A Study on Noise Reduction of Gear Pumps of Wheel Loaders Based on the ICA Model

Panling Huang, Liang Xu, Chuan Luo, Jianchuan Zhang, Chi Feng, Qi Zhang, Jun ZhouInternational Journal of Environmental Research and Public Health 2019 · DOI

TL;DR

This paper tackles the significant noise pollution from gear pumps in wheel loaders, a common issue in construction machinery. By developing a specialized test rig and employing Independent Component Analysis (ICA) to isolate gear pump noise, they identified key noise sources and implemented a novel strategy of increasing gear teeth number and modifying tooth profiles, leading to measurable noise reductions.

What Problem Does This Paper Address?

Construction machinery, particularly wheel loaders, are significant contributors to noise pollution. Gear pumps, essential for their hydraulic systems, are identified as a major source of this noise. This noise not only impacts environmental sustainability but also poses health risks to operators and can affect the market competitiveness of these machines. Existing noise reduction strategies often struggle to isolate the specific noise contributions of individual components like gear pumps, especially when other noisy elements like engines are present. The paper aims to address this by providing a method to accurately measure and reduce gear pump noise in a complex machinery environment.

Methods

The researchers employed a multi-pronged approach:

  • Noise Test Rig Development: A dedicated test rig was constructed to isolate the gear pump and minimize interference from other noise sources (like the engine). This rig included a motor, the gear pump, acoustic sensors, and a data acquisition system.
  • Independent Component Analysis (ICA): To separate the gear pump's noise signal from the mixed signals captured by the sensors (which included motor and accessory noise), the authors adapted ICA for the frequency domain. This involved using Fast Fourier Transform (FFT) to convert time-domain signals to frequency-domain, applying ICA for separation, and then Inverse Fast Fourier Transform (IFFT) to return the signals to the time domain. This frequency-domain ICA approach was chosen to handle the complexities of signal transmission, including time delays and path effects.
  • Theoretical and Simulation Analysis: The paper delved into the theoretical underpinnings of gear pump noise generation, focusing on flow pulsation and gear meshing impact. They analyzed how factors like tooth number and tooth profile influence these phenomena. Finite Element Analysis (FEA) was used to simulate meshing forces and stresses.
  • Noise Reduction Strategy: Based on their analysis, two primary noise reduction strategies were proposed and implemented:
  • Increasing Teeth Number: The number of teeth on the gear pump was increased from 9 to 12.
  • Gear Profile Modification: A "drum shape" modification was applied to the gear teeth profile to improve machining accuracy and reduce meshing impact.
  • Verification: The effectiveness of the optimized gear pump was verified through noise tests on the rig and subsequently on a wheel loader.

Datasets

Not explicitly discussed in the paper. The study relies on experimental measurements collected from a custom-built test rig and a wheel loader.

Key Insights and Ideas

  • ICA for Noise Source Separation is Powerful: The application of frequency-domain ICA proved crucial in isolating the gear pump's noise from a complex acoustic environment. This technique is a significant advancement for diagnosing noise in machinery.
  • Flow Pulsation and Meshing Impact are Key Noise Drivers: The paper reinforces that flow pulsation and gear impact are the primary culprits behind gear pump noise, rather than just the pump's inherent structure.
  • Teeth Number Matters for Flow Pulsation: Increasing the number of teeth on the gear pump directly reduces flow pulsation, a key factor in noise generation.
  • Profile Modification Enhances Meshing Stability: Modifying the gear tooth profile, specifically using a drum shape, effectively reduces meshing forces and impacts, leading to quieter operation.
  • Holistic Approach is Necessary: Effective noise reduction requires understanding the noise generation mechanism, accurately measuring the specific component's noise, and then applying targeted design modifications.

Key Results

  • Noise Reduction Achieved: The optimized gear pump demonstrated a reduction in:
  • Mixed noise value: 4.7 dB(A)
    • Gear pump noise: 1.7 dB(A)
    • Driver's position noise: 2.2 dB(A)
    • External radiation noise: 1 dB(A)
  • Flow Pulsation Reduction: Increasing teeth from 9 to 12 reduced flow pulsation by 5.7%.
  • Meshing Force Reduction: Simulation showed a reduction in maximum meshing force by approximately 30% with the 12-tooth gear pump.
  • Compliance with Standards: The driver's position noise of the optimized pump (79.3 dB(A)) fell below the CE certification value (<80 dB(A)).

Limitations and Future Work

  • External Radiation Noise: While significant improvements were made, the external radiation noise of the optimized pump (109.3 dB(A)) was still slightly above the CE certification value (<107 dB(A)). The authors suggest redesigning the engine's muffler structure as a next step to address this.
  • Engine Noise: The paper focuses solely on gear pump noise. The authors mention that engine noise reduction will be addressed in a future stage, indicating that a comprehensive solution for wheel loader noise would require tackling multiple sources.
  • Generalizability: The noise reduction theory developed for gear pumps is proposed to be extendable to other construction machinery fields, suggesting future work in applying these principles more broadly.

Summary updated

Abstract

In order to reduce the noise level of wheel loaders caused by gear pumps and realize environmentally sustainable development, the noise generation mechanism of a gear pump was studied, and the influence of flow pulsation and gear impact on noise was analyzed. In order to reduce the interference of other noise sources on the noise level of the gear pump, a noise test rig was established. The mixed noise signals obtained from the rig test were separated using the ICA model. The ICA model includes the following algorithms: The fast Fourier transform (FFT), independent component analysis (ICA) and inverse fast Fourier transform (IFFT). Some theories about the influence of the teeth number and teeth profile on noise were analyzed by theory and simulation. A noise reduction strategy was proposed by increasing the teeth numbers and modifying the teeth profile of the gear pump. The tests results showed that the noise values of the external and the driver's ear of the wheel loader were reduced to 1 and 2.2 dB (A), respectively. This proves the effectiveness of the optimization scheme of gear pump noise reduction.
BibTeX
@article{huang2019study,
  author = {Panling Huang and Liang Xu and Chuan Luo and Jianchuan Zhang and Chi Feng and Qi Zhang and Jun Zhou},
  title = {{A Study on Noise Reduction of Gear Pumps of Wheel Loaders Based on the ICA Model}},
  journal = {International Journal of Environmental Research and Public Health},
  year = {2019},
  doi = {10.3390/ijerph16060999},
}
Figure