Case-based adaptation for automotive engine electronic control unit calibration

doi:10.1016/j.eswa.2009.09.063

Expert Systems with Applications

Volume 37, Issue 4, April 2010, Pages 3184-3194

https://doi.org/10.1016/j.eswa.2009.09.063 Get rights and content

Abstract

The automotive engine performance is greatly affected by the calibration of its electronic control unit (ECU). The method for ECU calibration is traditionally done by trial-and-error. This traditional method consumes a large amount of time and money. To resolve this problem, case-based reasoning (CBR) is employed, so that an existing and effective ECU setup can be adapted to fit another similar class of engines. The adaptation procedure is done through a more sophisticated step called case-based adaptation (CBA) (Craw et al., 2001, Craw et al., 2006, Leake et al., 1996, Leake et al., 1997). CBA is an effective knowledge management tool, which can interactively learn the expert adaptation knowledge. The paper briefly reviews the methodologies of CBR and CBA. Then the application to ECU calibration is described via a case study. With CBR and CBA, the efficiency of calibrating an ECU can be enhanced. A prototype system has also been developed to verify the usefulness of CBR in ECU calibration.

Introduction

Modern automotive engines are controlled by the electronic control unit (ECU). The engine performance, such as power, torque, brake specific fuel-consumption and emission level, is significantly affected by the setup of control parameters in the ECU. Parameterization of electronic control unit (ECU) software is a major milestone in the development process for modern automotive engines. This process is known as ECU calibration or ECU tune-up. Fig. 1 shows a screenshot of calibration process in an ECU software. ECU calibration is engine dependent. In other words, an ECU setup is only valid on the same engine model. Traditionally, the ECU calibration is done by the vehicle manufacturer. However, in recent years, the programmable ECU (Fig. 2) and ECU read only memory (ROM) editors have been widely adopted by many performance cars. These devices allow the non-factory engineers to tune-up their engines according to different add-on components and driver’s requirements, and create business for aftermarket automotive industry.

Current practice of engine tune-up relies on the experience of the automotive engineer, who will handle a huge number of combinations of engine control parameters and carry out many engine testes on the dynamometer according to different combinations of engine control parameters. The relationship between the input and output parameters of a modern car engine is a complex multivariable nonlinear function, which is very difficult to be found (Li, 2005). Consequently, engine tune-up is usually done by trial-and-error method. This spends a large amount of time and money. In industrial practice, many automotive engineers like to tune-up an engine by referring to an existing base map, which is obtained from the past setup of a similar engine or the same engine. The parameters of the base map are then adjusted to fit different performance requirements of the same engine, or even to fit another but similar engine. This practice exactly fulfills the working environment of case-based reasoning (CBR), i.e., based on a retrieved similar case, adaptation is then performed to fit different new situations. So, a case-based reasoning and adaptation framework for computer-aided ECU calibration is presented in this study.

Section snippets

Case-based reasoning

Case-based reasoning (Armengol, 2007, Kolodner, 1993, Pal and Shiu, 2004) is a simple problem-solving paradigm that involves matching a current problem against similar problems that were successfully solved in the past. The process can be augmented by adapting solutions so that they can match the current problem more closely. There are many examples for CBR applications, e.g., an auto mechanic who fixes an engine by recalling another car that exhibited similar symptoms, a lawyer who advocates a

Case-based adaptation (CBA)

Adaptation in CBR sounds easy but very difficult to implement because no general rules can cover all situations even in a very specific domain. The only way to acquire adaptation knowledge is only done by consulting human domain experts for their ways to handle different problems (Seifert, 2005). However, even the human domain experts may not answer precisely and accurately how the problems can be handled. With the emergence of CBA, the previous difficulties could be alleviated. The general

Application to ECU calibration

In modern automotive engines, a lot of engine performance is affected by the control parameters in the ECU, such as power performance, idle speed performance and emission performance. As the scope of the problem domain is very wide, this project selects an engine ECU calibration for aftermarket power performance tune-up to demonstrate the effectiveness of the CBR and CBA methods. The engine power performance is usually expressed as a power curve against speeds. The engine power at specific

Experiments and results

A complete ECU calibration for a new engine model usually takes one to two years. As a matter of fact, the experienced engineer in HONDA B16A engine still spends at least half year to tune-up a similar model B18C engine, because a lot of trial-and-error on ECU setups and dynamometer tests are still required. In addition, ECU calibration is similar to planning and design domains where explicit knowledge is hard to explain and managed. Therefore the learning curve of ECU calibration for a novice

Conclusions

This paper describes the techniques of CBR and CBA applied in automotive ECU calibration. Although CBR was proposed and continually formulated since 25 years ago, a good and feasible adaptation technique for general case had not been established because adaptation knowledge is highly domain-specific and experience-driven. Hence, adaptation is traditionally a difficult task. In the research, CBA has been successfully applied to alleviate the difficulty. From the perspective of automotive ECU

Acknowledgements

The research is supported by the Science and Technology Development Fund of Macau, Grant 019/2007/A, and the University of Macau Research Grants RG057/08-09S/VCM/FST, and UL011/09-Y1/EME/WPK01/FST.

References (23)

EK. Burke et al.
Structured cases in case-based reasoning–reusing and adapting cases for timetabling problems
Knowledge-Based Systems
(2000)
S. Craw et al.
Learning adaptation knowledge to improve case-based reasoning
Artificial Intelligence
(2006)
E. Armengol
Usages of generalization in case-based reasoning
R. Bergmann et al.
Towards a new formal model of transformational adaptation in case-based reasoning
S. Craw et al.
Learning and applying case-based adaptation knowledge
F. Gebhardt et al.
Reasoning with complex cases
(1997)
H. Kelbassa
Optimal case-based refinement of adaptation rule bases for engineering design
J. Kolodner
Case-based reasoning
(1993)
D. Leake et al.
Acquiring case adaptation knowledge: A hybrid approach
D. Leake et al.
A case study of case-based CBR

G.Y. Li

Application of intelligent control and MATLAB to electronically controlled engines

(2005)

Cited by (17)

A framework for simulation-based engine-control unit inspection in manufacturing phase
2017, Control Engineering Practice
Citation Excerpt :
However, ECUs in the manufacturing phase also need the verification process in the unified way of the two ECU components by three major causes: 1) embedded system revision, 2) incomplete installation of an embedded system, and 3) quality variation of chipsets. The first cause originates from continuous ECU revision that is to calibrate an embedded system (Vong & Wong, 2010). Numerous embedded system versions can be generated during the revision process, and it is impossible to validate an ECU like verification in the development phase against every change (Boot, Richert, & Rukgauer, 1999).
This paper proposes a framework for electronic engine-control unit (ECU) inspection in manufacturing phase. Although various methodologies have been developed for the ECU verification at the development phase, the ECU verification in the manufacturing phase has rarely been brought into focus. ECUs in the manufacturing phase need the verification process in the unified way of the ECU software and hardware components by three major causes: 1) ECU software revision, 2) incomplete installation of software, and 3) quality variation of hardware. For the effective ECU verification in the manufacturing phase, we propose a simulation-based ECU testing methodology. The proposed framework configures virtual vehicle environment to simulate an ECU using a “virtual engine system (VES) model” that specifies the operations of every ECU function during a simulation. The proposed framework employs a reverse engineering approach to identify the operation state transition of an ECU function by reference vehicle states from vehicle experimental data. The main objective of this paper is to design a VES model-based inspection system that simulates an ECU without software modification in brief time for set-up and execution. The proposed framework has been implemented and successfully applied to a Korean automotive company.
Two-layer random forests model for case reuse in case-based reasoning
2015, Expert Systems with Applications
Citation Excerpt :
Manual reuse approach is executed by experts based on their experience; hence, this approach depends on subjective decisions, lacks reliability, and takes too much time (Jin, Cho, Hyun, & Son, 2012). For automatic case reuse, some researchers focused on knowledge acquisition, e.g., in Lee (2003), associative rules were extracted using data mining techniques, and in Vong and Wong (2010), case-based adaptation scheme was used to adapt cases. Some researchers used statistical approaches to adapt cases, such as k-nearest neighbor approach, multiple regression analysis (MRA) based approach (Jin et al., 2012), artificial neural network (ANN) based approach (Lotfy & Mohamed, 2002), genetic algorithm based approach (Mat Jani & Lee, 2008), and grey relational analysis (Hu et al., 2015).
Case reuse is important for case-based reasoning (CBR) because without it, a CBR system degrades to a case retrieval system, and a retrieved case generally cannot solve a problem directly. To improve the accuracy of case reuse, a two-layer random forests model is proposed and the framework of the corresponding CBR system is presented. First, clustering analysis is used to organize the cases in the case base, and gravitational self-organizing mapping algorithm is adopted to automatically detect the cluster number on the basis of the structure of the data. Then, a two-layer model scheme is proposed to model the mapping in every cluster. In this scheme, the first layer model obtains the pre-estimate of the output feature value of the query case, and the second layer model models the error of the pre-estimate, which is used to decrease the error generated by the first layer model. Random forests algorithm, which is a popular ensemble learning model, is adopted as a model in the two layers to improve accuracy and stability. Several benchmark datasets are used to validate the proposed two-layer model scheme, and the results demonstrate that it can improve the case reuse accuracy and stability. The proposed two-layer random forests model is applied to hydraulic generator design, and the results confirm that the proposed model is effective for case reuse.
Automatic shape adaptation for parametric solid models
2015, CAD Computer Aided Design
Citation Excerpt :
Some modern 3D cad systems have provided a similar adaptation mechanism in geometric design stage [70], where a part is set to be under-constrained, based on domain knowledge, and then the automatic shape adaptation of the part can be carried out in an assembly circumstance based on its mating relationships determined by human intervention. Vong et al. [71] employ case-based reasoning (CBR) to adapt an existing and effective ECU setup to fit another similar class of engine.
Adaptation, as is well known, plays a fundamental role in Case-Based Design. However, after decades of efforts, automatic adaptation approach is still rare. In common design works, the first thing one will usually do is choosing a start-up model (a candidate model) of moderate complexity based on a simple query model possessing primary design constraints. To enable the candidate model to smartly adapt its shape to that of the query model according to the embedded constraints, a novel automatic shape adaptation approach is proposed in this paper. First, to determine the corresponding faces between two non-preregistered models as relevant elements, a shape frame concept and its quantitative descriptor are defined. Second, to unify the representation of seemingly different but inherently consistent dimensions, a promotion method is adopted. Third, based on the corresponding faces and the promoted dimension representation, the corresponding dimensions between the two parametric solid models are identified. Finally, the parametric information is smoothly transferred from the query model to the candidate model as design constraints, and the shape of the candidate model is automatically adapted to the query model. Besides that, a prototype system is also implemented to verify the effectiveness of the automatic shape adaptation approach.
Integrating gray system theory and logistic regression into case-based reasoning for safety assessment of thermal power plants
2012, Expert Systems with Applications
Citation Excerpt :
As for the applications of CBR, the research achievements are more fruitful. CBR has been applied widely to health care, engineering design, classification, prediction, recommendation, technologies optimization, organizational behavior science, social learning, and many other fields, while some successful application instances have been achieved (Golosnoy & Okhrin, 2008; Marion et al., 2008; Owen, 2000; Sinha & May, 1996; Vong & Won, 2010). Meanwhile, infused with other methods, such as genetic algorithms, data mining, artificial neural networks, fuzzy set theory, and others, CBR is extended widely to wider research fields (Chang, Lai, & Robert Lai, 2006; Golobardes, Llorà, Salamó, & Martí, 2002; Zhuang, Churilov, Burstein, & Sikaris, 2009).
Safety assessment of thermal power plants (TPPs) is one of the important means to guarantee the safety of production in thermal power production enterprises. Due to various technical limitations, existing assessment approaches, such as analytic hierarchy process (AHP), Monte Carlo methods, artificial neural network (ANN), etc., are unable to meet the requirements of the complex security assessment of TPPs. Currently, most of the security assessments of TPP are completed by the means of experts’ evaluations. Accordingly, the assessment conclusions are greatly affected by the subjectivity of the experts. Essentially, the evaluation of power plant systems relies to a large extent on the knowledge and length of experience of the experts. Therefore in this domain case-based reasoning (CBR) is introduced for the security assessment of TPPs since this methodology models expertise through experience management. Taking the management system of TPPs as breakthrough point, this paper presents a case-based approach for the Safety assessment decision support of TPPs (SATPP). First, this paper reviews commonly used approaches for TPPs security assessment and the current general evaluation process of TPPs security assessment. Then a framework for the Management System Safety Assessment of Thermal Power Plants (MSSATPP) is constructed and an intelligent decision support system for MSSATPP (IDSS-MSSATPP) is functionally designed. IDSS-MSSATPP involves several key technologies and methods such as knowledge representation and case matching. A novel case matching method named Improved Gray CBR (IGCBR) has been developed in which a statistical approach (logistic regression) and Gray System theory are integrated. Instead of applying Gray System theory directly, it has been improved to integrate it better into CBR. In addition this paper describes an experimental prototype system of IDSS-MSSATPP (CBRsys-TPP) in which IGCBR is integrated. The experimental results based on a MSSATPP data set show that CBRsys-TPP has high accuracy and systematically good performance. Further comparative studies with several other common classification approaches also show its competitive power in terms of accuracy and the synergistic effects of the integrated components.
A case-based knowledge system for safety evaluation decision making of thermal power plants
2012, Knowledge-Based Systems
Citation Excerpt :
Different from ANNs and decision trees, it can address the problem of over fitting. Combined with other intelligent reasoning techniques (such as rule-based reasoning) [24], CBR has been applied widely to health care, engineering design, classification, prediction, recommendation, technologies optimization, organizational behaviour science, social learning, and numerous other fields, while a great many successful application instances have also been achieved [25–29]. In the field of evaluation research, there are also many articles concerning CBR, such as the CBR applications to software cost estimation [30], software effort estimation [31], risk assessment in audit judgment [32], risk analysis for electronic commerce [33], web break sensitivity evaluation in a paper machine [34], safety risk analysis in information safety systems [35], safety evaluation of process configuration [36], and so forth.
Safety assessment of thermal power plants (TPP) is an important means to ensure the safety of production in thermal power production enterprises. Modern information technology can play an important role in TPP safety assessment. The evaluation of power plant systems relies, to a large extent, on the knowledge and experience of the experts undertaking the task. Case-based reasoning (CBR) is introduced for the safety assessment of TPP since it models expertise through experience management. This paper provides a case-based approach for the Management System safety assessment decision making of TPP (MSSATPP). We introduce a case matching method named CBR-Grey, which integrates the Delphi approach and grey system theory. Based on this method, we implement a prototype of case-based knowledge system (CBRSYS-TPP) for the evaluation decision making of the panel of experts. Our experimental results based on a real-world TPP safety assessment data set show that CBRSYS-TPP has high accuracy and systematically good performance.
Case-based expert system using wavelet packet transform and kernel-based feature manipulation for engine ignition system diagnosis
2011, Engineering Applications of Artificial Intelligence
Citation Excerpt :
CBR is also an approach to incremental learning, since a new experience is retained each time a problem has been solved, making it immediately available for future problems. The CBR field has grown rapidly over the last two decades, as seen by its increased share of papers and successful applications in many practical problems (Kolodner, 1993; Perner, 2002; Perner, 2009; Kowalski et al., 2005; Vong and Wong, 2010). A typical CBR system is composed of four sequential steps (Fig. 8):
Whenever there is any fault in an automotive engine ignition system or changes of an engine condition, an automotive mechanic can conventionally perform an analysis on the ignition pattern of the engine to examine symptoms, based on specific domain knowledge (domain features of an ignition pattern). In this paper, case-based reasoning (CBR) approach is presented to help solve human diagnosis problem using not only the domain features but also the extracted features of signals captured using a computer-linked automotive scope meter. CBR expert system has the advantage that it provides user with multiple possible diagnoses, instead of a single most probable diagnosis provided by traditional network-based classifiers such as multi-layer perceptions (MLP) and support vector machines (SVM). In addition, CBR overcomes the problem of incremental and decremental knowledge update as required by both MLP and SVM. Although CBR is effective, its application for high dimensional domains is inefficient because every instance in a case library must be compared during reasoning. To overcome this inefficiency, a combination of preprocessing methods, such as wavelet packet transforms (WPT), kernel principal component analysis (KPCA) and kernel K-means (KKM) is proposed. Considering the ignition signals captured by a scope meter are very similar, WPT is used for feature extraction so that the ignition signals can be compared with the extracted features. However, there exist many redundant points in the extracted features, which may degrade the diagnosis performance. Therefore, KPCA is employed to perform a dimension reduction. In addition, the number of cases in a case library can be controlled through clustering; KKM is adopted for this purpose. In this paper, several diagnosis methods are also used for comparison including MLP, SVM and CBR. Experimental results showed that CBR using WPT and KKM generated the highest accuracy and fitted better the requirements of the expert system.

View all citing articles on Scopus

¹: Tel.: +853 83974458; fax: +853 28838314.

View full text

Case-based adaptation for automotive engine electronic control unit calibration

Abstract

Introduction

Section snippets

Case-based reasoning

Case-based adaptation (CBA)

Application to ECU calibration

Experiments and results

Conclusions

Acknowledgements

Knowledge-Based Systems

Artificial Intelligence

Usages of generalization in case-based reasoning

Towards a new formal model of transformational adaptation in case-based reasoning

Learning and applying case-based adaptation knowledge

Reasoning with complex cases

Optimal case-based refinement of adaptation rule bases for engineering design

Case-based reasoning

Acquiring case adaptation knowledge: A hybrid approach

A case study of case-based CBR

Application of intelligent control and MATLAB to electronically controlled engines