Control Concepts for DH Compact Stations

Similar documents
Dynamic simulation of DH house stations

How to avoid pressure oscillations in district heating systems

B1-1. Closed-loop control. Chapter 1. Fundamentals of closed-loop control technology. Festo Didactic Process Control System

REDUCING PROCESS VARIABLITY BY USING FASTER RESPONDING FLOWMETERS IN FLOW CONTROL

Multifunctional Thermostatic Circulation Valve MTCV

Improving the Control System for Pumped Storage Hydro Plant

Variable swirl diffusers

Analyzing Control Problems and Improving Control Loop Performance

PRO-CLEAN STRATIFIED TANK

CM 3310 Process Control, Spring Lecture 21

Alireza Mousavi Brunel University

Technology that counts

ENERGY DISTRIBUTION ANALYSIS IN A LOW HEAD FRANCIS TURBINE DURING THERMODYNAMIC EFFICIENCY MEASUREMENTS

Fundamentals of Ground Source GTHP Pump Technology

Dynamic performance of a district heating system in Madumvej, Denmark

Simulation Study on Pressure Control using Nonlinear Input/Output Linearization Method and Classical PID Approach

IMPROVED TECHNIQUE OF MULTI-STAGE COMPENSATION. K. M. Yanev A. Obok Opok

Air Cooler Industry AC-LN 1-7 / ACA-LN 2-7 / ACAF-LN 2-7

VSMP SSPMP SSIMF. We will first explain how the motion starts (onset) and then investigate the dynamics of this phenomenon.

MECHANICS AND CONTROL OF PUMPING A PLAYGROUND SWING AND ROBOTIC IMPLEMENTATION ABSTRACT INTRODUCTION NOMENCLATURE

Technology that counts

SIMULATION SUITE CHEMCAD SOFTWARE PROCESS CONTROL SYSTEMS PROCESS CONTROL SYSTEMS COURSE WITH CHEMCAD MODELS. Application > Design > Adjustment

ATT 310 USER MANUAL

Available online at ScienceDirect. Procedia Engineering 106 (2015 ) Dynamics and Vibroacoustics of Machines (DVM2014)

Control Loop Investigations

Gesellschaft für Schwerionenforschung mbh (GSI), Planckstrasse 1, D Darmstadt, Germany

Dr. OligoTM DNA / RNA / OLIGO SYNTHESIZERS

Joseph D. Galloway II Robotics and Motion Laboratory, Dept. of Mechanical Engineering San Antonio, TX, USA Gerardo Aaron Rios

Modelling the Dynamics of Flight Control Surfaces Under Actuation Compliances and Losses

881 Compact IC pro 882 Compact IC plus. Compact ion chromatography for intelligent, reliable and precise analysis

NEGST. New generation of solar thermal systems. Advanced applications ENEA. Comparison of solar cooling technologies. Vincenzo Sabatelli

Aspen Polymers. Conceptual design and optimization of polymerization processes

3D Simulation of the Plunger Cooling during the Hollow Glass Forming Process Model, Validation and Results

Stability Performance Dilemma in Hydronic Radiators with TRV

NR ROTARY RING TABLE: FLEXIBLE IN EVERY RESPECT

Determination of Dynamic Model of Natural Gas Cooler in CPS Molve III Using Computer

Design and analysis of the prototype of boiler for steam pressure control

flexotherm 230V Heat pump technical specification

405 Compact Orifice Series and 1595 Conditioning Orifice Plate Flow Test Data Book and Flow Handbook

Check-Q-meter. Table of contents. Features. Functions. RE 27551/ /10 Replaces: Type FD

Chapter 3. Regression-Based Models for Developing Commercial Demand Characteristics Investigation

CHAPTER 3 QUARTER AIRCRAFT MODELING

405 Compact Orifice Series and 1595 Conditioning Orifice Plate Flow Test Data Book and Flow Handbook

Technology that counts

Water terminal control solutions

405 Compact Orifice Series and 1595 Conditioning Orifice Plate Flow Test Data Book and Flow Handbook

Research Collection. Thermal issues in 5-axis machine tools. Conference Paper. ETH Library

VDN2.. VEN2.. VUN2.. Radiator valves. Building Technologies HVAC Products ACVATIX TM. NF norm, for 2-pipe heating systems

Chapter 3. State Feedback - Pole Placement. Motivation

COMPARISON OF MEASURED AND ANALYTICAL PERFORMANCE OF SHELL-AND-TUBE HEAT EXCHANGERS COOLING AND HEATING SUPERCRITICAL CARBON DIOXIDE

Laird Thermal Systems Application Note. Precise Temperature Control for Medical Analytical, Diagnostic and Instrumentation Equipment

MECHANICS OF PNEUMATIC ARTIFICIAL MUSCLE

UNCERTAINTY IN TORQUE CALIBRATION USING VERTICAL TORQUE AXIS ARRANGEMENT AND SYMMETRICAL TWO FORCE MEASUREMENT

CHAPTER 1 Basic Concepts of Control System. CHAPTER 6 Hydraulic Control System

Interactivity-Compatibility. TA Hydronic College Training

ClimateMaster s Tranquility 22. Raj Hiremath Director of Residential Marketing

HiLumi LHC FP7 High Luminosity Large Hadron Collider Design Study. Milestone Report. Cryogenic Scenarios for the Cold Powering System

Bruker Daltonics. EASY-nLC. Tailored HPLC for nano-lc-ms Proteomics. Nano-HPLC. think forward

1/54 Circulation pump, safety valve, expansion vessel

AERODROME METEOROLOGICAL OBSERVATION AND FORECAST STUDY GROUP (AMOFSG)

Variable Displacement Plug-In Motor A6VE. Series 6, for open and closed circuits Axial tapered piston, bent axis design

Design of Decentralised PI Controller using Model Reference Adaptive Control for Quadruple Tank Process

GT-POWER linearization and engine advanced control design applications

Kjeltec 2300 Analyzer Unit

R E F R I G E R A T I O N A N D A I R C O N D I T I O N I N G 2 December 2008 / 1

SRV02-Series Rotary Experiment # 1. Position Control. Student Handout

Latest generation testo air velocity & IAQ probes with cable handle. Instruction manual

ADAM PIŁAT Department of Automatics, AGH University of Science and Technology Al. Mickiewicza 30, Cracow, Poland

Energy flows and modelling approaches

P1 REVISION CHAPTER 1a Energy Tfr by Heating

SVSM User Training and Example Sequences

Quanser NI-ELVIS Trainer (QNET) Series: QNET Experiment #02: DC Motor Position Control. DC Motor Control Trainer (DCMCT) Student Manual

Professional Portfolio Selection Techniques: From Markowitz to Innovative Engineering

Model Predictive Controller of Boost Converter with RLE Load

= A x (t) + B utt), by d{ _-=-=-

Motion II. Goals and Introduction

Comparison of discharge measurements - Thermodynamic to US Clamp- On, stationary US and Needle Opening Curve

Fundamental study on simple quantitative approach of damping performance for semi-active damper

Cooler-Bypass Thermostat Valve, Size 10

FAVIMAT (AI)ROBOT2 FAVIGRAPH. Automatic Single-Fibre Testers

Eliminating Oscillations in TRV-Controlled Hydronic Radiators

Simply better pipetting!

ESTIMATION OF THE DOWNSTREAM FLOW OF VALVES. J.Prost, K.Käfer Institute for Waterpower and Pumps Vienna University of Technology 1040 Wien, Austria

Lecture 25: Tue Nov 27, 2018

Thermostatic expansion valves TUA/TUAE

Educational Objectives

Active elastomer components based on dielectric elastomers

Thermostatic expansion valves TUB/TUBE and TUC/TUCE

CARBON COMPLIANCE. Setting an appropriate limit for zero carbon new homes. National or Regional Weather: Implications for Carbon Compliance

COMPARISON OF TWO METHODS TO SOLVE PRESSURES IN SMALL VOLUMES IN REAL-TIME SIMULATION OF A MOBILE DIRECTIONAL CONTROL VALVE

Residual Chlorine Monitor

Tradeoffs and Limits of Performance

MASSACHUSETTS INSTITUTE OF TECHNOLOGY Department of Mechanical Engineering 2.04A Systems and Controls Spring 2013

Analysis and Experiments of the Linear Electrical Generator in Wave Energy Farm utilizing Resonance Power Buoy System

INTRODUCTION TO THE SIMULATION OF ENERGY AND STORAGE SYSTEMS

Thermo Scientific ConFlo IV Universal Interface. Continuous Flow Interface. Isotope Ratio MS

Certification of a High Capacity Force Machine for Testing of Load Cells According to OIML R60

LASER INTERFEROMETER GRAVITATIONAL WAVE OBSERVATORY - LIGO - CALIFORNIA INSTITUTE OF TECHNOLOGY MASSACHUSETTS INSTITUTE OF TECHNOLOGY

OPERATING MANUAL. EIB-Kombisensor AS 315 N

Simple circuits - 3 hr

Transcription:

ECHNICAL PAPER Control Concepts for DH Compact Stations Published in Euroheat and Power IIII 2004 English edition 11 Q 1 Q 2 22 12 21 D I S R I C H E A I N G A C A D E M Y

Herman Boysen Product Application Manager, Danfoss A/S District Heating Controls DK-6430 Nordborg Denmark elephone: +45 7588 4123 Fax: +45 7449 0395 boy@danfoss.com 11 Q 1 Q 2 21 22 12 Jan Eric horsen Application Centre, Danfoss A/S District Heating Controls DK-6430 Nordborg Denmark elephone: +45 7588 4494 Fax: +45 7449 0950 jet@danfoss.com Fig. 1 hermostatic controller, proportional controller, and a parallel and serial coupling of those 1. ABSRAC As a result of a wide range of functional requirements to DH house stations and flat stations, a variety of control concepts are commercially available on the market today. he specific behaviour, or control performance, of those concepts vary significantly and are depending on many factors. his article describes the basic functionality of four control concepts, a thermostatic controller, a proportional controller and a parallel and serial coupling of those. he focus is on plate heat exchanger control, and the investigation is based on dynamic simulations, i.e. on models verified up against practical measurements. Specific cases, but also the general behaviour of the concepts are described. Control performance is investigated regarding stability, temperature overshoots, stationary temperature deviations and settling time. Finally, some recommendations are made in regard to selecting the control concept depending on demands and supply conditions. 2. INRODUCION he trend is towards small compact multifunctional controllers, requiring lower system prices, improved control functions, space saving and improved flexibility rate. he most common functions to be integrated or combined are thermostatic control, proportional control, differential pressure control, idle control and DHW priority control. Both self-acting and electric controls and combinations of those are used [1/2]. o investigate functionality and potential of control components and concepts, dynamic simulation has been applied. 3. MODELLING OF CONPONENS he computational software tool used for this investigation is the well-known Simulink program package, by he MathWorks Inc. [3]. Simulink is used for modelling, simulating and analysing of dynamic systems. he principles for the component models are described in [4/5/6]. Here the presentation is limited to the thermostatic controller. he thermostatic controller model is based on a force balance, including the actuator bellow element, the springs and the O-ring friction, as shown in fig. 2. A linear valve characteristic is implemented. he first order thermal time constant, as a function of the flow rate, is included in the model. Bellow spring Fig. 2 Schematic presentation of thermostatic controller O-ring 4. CONROL CONCEPS INVESIGAED Concept 1: Heat exchanger control by thermo- static controller Concept 2: Heat exchanger control by propor- tional controller Bellow element Valve seat Concept 3: Heat exchanger control by parallel coupling of thermostatic and pro- portional controller Concept 4: Heat exchanger control by serial coupling of thermostatic and pro- portional controller o compare the results some identical boundary conditions are used for all four concepts. Set point Spring / valve spring D I S R I C H E A I N G A C A D E M Y 2

dp control valve = 1 bar, 11 = 65 C and 90 C, 21 =10 C. he simulations are to be regarded as specific cases, intending to investigate the concept specific behaviours. 4.1 Heat Exchanger Control by hermostatic Controller Fig. 3 and 4 show results for this concept. A linear valve characteristics is implemented, with a 22 deviation, disregarding Q 2 = 100 l/h of approx. 6 C for the 11 = 65 C case. For the 11 =90 C the value becomes approx. 4 C, which is explained by the lower primary flow rate needed, resulting in lower P-deviation. Normally 100 l/h is a low value for tapping flow. he general 22 temperature range, reflecting 11 = 65 C to 90 C and disregarding Q 2 = 100 l/h, is approx. 8 C. Fig. 3 Simulation result for thermostatic controller, 11 = 65 C Looking at peak temperatures, this concept has the highest peak; see fig. 4 at time 800 sec. Here the effect of low flow, Q 2, around the sensor, resulting in slow sensor becomes visible. Furthermore, there is no feed forward signal, like e.g. a proportional controller, to react immediately. Comparing the settling time, this concept has the longest in general. Especially looking at a step response where the system is only marginally stable, see fig. 4, time interval 500 to 600 sec. All P-controllers, like a thermostatic linear valve, will become unstable at a certain low value of Q 2, due to the loop-gain, d 22 /dkv-valve, goes towards infinity when Q 2 goes to zero. Fig. 4 Simulation result for thermostatic controller, 11 = 90 C 4.2 Heat Exchanger Control by Proportional Controller his concept is stable for all values of Q 2. Looking at the temperature deviation for 22, but disregarding the lowest Q 2 tapping steps of 100 l/h, the result is approx. 4 C for 11 = 65 C and for 11 = 90 C. he influence from hysteresis on the proportional controller is relatively more visible at low tapping flows compared to higher flows. Furthermore, the general 22 temperature level is significantly different comparing 11 = 65 C ( 22 level about 42 C) and 90 C ( 22 level about 58 C), resulting in a temperature Fig. 5 Simulation result for proportional controller, 11 = 65 C and 22 for 11 = 90 C D I S R I C H E A I N G A C A D E M Y 3

range of approx. 20 C, Q2 = 100 l/h disregarded. Simulation results for this concept are shown in fig. 5. Peak temperatures are relatively low and the settling time relatively short. he actual values are highly dependent on the dynamics of the heat exchanger, e.g. thermal capacity, which becomes more dominant at low flow ranges. 4.3 Heat Exchanger Control by Parallel Coupling of hermostatic and Prop. Controller Fig. 6 and 7 show simulation results for this concept. Since there are two valves in parallel, the thermostatic valve model implemented has half the capacity compared to concept 1. he proportional valve contribution is slightly lower compared to concept 2. At 11 = 90 C the main part of the flow belongs to the proportional controller. At 11 = 65 C the flow split is nearer to half and half. Benefits of this concept are the relatively small tap temperature deviation 22 in combination with the wide stable Q 2 tap flow range. Disregarding tap flows of Q 2 = 100 l/h, the deviation is approx. 5 C at 11 = 65 C and 2 C at 11 = 90 C. he general 22 temperature range, reflecting 11 = 65 C to 90 C, disregarding Q 2 = 100 l/h, is approx. 7 C. he situation of a totally closed thermostatic valve is shown in fig. 7 time range 600 to 700 sec. Here the proportional controller is determining the tap temperature, which becomes high due to the high proportional controller flow contribution. If the flow part of the thermostatic valve is small, oscillations will occur. But only the smaller part of the flow will oscillate, which results in insignificant tap temperature peak to peak values, see fig. 7 time range 0-100 sec. Fig. 6 Simulation result for thermostatic contr. parallel with prop. controller for 11 = 65 C Fig. 7 Simulation result for thermostatic contr. parallel with prop. controller for 11 = 90 C Considering temperature peaks and settling time, this concept performs better than concept 1. Regarding tap temperature range in general, this concept is performing likely or slightly better than concept 1 and clear better than concept 2. Comparing peaks and settling time to concept 2, the results depend on what step is regarded. Looking at fig. 7 and 5 at time 300 sec. setting time Fig. 8 Simulation result for thermostatic contr. serial with prop. controller for 11 = 65 C D I S R I C H E A I N G A C A D E M Y 4

time of concept 3. is shorter, since the thermostat gives a short boost on the primary flow. Looking at fig. 5 and 6 at time 600 sec. the settling time of concept 2 is shorter. he oscillation of the thermostatic valve is the explanation. In general, temperature peaks are of the same range where comparable. he fact, that the initial tap temperature is as different prior to some of the steps, a comparison is not straightforward to do in all cases. 4.4 Heat Exchanger Control by Serial Coupling of hermostatic and Prop. Controller Since this concept puts the proportional controller and thermostatic controller in series, the capacities of each one should be increased compared to the other concepts. he simulation results for this concept are shown in fig. 8 and 9. Looking at the temperature deviation for 22, but disregarding the lowest Q 2 tapping steps of 100 l/h, the result is approx. 7 C for 11 = 65 C and approx. 4 C for 90 C. he general 22 temperature range, reflecting 11 = 65 C to 90 C, disregarding Q 2 = 100 l/h, is approx. 10 C. Concept 4 gives a wider stability range compared to the thermostatic controller (concept 1), since the proportional controller acts as a limiter for the thermostatic controller gain, dkv/d 22, principally resulting in a larger P-band for the thermostatic controller. Most effect of reducing the thermostatic controller gain is, however, achieved at the lower supply temperature 11 = 65 C. At higher supply temperatures or higher supply differential pressure the damping effect of the proportional actuator decreases. he concept opens the possibility for decreasing the tap temperature range by increasing the proportional controller contribution. But the consequences are a reduced Q 2 flow range where stability is obtained at the high supply temperature For the 11 = 65 C case, there is almost no oscillation, compared to concept 1, fig. 3, which indicates the effect of the serial connected proportional actuator. At 11 = 90 C the oscillations are becoming more dominant, i.e. compared to concept 3, see fig. 7. he reason is that the Fig. 9 Simulation result for thermostatic contr. serial with prop. controller for 11 = 90 C able 1: Control characteristics for the four concepts thermostatic controller capacity becomes more dominant. Due to the serial coupling, concept 4 is able to control variations in supply differential pressure, without high 22 tap temperatures, which are drawbacks of concept 2 and 3. Regarding peak temperatures and settling times, this concept is to be ranged between concept 1 and 3. 5. CONROL PERFORMANCE SUMMARY In table 1, the general characteristics of the concepts are summarized. he concept potentials to handle variations in supply differential pressure is included, and stated in column dp range. he conclusion should be seen in relation to how each concept performs when designed for covering a typical primary temperature range as specified. 6. RECOMMANDAIONS As shown in the simulation cases, the four control concepts have their specific behaviour regarding heat exchanger control. If the supply conditions are constant, 11 and dp control valve the proportional controller concept is performing good in a simple way. If the supply temperature only is changing concepts 1, 3 and 4 are relevant. In this case concept 3 gives the best control performance followed by concept 4. If the supply differential pressure is changing only, concepts 1 and 4 are relevant, where concept 4 give the best control performance. If the proportional controller is equipped with a setting adjustment possibility, a manual seasonal adjustment due to varying supply conditions is a well-known compromise. o improve the control performance in all four cases at changing supply differential pressure, a differential pressure controller could be introduced. Besides this, the differential pressure controller introduces a flow limiting function, introducing hydraulic balance in the distribution system. As to stand-by operation (no tapping of DHW) the proportional controller in concept 2 and 4 provides immediately zero primary flow, which is an advantage in case where the scaling issue is relevant. In this case a bypass thermostat with D I S R I C H E A I N G A C A D E M Y 5

a relative low set point (~ 40-45 C) could keep the system warm at stand-by. On the opposite the concepts 1 and 4 give higher comfort (taping temperature reaches set point faster) when tapping after a long stand-by periode. 7. CONCLUSION his paper describes by means of case simulations the control performance of four concepts. Depending on the supply conditions and the demand to control performance, different concepts are to be preferred. here is no top of the line concept. he success depends on the demands and supply conditions. Due to this, the different concepts have their market share today. In practical DH house or flat station cases, where specific requirements are to be fulfilled and documented, laboratory tests are normally performed on prototype DH stations. ypical focus is on control performance, capacity considerations, heat loss and stand-by operation. References [1] L. Happe, Danfoss AVI Multifunctional Compact Controller, Offprint EuroHeat&Power, No 5009, September 2002. [2] Danfoss AB valve, Presented 2004 at Poznan International Installation Fair (INSALACJE 2004). [3] Simulink, Dynamic System Simulation for MALAB, Version 4, Users Guide, he Math Works, Inc. 2000. [4] Herman Boysen & Jan Eric horsen. Control Concepts for DH Compact Stations Investigated by Simulations. Presented at 9th International Symposium on District Heating and Cooling, Finland, August 2004. [5] J.E. horsen, Dynamic Simulations of DH House Stations, 7. Dresdener Fernwärme-Kolloquium, Germany 2002. Printed in EuroHeat&Power June 2003. [6] Atli Benonysson & Herman Boysen. Optimum control of heat exchangers. Presented at 5th International symposium on automation of district heating systems, Finland, August 1995 D I S R I C H E A I N G A C A D E M Y 6

Notes D I S R I C H E A I N G A C A D E M Y 7

For further information please contact: Danfoss A/S, District Heating Controls Att.: Mr. Herman Boysen, DK-6430 Nordborg, Phone +45 74 88 41 23, Fax +45 74 49 09 49, E-mail:boy@danfoss.com Notes D I S R I C H E A I N G A C A D E M Y Danfoss A/S, DK-6430 Nordborg, Denmark, Phone: +45 7488 2222, elefax: +45 7449 0949, www.danfoss.com VF.KA.L1.02 Produced by Danfoss