Applications Junqiang (James) Fan Fellow, Systems and Controls - PowerPoint PPT Presentation

Control Design and Verification with Physics Based Models for HVAC/R Applications Junqiang (James) Fan Fellow, Systems and Controls Engineering Sept 28, 2016

OUTLINE  Vapor compression refrigeration cycle  Model Based Control Development Process  Application Examples  Transportation Refrigeration  Commercial Refrigeration  Residential HVAC  Commercial Building HVAC  Conclusions 2

VAPOR COMPRESSION REFRIGERATION CYCLE 3

WHAT IS CONTROL OF HVAC/R? Reliably operating HVAC/R systems to be functional and energy efficient Control HVAC/R Plant Heat pump Fan-coil unit Equipment Increasing Complexity PID+ logic Measurements Supermarket refrigeration Condenser Condenser Condenser Fresh food Rack Fresh food Rack Fresh food Rack Space heating Space heating Space heating Controls Controls Controls Sanitary hot Sanitary hot Sanitary hot water water water Cold room evap. Cold room evap. Cold room evap. coordinated Systems PIDs+ logic Frozen food Rack Frozen food Rack Frozen food Rack Cold room evap. Cold room evap. Cold room evap. Multideck Multideck Multideck Island Island Island Serve Serve Serve - - - over over over Actuations Combi Combi Combi - - - freezer freezer freezer Large-scale systems Controller (building parameters /campus level) What’s important? What’s important? • • Control architecture & algorithm design Know the physics, systems objectives and limitations • • Implementation and test/verification Model the physics, component to system • • Tuning and commissioning System complexity 4 • Operation & upgrading

MODEL BASED CONTROL DEVELOPMENT PROCESS From requirements definition to field support Requirements Field upgrades and configuration Internet Hardware/software updates Modeling and simulation Operation Diagnostics and fault detection Symptom 1 Symptom 2 Symptom 3 Symptom 4 Symptom N Control design Commissioning V apor compresso r C ont rol er I nvert er syst em and Box heat ers - Verification and validation Software-in-the-loop Tuning guideline Rapid prototyping, Hardware-in-the-loop 5

APPLICATION EXAMPLES Equipment Pulsor TM : Truck Refrigeration Equipment Developed control architecture and algorithm for robust system performance and optimal efficiency Systems Condenser Condenser Condenser Control Fresh food Rack Fresh food Rack Fresh food Rack Controlled Controlled 3 Space heating Space heating Space heating Controls Controls Controls variable 1 variable 2 variable 3 Sanitary hot Sanitary hot Sanitary hot water water water 2.5 Cold room evap. Cold room evap. Cold room evap. CO2OLtec TM : Supermarket Refrigeration System 2 60% Frozen food Rack Frozen food Rack Frozen food Rack 1.5 66% 1 Developed control commissioning guidelines in use 39% Cold room evap. Cold room evap. Cold room evap. 0.5 Multideck Multideck Multideck by Carrier installers 0 Island Island Island Serve Serve Serve - - - over over over Default New Default New Default New Combi Combi Combi - - - tuning tuning tuning tuning tuning tuning freezer freezer freezer Infinity NG TM : Residential HVAC System 84 Zone3 SP 82 Zone temp, o F Zone3 temp Demonstrated HW-independent, model based 80 Zone4 SP Zone4 temp developed control algorithm on scalable SW platform 78 76 74 19 20 21 22 23 24 Time, hr Large Systems/Buildings Supervisory control algorithm : 10% to 15% energy consumption reduction. 6

PULSOR ™… TRUCK REFRIGERATION Architecture and algorithm design Verification and Modeling and Validation Algorithm design Requirements Simulation Rapid prototyping Discharge Pressure [bar] 35 30 25 20 15 10 Setpoint 5 0 0 1 2 3 4 5 6 7 8 Suction Pressure [bar] Active constraint control algorithm  Eliminated cycling Operating constraints  Better performance Product No control algorithm  Small (~kW) capacity changes during field trials  Air-cooled, standard vapor compression system  Single-input-multiple-output control (Hybrid control solution) 7

CO 2 OLTEC™… SUPERMARKET REFRIGERATION Faster and accurate system commissioning Commissioning Control analysis and design Requirements Modeling and guidelines Simulation Condenser Fresh food rack Fresh food Rack Controlled variable 1 Sanitary hot Space heating Space heating water Controls Controls Controls 70 Sanitary hot Cold room evap. water 60 Setpoint Frozen food rack 50 Control tuning 0 50 100 150 200 Cold Controlled variable 2 room evap. instructions 36 Multideck Island 34 Island Setpoint Serve over - Serve - over Combi - 32 Combi - 0 50 100 150 200 freezer freezer Product Controlled Control variable 1 Controlled  Large (~100kW) capacity variable 3 variable 2 3 2.5  CO 2 -based refrigeration system 2 60% 1.5 66%  Multiple-input-multiple-output control 1 39% 0.5 (100’s control loops) 0 Default New Default New Default New 2010 After Before After Before After Before tuning tuning tuning tuning tuning tuning  Site-specific configuration CCS using transitioned SW 8

CO 2 OLtec™: Gas Cooler Modeling More physics captured by 2-D cross-flow HX model versus 1-D counter flow HX model at reasonable cost of simulation speed 2-D Gas Cooler Model Front view Side view 9

INFINITY NG … RESIDENTIAL HVAC Software architecture and system control design System control algorithm Model-based control algorithm development Appl. SW Appl. SW Appl. SW Appl. SW comp. 1 comp. 3 comp. N comp. 5 Appl. SW Appl. SW comp. 2 comp. 4 Requirements Layered base New programming model software architecture HW resource mapping Data Dictionary Automatically generated code 2012 Final Code Product … Application SW User Interface Field trial results Automatically generated code 84 Zone3 SP 82 Zone temp, o F  North American Zone3 temp 80 Zone4 SP residential application Zone4 temp 78  Multiple-input-multiple-output System Control 76 control Control 74 algorithm 0 1 2 3 4 5 19 20 21 22 23 24  Large variety of configurations Time (hr) No control algorithm changes during field trials Hardware/software separation 10

INTEGRATED WHOLE-BUILDING HVAC MODEL Inputs 3 rd Floor Ret . Air Mass Flow, Individual Zone Temp. Temp., RH • Weather & Schedules Water-Side Sup. Controls (PI) Key Outputs Pressure & Temp. • AHU3 Chiller Plant Eqp. Power, Flow, Temp. Floor 3 • AHU Fan Power & 3 rd Floor Sup . Air Mass Flow, SAT Valve Pos. Temp., RH Control (PI) • Zone Temp., RH. 2 nd Floor Ret . Air Mass Flow, Chilled-Water Ret. Temp., RH Pressure & Temp. Water-Side Sup. Pressure & Temp. Chiller Plant Building AHU2 Floor 2 2 nd Floor Sup . Air Mass Flow, Temp., RH 1 st Floor Ret . Air Mass Flow, Temp., RH Water-Side Sup. Pressure & Temp. AHU1 CHWST CWST Floor 1 1 st Floor Sup . Air Mass Flow, DP Control (PI) Temp., RH 11

SUMMARY OF CASE STUDIES 8 Test Profiles 4 Case Configurations Web-Bulb Temp. (each case config.) Case Configurations Definitions Test Cases Test Case Scenarios Case Configuration 1 Medium Office + Primary-Only Chiller Plant Configuration Test 1 Miami Summer 1 Test 2 Miami Shoulder Case Configuration 2 Medium Office + Primary-Sec. Chiller Plant Configuration Test 3 Las Vegas Summer 2 Test 4 Las Vegas Shoulder Case Configuration 3 Large Hotel + Primary-Only Chiller Plant Configuration Test 5 Baltimore Summer 3 Test 6 Baltimore Shoulder Case Configuration 4 Large Hotel + Primary-Sec. Chiller Plant Configuration Test 7 Chicago Summer 4 Test 8 Chicago Shoulder 4 Chiller Plant Control Algorithms Control Algorithms Descriptions 1. Baseline Control Constant chilled-water supply temp. (CHWST) setpoint of 7°C. Load based chiller staging logic. 2. OAT-Based Reset A linear schedule to reset CHWST setpoint based on outdoor air temperature (ASHRAE 90.1). Load based chiller (ASHRAE 90.1) staging logic. 3. Heuristic-Based Trim-Respond logic resets CHWST setpoint based on the demand measured by AHU’s chilled-water valve (Trim-Respond) position . One request is generated when one chilled-water valve position becomes greater than a prescribed threshold (e.g., 90%). Load based chiller staging logic. 4. Low-Cost Optimal Maximize CHWST setpoint while performing real-time load estimation . Load based chiller staging logic. 12

LOW-COST OPTIMAL CONTROL Average Energy Savings (%) from Low-Cost Optimal Control 18.1 ~10% (hotel) ~15% (office) 12 11.5 6.4 Case Config. 1 (Office-PriOnly) Case Config. 2 (Office-PriSec ) Case Config. 3 (Hotel-PriOnly) Case Config. 4 (Hotel-PriSec) 13

CONCLUSIONS Better performing and more robust products  Physics based dynamic modeling and control enabling  Control architecture (actuation/sensing) trade-off analysis  Algorithm analysis and design  Installation/commissioning guidelines development  Software robustness testing  Equipment diagnostics development  No turn-backs or surprises after the products are developed/deployed 14