Kalevi Huhtala

Kalevi Huhtala

Auto-generated profile, based on publicly available information

Kalevi Huhtala

Unit of Automation Technology and Mechanical Engineering, Tampere University, Tampere, Finland

Research Focus & Metrics

Academic Career Length

26Years (2001-Present)

(Calculated once Scilit publication data is synced: current year − year of first publication + 1)

Representative Publications

View All Publications
6 August 2018 in Journal of Marine Engineering & Technology
Verified
Journal Article
Closed access
MJMiikka Jaurola
Miikka Jaurola
AHAnders Hedin
Anders Hedin
STSeppo Tikkanen
Seppo Tikkanen
KHKalevi Huhtala
Kalevi Huhtala
The conventional marine vessel power systems typically have the potential to improve their fuel consumption and their emissions. This can be done by redesigning the system configuration, the machinery and the power management strategy. The addition of options in power management allows for the running of individual power sources closer to their optimal operating point. However, this immediately raises questions about how to redesign the system and how to operate it to maximise the benefits. The information needed to answer these questions is often scattered around separate sectors of the marine industry. The system integrator needs to be able to combine the complex dependencies of these individual sectors to formulate the big picture that describes the whole power system. Numerical optimisation algorithms provide solutions to develop methodologies to solve multi-variable and potentially multi-objective problems. This literature review presents the authors' findings of design and power management optimisation cases in marine vessel power systems.

30 April 2015 in IEEE Transactions on Vehicular Technology
Verified
Journal Article
Closed access
JCJi-Wung Choi
Ji-Wung Choi
KHKalevi Huhtala
Kalevi Huhtala
This paper proposes a new efficient path-planning algorithm for articulated steering vehicles operating in semi-structured environments, in which obstacles are detected online by the vehicle's sensors. The first step of the algorithm is offline and computes a finite set of feasible motions that connect discrete robot states to construct a search space. The motion primitives are parameterized using Bézier curves and optimized as a nonlinear programming problem (NLP) equivalent to the constrained path planning problem. Applying the A* search algorithm to the search space produces the shortest paths as a sequence of these primitives. The sequence is drivable and suboptimal, but it can cause unnatural swerves. Therefore, online path smoothing, which uses a gradient-based method, is applied to solve another NLP. Numerical simulations demonstrate that performance of the proposed algorithm is significantly better than that of existing methods when determining constrained path optimization. Moreover, field experimental results demonstrate the successful generation of fast and safe trajectories for real-time autonomous driving.

1 January 2015 in Procedia Engineering
Verified
Journal Article
Open access
Matti LinjamaMatti Linjama
Matti Linjama
MPMiika Paloniitty
Miika Paloniitty
LTLassi Tiainen
Lassi Tiainen
KHKalevi Huhtala
Kalevi Huhtala
Digital hydraulic valve system is a new kind of hydraulic control valve assembly, which has potential to save energy and improve performance of valve controlled actuators. The typical digital valve system has 4-6 parallel connected on/off valves per control edge totaling 16-24 valves in the four-way valve configuration. The flow capacities of the parallel connected valves are set according to the powers of two such that it is possible to achieve 2N different flow rates with N valves. An alternative approach is to use equally sized valves, which means that N parallel connected valves give only N+1 different flow rates. This approach has several benefits, but the number of valves becomes very large. This paper shows that it is possible to implement valve assembly having 128 miniaturized valves, such that it can be installed instead of traditional CETOP3 servo or proportional valve. Careful electromagnetic optimization and mechatronic design are used together with novel manufacturing methods and new type of power electronics. The prototype is build and experimentally studied and results show that the performance of this kind of digital valve system is superior to traditional four-way control valves in terms of response time and fault tolerance

23 February 2016 in Journal of Engineering for Gas Turbines and Power
Verified
Journal Article
Closed access
PDF file
TKTomi R. Krogerus
Tomi R. Krogerus
MHMika P. Hyvönen
Mika P. Hyvönen
KHKalevi J. Huhtala
Kalevi J. Huhtala
Diesel engines are widely used due to their high reliability, high thermal efficiency, fuel availability, and low consumption. They are used to generate power, e.g., in passenger cars, ships, power plants, marine offshore platforms, and mining and construction machines. The engine is at heart of these applications, so keeping it in good working condition is vital. Recent technical and computational advances and environmental legislation have stimulated the development of more efficient and robust techniques for the diagnostics of diesel engines. The emphasis is on the diagnostics of faults under development and the causes of engine failure or reduced efficiency. Diesel engine fuel injection plays an important role in the development of the combustion in the engine cylinder. Arguably, the most influential component of the diesel engine is the fuel injection equipment; even minor faults can cause a major loss of efficiency of the combustion and an increase in engine emissions and noise. With increased sophistication (e.g., higher injection pressures) being required to meet continuously improving noise, exhaust smoke, and gaseous emission regulations, fuel injection equipment is becoming even more susceptible to failure. The injection systems have been shown to be the largest contributing factor in diesel engine failures. Extracting the health information of components in the fuel injection system is a very demanding task. Besides the very time-consuming nature of experimental investigations, direct measurements are also limited to selected observation points. Diesel engine faults normally do not occur in a short timeframe. The modeling of typical engine faults, particularly combustion related faults, in a controlled manner is thus vital for the development of diesel engine diagnostics and fault detection. Simulation models based on physical grounds can enlarge the number of studied variables and also obtain a better understanding of localized phenomena that affect the overall behavior of the system. This paper presents a survey of the analysis, modeling, and diagnostics of diesel fuel injection systems. Typical diesel fuel injection systems and their common faults are presented. The most relevant state of the art research articles on analysis and modeling of fluid injection systems as well as diagnostics techniques and measured signals describing the behavior of the system are reviewed and the results and findings are discussed. The increasing demand and effect of legislation related to diagnostics, especially on-board diagnostics (OBD), are discussed with reference to the future progress of this field.

25 July 2019 in Applied Sciences
Verified
Journal Article
Open access
PDF file
KMKrishna Mohan Mishra
Krishna Mohan Mishra
KHKalevi Huhtala
Kalevi Huhtala
In this paper, we propose a new algorithm for data extraction from time-series data, and furthermore automatic calculation of highly informative deep features to be used in fault detection. In data extraction, elevator start and stop events are extracted from sensor data including both acceleration and magnetic signals. In addition, a generic deep autoencoder model is also developed for automated feature extraction from the extracted profiles. After this, extracted deep features are classified with random forest algorithm for fault detection. Sensor data are labelled as healthy and faulty based on the maintenance actions recorded. The remaining healthy data are used for validation of the model to prove its efficacy in terms of avoiding false positives. We have achieved above 90% accuracy in fault detection along with avoiding false positives based on new extracted deep features, which outperforms results using existing features. Existing features are also classified with random forest to compare results. Our developed algorithm provides better results due to the new deep features extracted from the dataset when compared to existing features. This research will help various predictive maintenance systems to detect false alarms, which will in turn reduce unnecessary visits of service technicians to installation sites.