Data Processing Approaches to Measure Velocity of Electromagnetic Gun on Laser Screen in Complex Environment

Huiyan Hao, Wenyu Liu, Peng Xu, Hui ZhaoSensors 2022 · DOI

TL;DR

This paper tackles the challenge of accurately measuring the velocity of projectiles fired from electromagnetic guns in noisy environments. They propose a novel data processing approach combining Ensemble Empirical Mode Decomposition (EEMD) and Correlation Algorithm (CA) to effectively filter out interference and precisely identify the projectile's passage time, leading to more reliable velocity measurements.

What Problem Does This Paper Address?

Electromagnetic guns offer high-speed projectile launching, but accurately measuring their exit velocity is crucial for performance assessment and system development. Traditional laser screen-based velocity measurement systems face significant challenges in the harsh firing environment. This includes interference from vibration, electromagnetic noise, smoke, flare, and even fragments, which can distort or completely obscure the signal from the projectile passing through the laser screen. This makes it difficult to reliably extract the precise moment of passage, a critical factor for calculating velocity. The paper aims to overcome these environmental and signal-processing hurdles to achieve accurate and consistent velocity measurements.

Methods

The core of the paper's contribution lies in its proposed data processing pipeline. It consists of two main stages:

  1. Ensemble Empirical Mode Decomposition (EEMD): This technique is used as a preprocessing step to denoise the raw signals captured by the laser screens. EEMD is an adaptive method that decomposes a signal into a set of intrinsic mode functions (IMFs) and a residual. By adding white noise and averaging multiple decompositions, EEMD effectively mitigates issues like modal aliasing and baseline drift, which are common in non-stationary signals like those from electromagnetic gun launches. The authors specifically select and reconstruct certain IMFs (IMF5-IMF7) to retain the essential signal while removing high-frequency noise and low-frequency baseline drift.
  2. Correlation Algorithm (CA): Following EEMD denoising, a correlation algorithm is employed to identify the corresponding signals from two sequential laser screens. This is crucial for distinguishing the actual projectile signal from residual interference or signals from debris. The CA calculates the cross-correlation between the signals from the two laser screens. When the correlation coefficient exceeds a threshold (e.g., 0.9), the signals are considered to belong to the same object. This step helps in accurately associating the start and stop signals of the projectile.
  3. Feature Point Extraction: After signal identification, the paper uses the maximum slope point of the processed signal as the feature point. This point is identified as the moment the armature's trailing edge passes the laser screen. This precise timing is then used to calculate the time interval between the two laser screens.
  4. Velocity Calculation: Finally, the velocity is calculated using the standard formula: $v = s/t$, where $s$ is the known distance between the two laser screens and $t$ is the time interval determined by the maximum slope points.

Datasets

The paper utilizes experimental data collected from an electromagnetic gun firing system. They describe conducting six experiments where an armature passes through four laser screens. The data consists of four-channel signals captured by the laser screens for each firing event. No specific public datasets are mentioned; the data appears to be generated specifically for this study.

Key Insights and Ideas

  • EEMD's Efficacy in Harsh Environments: The paper demonstrates that EEMD is a powerful tool for cleaning up signals in extremely noisy environments, such as those encountered during electromagnetic gun launches. It effectively separates the desired signal from various interferences like vibration and baseline drift.
  • Correlation for Robust Identification: The use of a correlation algorithm is critical for distinguishing the actual projectile signal from spurious signals caused by debris or other transient events. This ensures that the timing is based on the intended target.
  • Maximum Slope as a Precise Feature: Identifying the maximum slope point as the passage time offers a robust and repeatable method for determining the precise moment the armature crosses the laser screen, even with complex signal shapes.
  • Combined Approach for Superior Accuracy: The synergy between EEMD for denoising and CA for identification leads to a significant improvement in the accuracy and reliability of velocity measurements compared to traditional methods in challenging conditions.

Key Results

The paper presents experimental results from six firings. Table 1 shows the velocities obtained for six individual armature launches using the proposed method, with values ranging from 682 m/s to 991 m/s.

More importantly, the reliability of the method is verified using a system with four laser screens (Q1, Q2, T1, T2) and two sets of measurements ($s_1$ and $s_2$). Table 2 summarizes the calculated velocities ($V_{1k}$ and $V_{2k}$), average velocities ($U_k$), and uncertainties ($\sigma$) for six signals. The relative uncertainties ($\sigma_0$) are consistently below 0.1%, indicating high consistency and credibility of the measurements. The authors conclude that the consistency of the velocity trend measured by the two test systems is very good, implying the algorithm's reliability.

Limitations and Future Work

  • Limitations:
    • The paper acknowledges that while EEMD reduces high-frequency interference, some interference from fragments or particles may still persist and affect signal identification.
  • The effectiveness of the correlation algorithm relies on a sufficiently strong correlation between the signals from the two laser screens. If the signals are too distorted, identification might still be challenging.
  • The paper mentions that the threshold setting for signal amplitude in the correlation algorithm is related to armature section size and other parameters, implying a need for careful calibration.
  • Future Work:
    • The authors suggest that the proposed EEMD and CA approach is self-adaptive and suitable for various object shapes passing through two screens, offering a high-precision method for effective armature information extraction, especially when ideal waveforms are not detected due to system disturbances. This implies potential for further refinement and application to more complex scenarios.

Summary updated

Abstract

The exit velocity of the armature is an important indicator in measuring the launching performance of the electromagnetic gun. The non-contact photoelectric detection technology with the use of a laser screen was applied to the measurement of the armature velocity of the electromagnetic gun. By means of taking the signals that pass through the laser screen obtained by the velocity measurement system as the research object, we solved problems such as the harsh test environment of the launch armature velocity of the electromagnetic gun, the interferences on the armature signal passing through the laser screen unavoidably caused by various factors such as vibration, electromagnetic interference, shock wave, flare, smoke and fragments, and even the non-recognition of the signal passing through the laser screen in severe cases. A data-processing algorithm that combines the Ensemble Empirical Mode Decomposition (EEMD) with Correlation Algorithm (CA) was proposed, with the aim of processing the signals passing through the laser screen, while using the maximum slope point as the time passing through the laser screen so as to calculate the velocity of the armature passing the laser screen. This method can effectively reduce the influence of interference on the test results, and the test results from two sets of velocity measuring systems show that the velocity obtained by the proposed approach is highly consistent.
BibTeX
@inproceedings{hao2022data,
  author = {Huiyan Hao and Wenyu Liu and Peng Xu and Hui Zhao},
  title = {{Data Processing Approaches to Measure Velocity of Electromagnetic Gun on Laser Screen in Complex Environment}},
  booktitle = {Sensors},
  year = {2022},
  doi = {10.3390/s22176573},
}
Figure