":59,"publishedAt":19},"01M72WQ8R4DZKK1CT397C0ARQD","hierarchical-taxonomy-clustering","Build unified multi-level category taxonomy from hierarchical product category paths from any e-commerce companies using embedding-based recursive clustering with intelligent category naming via weighted word frequency analysis.","https:\u002F\u002Fgithub.com\u002Fagentscope-ai\u002FPawBench\u002Fblob\u002Fmain\u002Fdata\u002Fpawbench-v1.0\u002Fassets\u002FT139_skillsbench_taxonomy-tree-merge\u002Fskills\u002Fhierarchical-taxonomy-clustering\u002FSKILL.md",[],[],{"skillNo":61,"name":62,"nameZh":6,"description":63,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":6,"sourceHtmlUrl":64,"categories":65,"keywords":66,"publishedAt":19},"01M72WQ1989ZX613G3PY8Z492J","nlp-research-repo-package-installment","Align Python version and repo-declared dependencies (requirements.txt \u002F environment.yml) before installing packages for NLP research code reproduction.","https:\u002F\u002Fgithub.com\u002Fagentscope-ai\u002FPawBench\u002Fblob\u002Fmain\u002Fdata\u002Fpawbench-v1.0\u002Fassets\u002FT138_skillsbench_simpo-code-reproduction\u002Fskills\u002Fnlp-research-repo-package-installment\u002FSKILL.md",[],[],{"skillNo":68,"name":69,"nameZh":6,"description":70,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":6,"sourceHtmlUrl":71,"categories":72,"keywords":73,"publishedAt":19},"01M72WPZ828E746PAB3WCCQARW","state-space-linearization","Linearizing nonlinear dynamics around operating points for control design.","https:\u002F\u002Fgithub.com\u002Fagentscope-ai\u002FPawBench\u002Fblob\u002Fmain\u002Fdata\u002Fpawbench-v1.0\u002Fassets\u002FT137_skillsbench_r2r-mpc-control\u002Fskills\u002Fstate-space-linearization\u002FSKILL.md",[],[],{"skillNo":75,"name":76,"nameZh":6,"description":77,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":6,"sourceHtmlUrl":78,"categories":79,"keywords":80,"publishedAt":19},"01M72WPXFXKWDFVTGAYX0B1TSA","mpc-horizon-tuning","Selecting MPC prediction horizon and cost matrices for web handling.","https:\u002F\u002Fgithub.com\u002Fagentscope-ai\u002FPawBench\u002Fblob\u002Fmain\u002Fdata\u002Fpawbench-v1.0\u002Fassets\u002FT137_skillsbench_r2r-mpc-control\u002Fskills\u002Fmpc-horizon-tuning\u002FSKILL.md",[],[],{"skillNo":82,"name":83,"nameZh":6,"description":84,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":6,"sourceHtmlUrl":85,"categories":86,"keywords":87,"publishedAt":19},"01M72WPW3863555HM0CACRCCFK","integral-action-design","Adding integral action to MPC for offset-free tension tracking.","https:\u002F\u002Fgithub.com\u002Fagentscope-ai\u002FPawBench\u002Fblob\u002Fmain\u002Fdata\u002Fpawbench-v1.0\u002Fassets\u002FT137_skillsbench_r2r-mpc-control\u002Fskills\u002Fintegral-action-design\u002FSKILL.md",[],[]]