Neural-network based analysis and prediction of a compressor’s characteristic performance map

Yu, Youhong; Chen, Lingen; Sun, Fengrui; Chen, Wu

doi:10.1016/j.apenergy.2006.04.005

Cited by 97 publications

(44 citation statements)

References 19 publications

Supporting

Mentioning

Contrasting

Unclassified

Order By: Relevance

“…Neural network based method was employed in [71][72][73] to develop the compressor map. A.Lazzaretto and A.Toffolo use the output power to predict the compressor performance (pressure, mass flow, efficiency, etc) through constructing a neural network [72].…”

Section: Compressor Modeling 221 Compressor Map Modeling Overviewmentioning

confidence: 99%

Control of an Ultrahigh-Speed Centrifugal Compressor for the Air Management of Fuel Cell Systems

Zhao

Blunier

Gao

et al. 2014

IEEE Trans. on Ind. Applicat.

View full text Add to dashboard Cite

Fuel cell is a power supply system, which takes advantage of the electro-chemical reaction between oxygen and hydrogen to produce electricity with water as its byproduct. Due to their lower pollution and high efficiency compared with fossil fuels, fuel cell systems are under extensive development for many power applications. Compared with high temperature fuel cells (such as molten carbonate fuel cells or solid oxide fuel cells), the proton exchange membrane fuel cell (PEMFC) is more suitable for automotive applications because of its low temperature working condition, thereby a fast startup.Air compressor supplying the oxygen to the stack is an important component in the fuel cell systems. The compressor can consumes up to 20 % of the generated power in the most severe cases. The selecting of the compressor and corresponding control are directly related to the performance of the fuel cell. The size and weight of the air compressor has to be reduced to make them more feasible for automotive applications. Moreover, the control of the air compression system is also an important issue, which affects the efficiency and the safety of the fuel cell. To avoid oxygen starvation of the stack, the mass flow of the supplied air has to be controlled appropriately according to the load demand. Meanwhile, the pressure should not have large deviations or ripples which may damage the stack membrane.In this thesis, the requirements to the compressor for fuel cell applications are discussed. And several kinds of compressor are compared, including positive displacement types and dynamic superchargers. In the view of compactness, an ultra-high speed, up to 280,000 rpm, centrifugal compressor is adopted. The centrifugal compressor which is a kind of dynamics superchargers converts the kinetic energy of the air to the pressure. Compared with positive displacement compressors, it has the advantages of compactness, high efficiency and low noise, making it more suitable for automotive applications. However, the difficulty of the control as well as the surge iv constraint obstructs its industrial implementation for fuel cells.In this dissertation, The model of the air compression system is developed and the coupling between the model inputs and outputs are analyzed. The compressor map is measured and then fitted by a two hidden layer neural network model. The relative gain array (RGA) method is used for the coupling analysis. It shows that the coupling degree varies as the operation point changes because of the nonlinearity of the system. The second part of this work focuses on solving the control problems to adapt the centrifugal compressor to the fuel cell system. The objective is to control the mass flow and pressure simultaneously according to the fuel cell load demand. The problem lies in the coupling between the mass flow control and pressure control because the variations of one variable will also lead to the deviation of the other one. The interaction between the two control loops declines the performance of the fuel cell...

show abstract

Section: Compressor Modeling 221 Compressor Map Modeling Overviewmentioning

confidence: 99%

Control of an Ultrahigh-Speed Centrifugal Compressor for the Air Management of Fuel Cell Systems

Zhao

Blunier

Gao

et al. 2014

IEEE Trans. on Ind. Applicat.

View full text Add to dashboard Cite

show abstract

“…Jiang et al [5] and Yu et al [6] performed this type of research studies for centrifugal compressor and axial compressor, respectively. However, due to the complexity of the models and the amount of initial data required, in the general area of pipeline simulations the experimentally determined characteristic curves are commonly used.…”

Section: Centrifugal Compressormentioning

confidence: 99%

“…(6) are calculated from the original virial coefficients by equating Eqs. (5) and (6) and solving the original virial expansion for p…”

Section: Centrifugal Compressormentioning

confidence: 99%

Exergy-based analysis of gas transmission system with application to Yamal-Europe pipeline

2011

View full text Add to dashboard Cite

“…A comparison across different ANNs in predicting axial compressor performance map is presented by the present authors. 13,14 In a different study, Yu et al 15 applied back-propagation neural network to predict the compressor characteristic map thorough twice training.…”

Section: Introductionmentioning

confidence: 99%

Application of integrated fuzzy logic and neural networks to the performance prediction of axial compressors

Gholamrezaei

Ghorbanian

2015

Proceedings of the Institution of Mechanical Engineers, Part A:

View full text Add to dashboard Cite

An integrated fuzzy logic-neural network methodology is presented as a mean to improve the reconstruction of the performance map of axial compressors and fans. The learning capability of artificial neural network technique is integrated to the knowledge aspect of fuzzy inference system to offer enhanced prediction capabilities compared to using a single methodology independently. The proposed technique incorporates information of experimental data on surge, operating, and choke lines at any arbitrary but fixed rotational speed. A comparison of the predicted results with experimental data reveals a very good agreement. The proposed technique has the capability to model the nonlinear surge line as well as the kink in the performance map. Application of the method for compressor map generation showed that the proposed technique is robust and capable of enhancing any performance simulation tool used for the dynamic simulation and condition monitoring.

show abstract

Neural-network based analysis and prediction of a compressor’s characteristic performance map

Cited by 97 publications

References 19 publications

Control of an Ultrahigh-Speed Centrifugal Compressor for the Air Management of Fuel Cell Systems

Control of an Ultrahigh-Speed Centrifugal Compressor for the Air Management of Fuel Cell Systems

Exergy-based analysis of gas transmission system with application to Yamal-Europe pipeline

Application of integrated fuzzy logic and neural networks to the performance prediction of axial compressors

Contact Info

Product

Resources

About