Improving accuracy of cavitation severity recognition in axial piston pumps by denoising time-frequency images

Author(s):  
Qun Chao ◽  
Xiaoliang Wei ◽  
Junbo Lei ◽  
Jianfeng Tao ◽  
Cheng-Liang Liu

Abstract Vibration signal is a good indicator of cavitation in axial piston pumps. Some vibration-based machine learning methods have been developed for recognizing the pump cavitation. However, their fault diagnostic performance is often unsatisfactory in industrial applications due to the sensitivity of the vibration signal to noise. In this paper, we presented an intelligent method to recognize the cavitation severity of an axial piston pump under noisy environment. First, we adopted short-time Fourier transformation to convert the raw vibration data into spectrograms that acted as input images of a modified LeNet-5 convolutional neural network (CNN). Second, we proposed a denoising method for the converted spectrograms based on frequency spectrum characteristics. Finally, we verified the proposed method on the dataset from a test rig of high-speed axial piston pump. The experimental results indicate that the denoising method significantly improves the diagnostic performance of the CNN model under noisy environment. For example, the accuracy rate of the cavitation recognition increases from 0.52 to 0.92 at SNR of 4 dB by the denoising method.

Micromachines ◽  
2021 ◽  
Vol 12 (4) ◽  
pp. 398
Author(s):  
Ying Li ◽  
Xing Chen ◽  
Hao Luo ◽  
Jin Zhang

The manufacturing development of axial piston pumps usually takes the trend of high speed and miniaturization, and increases power density. Axial piston pumps are usually characterized as high speed to improve the power density; thus, high-speed churning losses caused by the internal rotating components stirring the oil can increase significantly. In order to improve the efficiency, more attention should be given to the churning losses in axial piston pumps, especially in high-speed conditions. Using the method of least-squares curve fitting, this paper establishes a series of formulas based on the churning losses test rig over a wide range of speeds, which enable accurate predictions of churning losses on the cylinder block and pistons. The reduction coefficient of flow resistance of multi-pistons as calculated. The new churning losses formula devoted to the cylinder block and rotating pistons was validated by comparison with experimental evidence in different geometries of axial piston pumps. According to the prediction model of churning losses, some valuable guidance methods are proposed to reduce the energy losses of the axial piston pump, which are the theoretical support for the miniaturization of axial piston pump manufacturing.


1994 ◽  
Vol 116 (2) ◽  
pp. 281-287 ◽  
Author(s):  
R. M. Harris ◽  
K. A. Edge ◽  
D. G. Tilley

The suction dynamics of axial piston pumps become more critical if the pump is to be used at high speeds. In order to prevent air-release and cavitation from occurring within the pump it is necessary to pressurise the pump inlet. As the speed of a pump is increased, higher boost pressures are required, due to the extra losses incurred through the suction line and portplate at the higher flowrates. However, the lack of data regarding axial piston pump behavior at high speeds creates problems for the system designer in selecting suitable boost conditions and for the pump designer in selecting the portplate configuration that is required to reduce fluid-borne-noise levels. This paper discusses the suction performance of piston pumps, and presents experimental and simulation results exploring the behavior of a high-speed axial-piston pump. Different air-release and cavitation models that are suitable for simulation studies are investigated.


Author(s):  
Shu Wang

The timing definition of valve plates is one of the most complex topics in the piston pump designs because it affects many pump characteristics (such as efficiency, swashplate stroking, stabilities, noise, etc.). In the study, the pressure carryover is introduced and defined as the average angular positions to locate piston pressure transitions from the top dead center (TDC) or bottom dead center (BDC) in the piston pump. Pressure carryover presents the overall outcome of the pressure transitions within piston bores. The new pressure carryover definition is derived by the timing angles and other geometrics of valve plates that is an approximation of the practical pressure transitions. The pressure carryover also determines the containment forces and moments on the swashplate produced by the pumping pistons. The relationship between the pressure carryover angle and the containment moment has been developed and analyzed in the study. The amplitudes and frequencies of the forces and moments can be changed by varying the pressure carryover angle that produce different tonalities and control efforts for the swashplate type axial-piston pumps. Therefore, the pressure carryover is the most important and straightforward connection between pump dynamics and valve plate designs. In order to optimize the pump performance, the piston pressure carryover might be investigated thoroughly for the pump and its controller designs.


Sensors ◽  
2020 ◽  
Vol 20 (22) ◽  
pp. 6576
Author(s):  
Shengnan Tang ◽  
Shouqi Yuan ◽  
Yong Zhu ◽  
Guangpeng Li

A hydraulic axial piston pump is the essential component of a hydraulic transmission system and plays a key role in modern industry. Considering varying working conditions and the implicity of frequent faults, it is difficult to accurately monitor the machinery faults in the actual operating process by using current fault diagnosis methods. Hence, it is urgent and significant to investigate effective and precise fault diagnosis approaches for pumps. Owing to the advantages of intelligent fault diagnosis methods in big data processing, methods based on deep learning have accomplished admirable performance for fault diagnosis of rotating machinery. The prevailing convolutional neural network (CNN) displays desirable automatic learning ability. Therefore, an integrated intelligent fault diagnosis method is proposed based on CNN and continuous wavelet transform (CWT), combining the feature extraction and classification. Firstly, CWT is used to convert the raw vibration signals into time-frequency representations and achieve the extraction of image features. Secondly, a new framework of deep CNN is established via designing the convolutional layers and sub-sampling layers. The learning process and results are visualized by t-distributed stochastic neighbor embedding (t-SNE). The results of the experiment present a higher classification accuracy compared with other models. It is demonstrated that the proposed approach is effective and stable for fault diagnosis of a hydraulic axial piston pump.


2012 ◽  
Vol 6 (4) ◽  
pp. 410-417 ◽  
Author(s):  
Tetsuhiro Tsukiji ◽  
◽  
Eishin Noguchi ◽  
Futoshi Yoshida ◽  

In the present study, we succeed in observing cavitating flow near a notch (V-shaped groove) in a valve plate in an axial piston pump, and we improve an oil hydraulic ball valve, using the visualization technique. Our model of the axial piston pump, is designed to allow the jet flow near the notch and the cavitation cloud to be observed clearly from two directions using a high-speed video camera. Computational Fluid Dynamics (CFD) is employed to estimate the occurrence and the region of the cavitation cloud. It is found that our CFD method is very useful for estimating the region of the cavitation cloud. It is further found that adding notches serves to greatly reduce the cavitation region. Using a commercially available digital video camera, a high-speed video camera, and X-rays source, we also succeed in improving an oil hydraulic ball valve by preventing vibration, cavitation, and the noise.


Author(s):  
N.A. Belov ◽  
O.F. Nikitin

The article considers the flow of the working fluid in the discharge channel of the axial piston pump with end distribution. Geometric region shapes of the channels, currently used in axial piston pumps, negatively affecting the dynamic parameters of the flow flowing through it, are determined by numerical simulation. The configuration of the channel cavity allowing a more uniform distribution of dynamic parameters over the volume of the fluid flow is proposed. The optimal ratio between the reference dimensions adopted for constructing a three-dimensional model of the channel was determined based on the study of the dependence of the power factor value, the amount of movement in the output section vs the shape of the channel. Energy loss due to flowing the working fluid through the channel is reduced. The resulting force effect on the discharge pipe and other elements connected to the pump is reduced and the vibroacoustic characteristics of the pump unit are improved.


2018 ◽  
Vol 70 (6) ◽  
pp. 1137-1144
Author(s):  
Zhanling Ji

Purpose High pressure and high speed of the axial piston pump can improve its power density, but they also deteriorate the thermal-fluid-structure coupling effect of the friction pairs. This paper aims to reveal the coupling mechanism of the pump, for example, valve plate pair, by carrying out research on multi-physics field coupling. Design/methodology/approach Considering the influences of temperature on material properties and thermal fluid on structure, the thermal-fluid elastic mechanics model is established. A complete set of fast and effective thermal-fluid-structure coupling method is presented, by which the numerical analysis is conducted for the valve plate pair. Findings According to calculations, it is revealed that the temperature and pressure evolution laws of oil film with time, the pressure distribution law of the fluid, stress and displacement distribution laws of the solid in the valve plate pair. In addition, the forming history of the wedge-shaped oil film and mating clearance change law with rotational speed and outlet pressure in the valve plate pair are presented. Originality/value For an axial piston pump operating under high speed, high pressure and wide temperature range, the multi-physics field coupling analysis is an indispensable means and method. This paper provides theoretical evidence for the development of the pump and lays a solid foundation for the research of the same kind of problem.


2006 ◽  
Vol 129 (4) ◽  
pp. 456-468 ◽  
Author(s):  
Alessandro Roccatello ◽  
Salvatore Mancò ◽  
Nicola Nervegna

Analysis of a variable displacement axial piston pump, as in other complex fluid power and mechanical systems, requires appropriate insight into three multidisciplinary domains, i.e., hydraulics, mechanics and tribology. In recent years, at FPRL, modelling of axial piston pumps has evolved in AMESim (one-dimensional code) where a three-dimensional mechanical approach has required generation of proprietary libraries leading to the evaluation of internal forces/reactions in all pump subsystems. Tribologic aspects in axial piston pumps modelling are also being investigated but AMESim, in this respect, does not appear as the appropriate computational environment. Consequently, a new approach has been initiated grounded on MSC.ADAMS. In this perspective, the paper details how the model has been developed through proprietary macros that automatically originate all pump subsystems parametrically and further apply required constraints and forces (springs, contacts and pressure forces). The ADAMS environment has also been selected due to co-simulation capabilities with AMESim. Accordingly, the paper elucidates how the entire modelling has been construed where hydraulics is managed in AMESim while ADAMS takes care of mechanics. A comparison between simulated and experimental steady-state characteristics of the axial pump is also presented. As such this paper indicates an innovative methodology for the analysis of complex fluid power systems in the hope that, eventually, tribology will also fit into the scene.


Sign in / Sign up

Export Citation Format

Share Document