An Engine Coolant Temperature Model and Application for Cooling System Diagnosis

Transcription

1 SAE TECHNICAL PAPER SERIES An Engine Coolant Temperature Model and Application for Cooling System Diagnosis In Kwang Yoo, Kenneth Simpson, Myron Bell and Stephen Majkowski Delphi Automotive Systems Reprinted From: Electronic Engine Controls 2000: Modeling, Neural Networks, OBD, and Sensors (SP 1501) SAE 2000 World Congress Detroit, Michigan March 6 9, Commonwealth Drive, Warrendale, PA U.S.A. Tel: (724) Fax: (724)

2 The appearance of this ISSN code at the bottom of this page indicates SAE s consent that copies of the paper may be made for personal or internal use of specific clients. This consent is given on the condition, however, that the copier pay a $7.00 per article copy fee through the Copyright Clearance Center, Inc. Operations Center, 222 Rosewood Drive, Danvers, MA for copying beyond that permitted by Sections 107 or 108 of the U.S. Copyright Law. This consent does not extend to other kinds of copying such as copying for general distribution, for advertising or promotional purposes, for creating new collective works, or for resale. SAE routinely stocks printed papers for a period of three years following date of publication. Direct your orders to SAE Customer Sales and Satisfaction Department. Quantity reprint rates can be obtained from the Customer Sales and Satisfaction Department. To request permission to reprint a technical paper or permission to use copyrighted SAE publications in other works, contact the SAE Publications Group. All SAE papers, standards, and selected books are abstracted and indexed in the Global Mobility Database No part of this publication may be reproduced in any form, in an electronic retrieval system or otherwise, without the prior written permission of the publisher. ISSN Copyright 2000 Society of Automotive Engineers, Inc. Positions and opinions advanced in this paper are those of the author(s) and not necessarily those of SAE. The author is solely responsible for the content of the paper. A process is available by which discussions will be printed with the paper if it is published in SAE Transactions. For permission to publish this paper in full or in part, contact the SAE Publications Group. Persons wishing to submit papers to be considered for presentation or publication through SAE should send the manuscript or a 300 word abstract of a proposed manuscript to: Secretary, Engineering Meetings Board, SAE. Printed in USA

3 An Engine Coolant Temperature Model and Application for Cooling System Diagnosis Copyright 2000 Society of Automotive Engineers, Inc. In Kwang Yoo, Kenneth Simpson, Myron Bell and Stephen Majkowski Delphi Automotive Systems ABSTRACT A coolant temperature model of an internal combustion engine has been formulated to meet the new On-Board Diagnostics II (OBD II) requirement for coolant temperature rationality. The model utilizes information available within the production Engine Control Module (ECM). The temperature prediction capability has been tested for various real-world driving conditions and cycles along with regulated drive cycles. The model can be calibrated to find the appropriate timing for initiation of a diagnostic algorithm for engine cooling system and Coolant Temperature Sensor (CTS) faults. A diagnostic scheme has been developed to detect and isolate various types of cooling system failures using engine soak time information available from a low power timer in the ECM. This model-based fault detection scheme will meet the new California Air Resource Board (CARB) regulations for the high-sided cooling system rationality and allow more accurate and timely repair of system faults at automotive service centers. describes a relatively simple method for modeling engine coolant temperature and the implementation of the associated diagnostic on a production ECM that allows an engine management system to meet the new OBD II requirement. The coolant temperature model is constructed using available information within the production ECM, such as calculated mass air flow, startup coolant temperature, engine speed, vehicle speed, deactivation of fuel, air temperature, etc. MODEL DESCRIPTION A typical internal combustion engine cooling system consists of an engine block (cooling jackets), thermostat, Coolant Temperature Sensor (CTS), radiator, and coolant circulatory pump. Figure 1 shows a schematic diagram of a typical engine cooling sub-system. INTRODUCTION Automobile manufacturers have incorporated logic to disable selected monitoring strategies when the coolant temperature is too high [1]. The CARB is mandating high-sided coolant temperature rationality monitoring to address in situations when these diagnostic disablement strategies prevent other emission related faults from being recorded for prolonged periods of time. Most of the current diagnostic schemes for comprehensive component monitoring of the engine cooling system may not be sufficient to address the new high-sided coolant temperature rationality requirement. Previous studies [2-5] show detail and extensive modeling of each of the engine cooling system components and different measurement technologies. Most of these models are aimed at anticipation of the results of the cooling system components before prototypes are built in a development environment. However, few have focused on coolant temperature modeling for use within the production ECM. This paper Figure 1. Schematic diagram of an engine cooling subsystem An engine cooling system can be viewed as an energy storage device composed of coolant, engine oil, engine block and head. The energy level can be expressed as the temperature of the system. There is typically one source of energy (i.e., combustion energy) in an internal combustion engine and several sources of energy loss. The intake air drawn into the engine is mixed with the injected fuel and combusts inside of the engine. Some of the combustion energy produces engine power and the rest of it are transferred to the ambient air, the engine 1

4 block, oil and engine coolant. Figure 2 shows the energy transfer that occurs in the various engine subsystems. Figure 3. Engine cooling system model structure Figure 2. Energy transfer in the engine cooling system Figure 4 depicts the inputs and output of the engine and radiator coolant temperature models. In order to describe an engine s heat energy accumulation, engine parameters such as the heat transfer rate from the engine to the ambient air, the intake air-flow rate, the engine power production, and the exhaust gas temperature, are necessary. Some of these parameters are available in a production ECM, but other information such as the exhaust gas temperature and the engine power production is generally not available or requires the addition of sensors to obtain their values. A way to avoid using the engine power and the exhaust gas temperature information is to employ a heat rejection concept. Through this heat rejection test, it is possible to find the portion of fuel energy that is transferred to the coolant. Generally, considerable test resources and an adequate environmentally controlled facility are needed to map detailed heat rejection characteristics. The coolant temperature model in this study has been developed to instead use simple tests in the vehicle to determine the fuel energy contribution to heat up the coolant. STRUCTURE OF THE COOLANT TEMPERATURE MODEL CONCEPT The engine cooling system model consists of two base sets of models (Figure 3): one is the engine coolant temperature model and the other is the radiator coolant temperature model. The engine coolant temperature model includes heat rejection energy from injected fuel, heat loss to the ambient air, coolant flow rate and the effect of the thermostat. The radiator coolant temperature model includes radiator heat transfer incorporated with the coolant flow and circulation. These two models interact together as the engine coolant temperature warms up allowing coolant flow to the radiator by opening the thermostat. First, the engine side coolant warms up followed by the coolant at the engine and radiator beginning to mix with thermostat opening. Finally, the coolant continuously circulates through the radiator with thermostat fully open. Figure 4. Inputs and output of the cooling system model FUEL ENERGY SUPPLY The supplied fuel energy to the engine can be approximated as a function of fuel flow rate or as a combination of air mass flow and the commanded fuel to air ratio from the controller. A large portion of the total injected fuel energy is lost through engine power production, hot exhaust gas, and heat transfer through the radiator and engine. The rest of the injected fuel energy contributes to heat up the coolant temperature (i.e. heat rejection to the coolant). The heat rejection coefficient to the engine coolant, η, is simplified as a linear function of calculated mass airflow in this modeling process: the mass air flow could either be measured from a sensor or calculated if a manifold absolute pressure sensor is used. When the commanded nominal air to fuel ratio is lean, the combustion heat energy is limited to the amount of fuel injected. 2

5 Otherwise, the combustion heat energy is limited to the amount of intake air. Knowing the ECM command parameter, i.e. the number of cylinders that are injecting fuel, the total fuel energy can be formulated: For lean air to fuel mixture; Q For rich air to fuel mixture; Where, rej N η N N η N fueling cyl fueling cyl M air M fuel Q LHV ( F / A) com QLHV N 1 fueling Q rej η M air ( ) QLHV Ncyl Stoic (1.1) (1.2) η = apfactor M air + bpfactor (1.3) HEAT LOSS TO AIR The heat loss from the engine block to the ambient air stems mainly from the convection heat transfer. Therefore, the radiation and conduction heat transfer can be ignored: Qair = heng Aeng ( Teng Tair ) (1.4) The coefficient for convection heat-transfer rate, i.e. h eng, is dependent on both vehicle speed and cooling fan operation status. However, it can be simplified and described just as a function of cooling fan-operating status, since vehicle speed has a minimal impact on heat transfer directly from the engine block. THERMOSTAT MODEL The thermostat begins to open when the coolant temperature warms up to a certain level. The thermostat continues to open more up to the point that it is mechanically restricted. The engine coolant flow rate is dependent on the cross sectional area of the thermostat. This opening area can be simplified to a linear relationship of temperature from the thermostat opening coolant temperature to that which causes the opening of thermostat to be at its maximum. for, Teng < Tstat _ min, Athermostat = 0 for, Teng > Tstat _ max, Athermostat = 1 for, T T T, stat _ min eng stat _ max ( Teng Tstat _ min ) Athermostat = ( Tstat _ max Tstat _ min ) (1.5) where, T stat _ min is the engine coolant temperature causing the thermostat begin to open and T stat _ max is the engine coolant temperature for maximum lift of the thermostat. COOLANT FLOW RATE The engine coolant flow rate is primarily dependent upon the engine speed because typical water pumps are directly connected with the 3 crankshaft. Defining C coolflow as a compensation gain for the coolant flow rate, which is dependent on the engine speed, the coolant flow rate can be formulated as; M c = Athermostat Ccoolflow RPM (1.6) HEAT LOSS VIA RADIATOR COOLING SYSTEM The heat loss via the radiator cooling system can be expressed as the temperature difference between the engine outlet and engine inlet points, where the engine outlet temperature is assumed to be the same as the engine coolant temperature. Q rad = M c Cc ( Teng Teng _ in ) (1.7) ENGINE COOLANT TEMPERATURE MODEL From the energy balance, the engine coolant temperature can be formulated using previously defined parameters; Qeng = TengM ecc = Qrej Qair Qrad 1 1 Teng + ( heng Aeng + M ccc ) Teng + ( heng AengTair + M ecc M ecc N fueling 1 M cccteng _ in η M air QLHV ) = (1.8) Ncyl Stoic RADIATOR COOLANT TEMPERATURE MODEL For a normal production vehicle, the coolant temperature at the engine inlet from the radiator is not available. Therefore, the coolant temperature at the inlet of the engine needs to be formulated. From the energy balance between the lost energy through the radiator and the lost energy of coolant, it can be formulated as: Qrad = Qloss = Arad hrad ( Teng Tair ) = M ccc( Teng Teng T eng _ Aradhrad in = Teng ( Teng Tair ) (1.9) M ccc The heat transfer coefficient of the radiator, i.e. h rad, is dependent on the vehicle speed and cooling fan operating status. MODEL VALIDATION The coolant temperature model was developed in MATLAB and SIMULINK initially, then the model was translated into production ECM C code after initial model validation. Taking vehicle data and analyzing it with the SIMULINK model validated the model. The SIMULINK predicted results were verified to match the actual coolant warm-up profile. To verify the production software, additional testing was performed and the coolant model prediction was obtained from the ECM, via the production C software. In addition, the other required model inputs were collected. These inputs were then analyzed with the SIMULINK model. By comparing the ECM predicted temperature to the SIMULINK results, the production software was verified. The model calibration has been done utilizing the Delphi AutoTune ToolBox, which was developed in parallel with the coolant model algorithm. The ToolBox automatically optimizes the calibration data by using the least square method, _ in)

6 and provides the calibration values for the production software. All the data has been measured using a production test vehicle with a 4-cylinder engine equipped with a Delphi engine management system. The coolant temperature model has been validated at various driving and ambient temperature conditions, such as FTP, light load (i.e. idle) and steady state vehicle driving conditions. The FTP type of driving has been performed at different environmental temperature conditions (-30 ~ 45 C) to verify the model validity for different temperature ranges (APPENDIX 1). The model has been tested at various steady state driving conditions in which the vehicle speed (20 ~ 100 KPH) has been kept constant with no transient throttle maneuver (APPENDIX 2). Although the temperature model is quite accurate, some errors exist, stemming from a variety of sources. One of the sources of the model error originates from the simplification of the engine subsystem models to make it a reasonable size for use in a production ECM. Another model error stems from the lack of heater and blower motor status information at the ECM. The coolant temperature model has significant error when the heater and blower are turned on maximum. Having no compensation for the heater core and blower fan causes this error, since the heater and blower status is not available in a typical production ECM. The estimation of the error has been tested at low temperature conditions and light load conditions (APPENDIX 3). It shows that the heater and blower effects are dominant at light load and cold ambient conditions. DIAGNOSTIC ALGORITHM The diagnostic algorithm has been built to detect and isolate engine cooling system-related faults. There are four types of faults that are demonstrated in this paper (Figure 5), including stuck-open thermostat (i.e., too much cooling), stuckclosed thermostat (i.e., lack of cooling), skew-high and skew-low CTS. The diagnostic algorithm depends on the model temperature to decide the initiation point of each of the diagnostics. The algorithm is designed to perform only after a sufficiently long engine off (soak) period to utilize the temperature difference of intake air and engine at start-up as a first indication of the fault type. Then, the diagnostic initiation time point is decided utilizing the model temperature. When each diagnosis is authorized to run, the measured coolant temperature is compared to a certain threshold and the result is compared with the temperature difference of intake air and coolant at start up. With these comparison results, the failure types are distinguishable in the ECM. Figure 5. Structure of the cooling system diagnostic algorithm Limitations on acceptable vehicle driving conditions are imposed to constrain the model error and improve the calibration process. If a vehicle has been driven in a prolonged idle operation or in an extended fuel cut-off engine driving condition, the diagnostics are inhibited. A delay time factor, also, has been implemented before running the actual initiation of the diagnostic algorithm to compensate the model for error caused by heater and blower operation. DIAGNOSTIC ALGORITHM VALIDATION The diagnostic algorithm has been tested with a stuck-open thermostat, and with skew-high and skew-low CTS. The tests have been performed at various environmental temperature conditions and different driving conditions with the failed parts. Detection of each failure is demonstrated in the Appendices (APPENDIX 4 ~ 6) by applying the diagnostic algorithm as explained. Further, the accuracy of the modeled coolant temperature provides for a solid replacement value for an erroneous coolant temperature sensor. Many engine control algorithms rely on an accurate coolant temperature value, including fuel, spark, cooling fan control, and other diagnostics. In the presence of a coolant temperature sensor fault, indicated by the presence of a fault code, the model temperature may be used to substitute for the incorrect sensor value. 4

7 CONCLUSION The coolant temperature model has been designed for the diagnosis of the engine cooling system. Utilizing the coolant temperature model, the developed diagnostic algorithm has a capability of detecting and isolating engine cooling system related failures. Using the model temperature can increase the detection sensitivity and the model can be used for the default temperature if the CTS is found to be faulty. Although categorization of the cooling system related faults were presented, other cooling system related rationality checks could be possible. The diagnostic algorithm, with the coolant temperature model, meets the latest CARB requirements. REFERENCES 1. California Air Resources Board, On-Board Diagnostic II Compliance Guidelines, CARB Mailout #MSC 98-01, Jean-Claude Corbel, An Original Simulation Method for Car Engine Cooling Systems: A Modular System, SAE Technical Paper Series Ngy Srun AP, A Simple Engine Cooling System Simulation Model, SAE Technical Paper Series , Kader A. Fellague, S. Huey Hu, and Donald A. Willoughby, Determination of the Effects of Inlet Air Velocity and Temperature Distributions on the Performance of an Automotive Radiator, SAE Technical Paper Series Edward C. Chiang, George P.C. Huang, Zintal Chang and John H. Johnson, A One Dimensional Transient Compressible Flow Model for Cooling Air Flow Rate Computation, SAE Technical Paper Series J.B. Heywood, Internal Combustion Engine Fundamentals, McGraw-Hill Book Company, New York, Frank P. Incropera, David P. Dewitt, Introduction to Heat Transfer,John Wiley & Sons Inc. New York, CONTACTS Stephen Majkowski is an OBD II systems manager employed by Delphi Automotive Systems, working in Engine Management Systems. His address is steve.majkowski@delphiauto.com. DEFINITIONS, ACRONYMS, ABBREVIATIONS A thermostat : Thermostat opening area coefficient [none] A : Surface area of a radiator [m 2 rad ] A : Surface area of an engine [m 2 eng ] apfactor : Gain factor for heat rejection coefficients [none] bpfactor : Offset factor for heat rejection coefficients [none] CARB: California Air Resource Board CTS: Coolant Temperature Sensor C c : Specific heat for coolant [KJ/kg K] C r : Specific heat for the radiator coolant [kj/kg K] ECM: Engine Control Module ( F / A) com : ECM commanded air to fuel ratio [none] FTP: Federal Test Procedure h : Engine block heat transfer coefficient [kj/sec K m 2 eng ] h : Radiator heat transfer coefficient [kj/sec K m 2 rad ] M e : Effective engine mass [kg] M fuel : Mass flow rate of fuel [kg] M c : Mass flow rate of coolant [kg/sec] N fueling : Number of injector fueling [none] N cyl : Number of cylinder [none] OBD II: On-Board Diagnostics II Q rej : Heat rejection energy flow rate [kj/sec] Q air : Heat energy loss to the ambient [kj/sec] Q rad : Radiator heat transfer rate [kj/sec] Q LHV : Fuel lower heating value [kj/kg] Stoic : Stoichiometric air to fuel ratio [none] T eng : Engine coolant model temperature [ K] : Measured intake air temperature [ K] T air T stat _ min T stat _ max T eng _ out : Minimum temperature for thermostat open [ K] : Maximum temperature of thermostat to max lift [ K] : Coolant temperature input to the radiator [ K] T eng _ in = Tr : Coolant temperature at radiator to engine [ K] η: Heat rejection coefficient [none] In Kwang Yoo is an OBD II systems engineer employed by Delphi Automotive Systems, working in Engine Management Systems. His address is in.kwang.yoo@delphiauto.com. Ken Simpson is an OBD II systems engineer employed by Delphi Automotive Systems, working in Engine Management Systems. His address is ken.simpson@delphiauto.com. Myron Bell is a software engineer employed by Delphi Automotive Systems, working in Engine Management Systems. His address is mick.h.bell@delphiauto.com. 5

8 APPENDIX 1: COOLANT TEMPERATURE MODEL VALIDATION AT FTP Figure A1. Coolant temperature model validation at FTP 6

9 APPENDIX 2: COOLANT TEMPERATURE MODEL VALIDATION AT CONSTANT VEHICLE SPEED Figure A2. Coolant temperature model validation at constant vehicle speed 7

10 APPENDIX 3: MODEL ERROR ESTIMATION WITH HEATER AND BLOWER MAXIMUM OPERATION Figure A3. Model error estimation with heater and blower maximum 8

11 APPENDIX 4: STUCK-OPEN THERMOSTAT FAILURE DIAGNOSIS Figure A4. Stuck-open thermostat failure diagnosis 9

12 APPENDIX 5: SKEW-LOW COOLANT TEMPERATURE SENSOR DIAGNOSIS Figure A5. Skew-low coolant temperature sensor 10

13 APPENDIX 6: SKEW-HIGH COOLANT TEMPERATURE SENSOR DIAGNOSIS Figure A6. Skew-high coolant temperature sensor 11