冷冻空调自动控制- 自控系统设计实例.ppt

冷凍空調自動控制- 自控系統設計實例 Power Consumption Control for Optical Fiber Communication System Modeling The dominate heat transfer is Radiation P : TEC (Thermal Electric Cooler) 功率控制 T :光通訊模組溫度 System Modeling mc : 光通訊模組質量與比熱 P : TEC (Thermal Electric Cooler) 功率控制 該值並已減去光通訊模組額定發熱量 T : 光通訊模組溫度 To: 環境溫度 System Modeling Controller Design Ideal System Performance Real System Requirement Direct 1/f Noise Spectral Density Control Noise Spectral Density Control 結論 * * 李達生 Focusing here… 概論 自動控制理論發展 自控系統設計實例 Laplace Transform 冷凍空調自動控制 控制系統範例 控制元件作動原理 控制系統除錯 自動控制理論 系統穩定度分析 系統性能分析 PID Controller 自動控制實務 節能系統控制 訊號擷取系統 訊號雜訊處理 快速溫控系統 With the same stability, power consumption just half of the traditional one. Traditional Design Diode Laser DWDM Signal Output Signal Monitoring New Design Temperature Effects on Laser Diode Wavelength Shift ~ 0.3 nm/ oC High Precision Temperature Control is Necessary TEC Thermal Control Module is Widely Employed 0.01 oC Temperature Control is Achievable but Expensive Beam Point Stability ~ 1 mrad/ oC 多波分工器(DWDM, Dual Wavelength Demultiplex Module)與半導體雷射(Diode Laser)訊號源整合溫控系統模組之設計 Radiation Signal Output Signal Monitoring T P TEC Shielding Const Set T(t) = T(t)-To We got 0 Open Loop Transfer Function Assuming Aluminum material A proportional and Integral controller can easily achieve robust control P Kp + KI/s Plant a T e b P T e b 0.63 0.37 (under damping) 20 sec Steady State Error less than 1% ~ 0.1 degC P T e b +/-0.2 degC Stable in 0.01 degC The target can’t be achieved by the traditional control method!! PI Control can achieve relative stable after ts stable time. Then the 0.01 degC control will follow the Impulse Response generate noise power spectrum density MCU I2C Bus According to the spectral density, nano degC resolution could be possible. However, limited by the A/D, 10-3 degC resolution is obtained. MCU I2C Bus 結論 實際測試結果由於作為量測基準之熱電偶精度不足，無法證實溫控系統可達成 +/- 0.01 oC之測試規範 經三小時穩定溫控，熱電偶讀數波動始終低於其額定噪音範圍內，應可視為溫控系統能達成控制要求