liu.seSearch for publications in DiVA
Change search
Link to record
Permanent link

Direct link
Publications (10 of 27) Show all publications
Westny, T., Mohammadi, A., Jung, D. & Frisk, E. (2024). Stability-Informed Initialization of Neural Ordinary Differential Equations. In: Neil Lawrence (Ed.), Proceedings of the 41 st International Conference on Machine Learning, Vienna, Austria. PMLR 235, 2024: . Paper presented at International Conference on Machine Learning, 21-27 July 2024, Vienna, Austria (pp. 52903-52914). PMLR, 235
Open this publication in new window or tab >>Stability-Informed Initialization of Neural Ordinary Differential Equations
2024 (English)In: Proceedings of the 41 st International Conference on Machine Learning, Vienna, Austria. PMLR 235, 2024 / [ed] Neil Lawrence, PMLR , 2024, Vol. 235, p. 52903-52914Conference paper, Published paper (Refereed)
Abstract [en]

This paper addresses the training of Neural Ordinary Differential Equations (neural ODEs), and in particular explores the interplay between numerical integration techniques, stability regions, step size, and initialization techniques. It is shown how the choice of integration technique implicitly regularizes the learned model, and how the solver’s corresponding stability region affects training and prediction performance. From this analysis, a stability-informed parameter initialization technique is introduced. The effectiveness of the initialization method is displayed across several learning benchmarks and industrial applications.

Place, publisher, year, edition, pages
PMLR, 2024
Series
Proceedings of Machine Learning Research, ISSN 2640-3498
National Category
Computational Mathematics
Identifiers
urn:nbn:se:liu:diva-210226 (URN)
Conference
International Conference on Machine Learning, 21-27 July 2024, Vienna, Austria
Note

Funding: This research was supported by the Strategic Research Area at Linköping-Lund in Information Technology (ELLIIT) and the Wallenberg AI, Autonomous Systems and Software Program (WASP) funded by the Knut and Alice Wallenberg Foundation. Computations were enabled by the Berzelius resource provided by the Knut and Alice Wallenberg Foundation at the National Supercomputer Centre. The authors would like to thank the reviewers for their insightful comments and suggestions, which have significantly improved the manuscript.

Available from: 2024-12-03 Created: 2024-12-03 Last updated: 2024-12-03Bibliographically approved
Mohammadi, A., Krysander, M. & Jung, D. (2022). Analysis of grey-box neural network-based residuals for consistency-based fault diagnosis. In: : . Paper presented at 11th IFAC Symposium on Fault Detection, Supervision and Safety for Technical Processes SAFEPROCESS 2022. Pafos, Cyprus, 8-10 June 2022 (pp. 1-6). Elsevier, 55(6)
Open this publication in new window or tab >>Analysis of grey-box neural network-based residuals for consistency-based fault diagnosis
2022 (English)Conference paper, Published paper (Refereed)
Abstract [en]

Data-driven fault diagnosis requires training data that is representative of the different operating conditions of the system to capture its behavior. If training data is limited, one solution is to incorporate physical insights into machine learning models to improve their effectiveness. However, while previous works show the usefulness of hybrid approaches for isolation of faults, the impact of training data must be taken into consideration when drawing conclusions from data-driven residuals in a consistency-based diagnosis framework. By giving an understanding of the physical interaction between the signals, a hybrid fault diagnosis approach, can enforce model properties of residual generators to isolate faults that are not represented in training data. The objective of this work is to analyze the impact of limited training data when training neural network-based residual generators. It is also investigated how the use of structural information when selecting the network structure is a solution to limited training data and how to ameliorate the performance of hybrid approaches in face of this challenge.

Place, publisher, year, edition, pages
Elsevier, 2022
Series
IFAC papers online, E-ISSN 2405-8963 ; 6
Keywords
Grey-box recurrent neural networks, structural analysis, fault diagnosis, machine learning, model-based diagnosis, anomaly classification
National Category
Electrical Engineering, Electronic Engineering, Information Engineering
Identifiers
urn:nbn:se:liu:diva-188245 (URN)10.1016/j.ifacol.2022.07.097 (DOI)000858756200001 ()
Conference
11th IFAC Symposium on Fault Detection, Supervision and Safety for Technical Processes SAFEPROCESS 2022. Pafos, Cyprus, 8-10 June 2022
Available from: 2022-09-07 Created: 2022-09-07 Last updated: 2022-10-20
Frisk, E., Jarmolowitz, F., Jung, D. & Krysander, M. (2022). Fault Diagnosis Using Data, Models, or Both – An Electrical Motor Use-Case. In: : . Paper presented at 11th IFAC Symposium on Fault Detection, Supervision and Safety for Technical Processes SAFEPROCESS 2022. Pafos, Cyprus, 8-10 June 2022 (pp. 533-538). Elsevier, 55(6)
Open this publication in new window or tab >>Fault Diagnosis Using Data, Models, or Both – An Electrical Motor Use-Case
2022 (English)Conference paper, Published paper (Refereed)
Abstract [en]

With trends as IoT and increased connectivity, the availability of data is consistently increasing and its automated processing with, e.g., machine learning becomes more important. This is certainly true for the area of fault diagnostics and prognostics. However, for rare events like faults, the availability of meaningful data will stay inherently sparse making a pure data-driven approach more difficult. In this paper, the question when to use model-based, data-driven techniques, or a combined approach for fault diagnosis is discussed using real-world data of a permanent magnet synchronous machine. Key properties of the different approaches are discussed in a diagnosis context, performance quantified, and benefits of a combined approach are demonstrated.

Place, publisher, year, edition, pages
Elsevier, 2022
Series
IFAC papers online, E-ISSN 2405-8963
Keywords
fault diagnosis, model-based diagnosis, data-driven diagnosis, sparse data
National Category
Electrical Engineering, Electronic Engineering, Information Engineering
Identifiers
urn:nbn:se:liu:diva-188246 (URN)10.1016/j.ifacol.2022.07.183 (DOI)000884499400003 ()
Conference
11th IFAC Symposium on Fault Detection, Supervision and Safety for Technical Processes SAFEPROCESS 2022. Pafos, Cyprus, 8-10 June 2022
Available from: 2022-09-07 Created: 2022-09-07 Last updated: 2022-12-06
Skarman, F., Gustafsson, O., Jung, D. & Krysander, M. (2020). Acceleration of Simulation Models Through Automatic Conversion to FPGA Hardware. In: 2020 30th International Conference on Field-Programmable Logic and Applications (FPL): . Paper presented at 30th International Conference on Field-Programmable Logic and Applications (FPL), Gothenburg, Sweden, 31 Aug.-4 Sept. 2020 (pp. 359-360). IEEE
Open this publication in new window or tab >>Acceleration of Simulation Models Through Automatic Conversion to FPGA Hardware
2020 (English)In: 2020 30th International Conference on Field-Programmable Logic and Applications (FPL), IEEE , 2020, p. 359-360Conference paper, Published paper (Refereed)
Abstract [en]

By running simulation models on FPGAs, their execution speed can be significantly improved, at the cost of increased development effort. This paper describes a project to develop a tool which converts simulation models written in high level languages into fast FPGA hardware. The tool currently converts code written using custom C++ data types into Verilog. A model of a hybrid electric vehicle is used as a case study, and the resulting hardware runs significantly faster than on a general purpose CPU.

Place, publisher, year, edition, pages
IEEE, 2020
Keywords
FPGA, High Level Synthesis, Dynamic Programming, Hybrid Electric Vehicles
National Category
Computer Engineering
Identifiers
urn:nbn:se:liu:diva-171274 (URN)10.1109/FPL50879.2020.00068 (DOI)000679186400056 ()9781728199023 (ISBN)9781728199030 (ISBN)
Conference
30th International Conference on Field-Programmable Logic and Applications (FPL), Gothenburg, Sweden, 31 Aug.-4 Sept. 2020
Available from: 2020-11-12 Created: 2020-11-12 Last updated: 2021-08-27Bibliographically approved
Jung, D., Dong, Y., Frisk, E., Krysander, M. & Biswas, G. (2020). Sensor selection for fault diagnosis in uncertain systems. International Journal of Control, 93(3), 629-639
Open this publication in new window or tab >>Sensor selection for fault diagnosis in uncertain systems
Show others...
2020 (English)In: International Journal of Control, ISSN 0020-7179, E-ISSN 1366-5820, Vol. 93, no 3, p. 629-639Article in journal (Refereed) Published
Abstract [en]

Finding the cheapest, or smallest, set of sensors such that a specified level of diagnosis performance is maintained is important to decrease cost while controlling performance. Algorithms have been developed to find sets of sensors that make faults detectable and isolable under ideal circumstances. However, due to model uncertainties and measurement noise, different sets of sensors result in different achievable diagnosability performance in practice. In this paper, the sensor selection problem is formulated to ensure that the set of sensors fulfils required performance specifications when model uncertainties and measurement noise are taken into consideration. However, the algorithms for finding the guaranteed global optimal solution are intractable without exhaustive search. To overcome this problem, a greedy stochastic search algorithm is proposed to solve the sensor selection problem. A case study demonstrates the effectiveness of the greedy stochastic search in finding sets close to the global optimum in short computational time.

Place, publisher, year, edition, pages
Taylor & Francis, 2020
Keywords
Fault diagnosis, fault detection and isolation, sensor selection
National Category
Electrical Engineering, Electronic Engineering, Information Engineering Computer Engineering
Identifiers
urn:nbn:se:liu:diva-117176 (URN)10.1080/00207179.2018.1484171 (DOI)000525971000025 ()
Note

The previous status of this article was Manuscript.

Available from: 2015-04-21 Created: 2015-04-21 Last updated: 2021-12-28Bibliographically approved
Jung, D. & Sundström, C. (2019). A Combined Data-Driven and Model-Based Residual Selection Algorithm for Fault Detection and Isolation. IEEE Transactions on Control Systems Technology, 27(2), 616-630
Open this publication in new window or tab >>A Combined Data-Driven and Model-Based Residual Selection Algorithm for Fault Detection and Isolation
2019 (English)In: IEEE Transactions on Control Systems Technology, ISSN 1063-6536, E-ISSN 1558-0865, Vol. 27, no 2, p. 616-630Article in journal (Refereed) Published
Abstract [en]

Selecting residual generators for detecting and isolating faults in a system is an important step when designing model-based diagnosis systems. However, finding a suitable set of residual generators to fulfill performance requirements is complicated by model uncertainties and measurement noise that have negative impact on fault detection performance. The main contribution is an algorithm for residual selection that combines model-based and data-driven methods to find a set of residual generators that maximizes fault detection and isolation performance. Based on the solution from the residual selection algorithm, a generalized diagnosis system design is proposed where test quantities are designed using multivariate residual information to improve detection performance. To illustrate the usefulness of the proposed residual selection algorithm, it is applied to find a set of residual generators to monitor the air path through an internal combustion engine.

Place, publisher, year, edition, pages
IEEE, 2019
Keywords
Automotive applications, change detection algorithms, fault detection, fault diagnosis, machine learning.
National Category
Control Engineering
Identifiers
urn:nbn:se:liu:diva-149582 (URN)10.1109/TCST.2017.2773514 (DOI)000457619300014 ()
Note

Funding agencies: Volvo Car Corporation Gothenburg Sweden

Available from: 2018-07-08 Created: 2018-07-08 Last updated: 2019-02-20Bibliographically approved
Jung, D., Ahmed, Q. & Rizzoni, G. (2018). Design Space Exploration for Powertrain Electrification using Gaussian Processes. In: 2018 Annual American Control Conference (ACC): . Paper presented at American Control Conference (pp. 846-851).
Open this publication in new window or tab >>Design Space Exploration for Powertrain Electrification using Gaussian Processes
2018 (English)In: 2018 Annual American Control Conference (ACC), 2018, p. 846-851Conference paper, Published paper (Refereed)
Abstract [en]

Design space exploration of hybrid electric vehicles is an important multi-objective global optimization problem. One of the main objectives is to minimize fuel consumption while maintaining satisfactory driveability performance and vehicle cost. The design problem often includes multiple design options, including different driveline architectures and component sizes, where different candidates have various trade-offs between different, in many cases contradictory, performance requirements. Thus, there is no global optimum but a set of Pareto-optimal solutions to be explored. The objective functions can be expensive to evaluate, due to time-consuming simulations, which requires careful selection of which candidates to evaluate. A design space exploration algorithm is proposed for finding the set of Pareto-optimal solutions when the design search space includes multiple design options. As a case study, powertrain optimization is performed for a medium-sized series hybrid electric delivery truck.

Keywords
Gaussian processes;hybrid electric vehicles;Pareto optimisation;power transmission (mechanical);hybrid electric vehicles;multiobjective global optimization problem;satisfactory driveability performance;vehicle cost;Pareto-optimal solutions;powertrain optimization;medium-sized series hybrid electric delivery truck;powertrain electrification;Linear programming;Space exploration;Mechanical power transmission;Gaussian processes;Optimization;Hybrid electric vehicles;Fuels
National Category
Energy Engineering
Identifiers
urn:nbn:se:liu:diva-151301 (URN)10.23919/ACC.2018.8430899 (DOI)
Conference
American Control Conference
Available from: 2018-09-17 Created: 2018-09-17 Last updated: 2018-09-17
Tamilarasan, S., Jung, D. & Guvenc, L. (2018). Drive Scenario Generation Based on Metrics for Evaluating an Autonomous Vehicle Controller. In: SAE Technical Paper: . Paper presented at SAE World Congress. SAE International
Open this publication in new window or tab >>Drive Scenario Generation Based on Metrics for Evaluating an Autonomous Vehicle Controller
2018 (English)In: SAE Technical Paper, SAE International , 2018Conference paper, Published paper (Refereed)
Abstract [en]

An important part of automotive driving assistance systems and autonomous vehicles is speed optimization and traffic flow adaptation. Vehicle sensors and wireless communication with surrounding vehicles and road infrastructure allow for predictive control strategies taking near-future road and traffic information into consideration to improve fuel economy. For the development of autonomous vehicle speed control algorithms, it is imperative that the controller can be evaluated under different realistic driving and traffic conditions. Evaluation in real-life traffic situations is difficult and experimental methods are necessary where similar driving conditions can be reproduced to compare different control strategies. A traditional approach for evaluating vehicle performance, for example fuel consumption, is to use predefined driving cycles including a speed profile the vehicle should follow. However, if the vehicle speed is part of the vehicle control output, a different vehicle evaluation framework is necessary. Here, speed constraints are defined based on route and traffic conditions, such as speed limits, traffic signs and signals, and the locations of surrounding vehicles. Hence, route generation is an important task for evaluating speed control algorithms. A route is a distance-based description of the road conditions and locations of traffic signs and signals. A driving scenario is defined as a route which also includes information about traffic density and the location of surrounding traffic as function of time. It is discussed how driving scenarios can be used to evaluate and compare different speed control algorithms. The generation of driving scenarios is performed in two steps, route generation and traffic data generation. First, two approaches are discussed for generating the route conditions, such as varying speed limits and locations of traffic signals, either using real road map data or to recreate from vehicle speed data. In a second step, traffic conditions are simulated using the software SUMO to generate speed profiles of surrounding vehicles on the road. To assure that the selected driving scenarios represent varying driving conditions, a set of metrics is selected and used for driving scenario selection.

Place, publisher, year, edition, pages
SAE International, 2018
National Category
Control Engineering
Identifiers
urn:nbn:se:liu:diva-149583 (URN)10.4271/2018-01-0034 (DOI)
Conference
SAE World Congress
Available from: 2018-07-08 Created: 2018-07-08 Last updated: 2018-09-17
Oruganti, P. S., Ahmed, Q. & Jung, D. (2018). Effects of Thermal and Auxiliary Dynamics on a Fuel Cell Based Range Extender. In: SAE Technical Paper: . Paper presented at SAE World Congress. SAE International
Open this publication in new window or tab >>Effects of Thermal and Auxiliary Dynamics on a Fuel Cell Based Range Extender
2018 (English)In: SAE Technical Paper, SAE International , 2018Conference paper, Published paper (Refereed)
Abstract [en]

Batteries are useful in Fuel Cell Hybrid Electric Vehicles (FCHEV) to fulfill transient demands and for regenerative braking. Efficient energy management strategies paired with optimal powertrain design further improves the efficiency. In this paper, a new methodology to simultaneously size the propulsive elements and optimize the power-split strategy of a Range Extended Battery Electric Vehicle (REBEV), using a Polymer Electron Membrane Fuel Cell (PEMFC), is proposed and preliminary studies on the effects of the driving mission profile and the auxiliary power loads on the sizing and optimal performance of the powertrain design are carried out. Dynamic Programming is used to compute the optimal energy management strategy for a given driving mission profile, providing a global optimal solution. The component sizing problem is performed using a machine learning based, guided design space exploration to find the set of Pareto-optimal solutions that give the best trade-offs between the different objectives. The powertrain model includes the dynamic behavior of the fuel cell system compressor and a battery lumped parameter thermal model along with the quasi-static semi-empirical model of the fuel cell and a zero-order battery model. Initial results indicate an increase in the Pareto-optimal sizes with the inclusion of thermal management.

Place, publisher, year, edition, pages
SAE International, 2018
National Category
Energy Engineering
Identifiers
urn:nbn:se:liu:diva-151299 (URN)10.4271/2018-01-1311 (DOI)
Conference
SAE World Congress
Available from: 2018-09-17 Created: 2018-09-17 Last updated: 2018-09-17
Jung, D., Ahmed, Q., Zhang, X. & Rizzoni, G. (2018). Mission-based Design Space Exploration for Powertrain Electrification of Series Plugin Hybrid Electric Delivery Truck. In: WCX World Congress Experience: . Paper presented at SAE World Congress. SAE International
Open this publication in new window or tab >>Mission-based Design Space Exploration for Powertrain Electrification of Series Plugin Hybrid Electric Delivery Truck
2018 (English)In: WCX World Congress Experience, SAE International , 2018Conference paper, Published paper (Refereed)
Abstract [en]

Hybrid electric vehicles (HEV) are essential for reducing fuel consumption and emissions. However, when analyzing different segments of the transportation industry, for example, public transportation or different sizes of delivery trucks and how the HEV are used, it is clear that one powertrain may not be optimal in all situations. Choosing a hybrid powertrain architecture and proper component sizes for different applications is an important task to find the optimal trade-off between fuel economy, drivability, and vehicle cost. However, exploring and evaluating all possible architectures and component sizes is a time-consuming task. A search algorithm, using Gaussian Processes, is proposed that simultaneously explores multiple architecture options, to identify the Pareto-optimal solutions. The search algorithm is designed to carefully select the candidate in each iteration which is most likely to be Pareto-optimal, based on the results from previous candidates, to reduce computational time. The powertrain of a medium-sized series plugin hybrid electric delivery truck with a range extender is optimized for different driving missions. Three different powertrain architectures are included in the design space exploration and the fuel economy is evaluated using a simulation model of the powertrain and Dynamic Programming. Results from the analysis show which ranges of powertrain component sizes are recommended for the different types of driving scenarios.

Place, publisher, year, edition, pages
SAE International, 2018
National Category
Energy Engineering
Identifiers
urn:nbn:se:liu:diva-151300 (URN)10.4271/2018-01-1027 (DOI)
Conference
SAE World Congress
Available from: 2018-09-17 Created: 2018-09-17 Last updated: 2018-09-17
Organisations
Identifiers
ORCID iD: ORCID iD iconorcid.org/0000-0003-0808-052X

Search in DiVA

Show all publications