KR20240081917A - Diagnosis device for rotating machinery using displacement signal analysis - Google Patents

Abstract

By receiving acceleration from a rotating machine and performing two integrations through a signal processing process that effectively removes harmonic noise and low-frequency drift accompanying the use of an acceleration sensor, the vibration speed and rotation trajectory of the rotating body are provided, and the rotation trajectory is provided in the frequency domain. This is about a technology that provides the operator with a rotation trajectory in the time domain reconstructed at the rotation frequency and its high frequency, allowing the operator to intuitively understand the operating state.

Description

Diagnosis device for rotating machinery using displacement signal analysis}

The present invention relates to a device for diagnosing abnormal conditions of a rotating machine including a rotating shaft using displacement signal analysis.

Rotation trajectories are widely used to diagnose abnormal conditions in rotating machines. Rotating machines used in industry are usually supported by bearings and are connected to other machines through shaft couplings to provide rotation or rotational power. Malfunctions in rotating machinery can be repaired at low cost if diagnosed early. Malfunction of rotating machinery can result in a sudden increase in vibration or shock due to stoppage, and can cause damage to surrounding devices, so the condition and lifespan of rotating machinery equipment needs to be monitored using an appropriate diagnostic device.

In the case of severe abnormalities in rotating machinery equipment, which is widely used in industrial settings, the abnormal condition can be determined based on the level of noise generated by vibration using not only general sensors but also inexpensive acoustic sensors. However, in order to detect minute vibration changes in the early stage of failure or gradual changes in state before failure, diagnostic techniques with excessive operating costs are required, resulting in a high cost threshold. In contrast, sensor systems that are easy to install and low cost have limitations in that they do not provide sufficient precision and reliability required to monitor abnormal conditions. The present invention develops a technology that is relatively easy to install, obtains highly reliable results, and lowers the cost threshold by monitoring vibration with an acceleration sensor.

The rotation trajectory uses a method of obtaining the vibration displacement of the rotating body by integrating the acceleration twice. Typically, an acceleration sensor contains a piezoelectric element inside and converts the force applied by the internal reference mass into an electrical signal to output the amount of vibration of the sensor mounting area. Acceleration sensors are convenient because they do not need to be installed directly facing a vibrating rotating body. On the other hand, noise and drift are problems. Since the system where rotating machinery is installed is an environment where various vibrations exist, various background noises are introduced in addition to the vibration components related to abnormal conditions. For example, noise generated by electrical and mechanical factors in the acceleration sensor itself. Additionally, when an abnormal condition occurs, additional heat is usually generated in the system. Heat flowing into the acceleration sensor causes signal drift along with noise. In order to easily diagnose abnormalities in rotating machinery equipment, error factors caused by such noise and drift must be easily excluded. The information required for diagnosing the condition of rotating machinery equipment is mainly included in the frequency signal components linked to the rotation speed of the rotating body. It is possible to exclude signals that change static components at very low frequencies or signals with frequencies several times higher than the rotation frequency, and limit the information needed for diagnosing abnormal conditions by focusing on components linked to rotation speed. By exchanging this as a signal converted to the frequency domain, for example, only a small amount of data can be transmitted to a mobile terminal, while providing a rotation trajectory containing sufficient state information so that an ordinary worker can intuitively understand it.

By receiving acceleration input from rotating machinery and performing two integrations through a signal processing process that effectively removes the high-frequency noise and low-frequency drift that accompany the use of acceleration sensors, the vibration speed and rotation trajectory of the rotating body are provided, and the rotation trajectory is reconstructed. By providing the operator with a rotation trajectory in the time domain that is made in the frequency domain by hardwired, FFT, and iFFT and reconstructed with the rotation frequency and its harmonics, it has the effect of providing the operator with an intuitive understanding of the operating state.

A rotating axis generates vibrations in all six degrees of freedom directions, and the eccentric mass from the rotation center axis cannot be completely removed. For this reason, the rotation trajectory component due to vibration synchronized to the rotation speed is the most dominant in the entire rotation trajectory. In most cases, the rotation axis is connected to additional equipment, and various mechanical elements such as couplings, gears, and transmissions are directly or indirectly connected to the rotation axis, so it is possible to monitor various operating conditions by analyzing the rotation trajectory. When an abnormal condition occurs in the rotation system, additional vibration is generated, and the abnormal condition is monitored by increasing the vibration component at an integer multiple of the rotation frequency. The rotating machinery equipment diagnosis device receives vibration signals from the rotating shaft of the rotating machinery equipment from a pair of acceleration sensors mounted on the rotating body housing, performs signal processing and abnormality diagnosis tasks, and provides diagnostic information, including the rotation trajectory, to the user via wired or wireless means. send. The rotating machinery equipment diagnosis device includes multiple ADCs at the input terminal, and the main processor for signal processing includes FFT and can perform hard-wired FFT. In order to acquire a rotation trajectory from a rotation axis, the most basic method is to install two displacement sensors arranged perpendicular to each other on a plane perpendicular to the rotating axis. However, in this case, the shape precision of the circumferential surface of the target measured by the displacement sensor must be very high, and in the case of general rotating bodies, it is rare for free space to be secured in advance for installing the displacement sensor. Although it varies depending on the structure of the rotor, installing a displacement sensor in a general rotor structure that takes into account lubrication, cooling, and prevention of foreign matter inflow not only incurs additional costs, but it is often difficult to modify the system. Therefore, when applying to more general-purpose rotating machinery, an acceleration sensor that can be easily mounted on the rotating body housing is used. When using an acceleration sensor, even when it is not in an abnormal state, vibration occurs due to rotation in most rotating bodies, and in most cases, the rotating body is not a single rigid body, but multiple elements are connected and supported by various bearings, so it generally generates an acceleration signal. The displacement component is obtained by integrating twice, which is only valid when the acceleration signal is ideal and generates a complex vibration signal. When errors in the offset component accumulate, errors accumulate in the velocity and displacement signals obtained through integration, reducing the reliability of abnormal state diagnosis. Accordingly, an acceleration sensor including a reference mass and piezoelectric elements is constructed, and the rotation trajectory is reconstructed through iFFT. From the signal information obtained from the FFT unit, the radius of the rotation trajectory, phase, and frequency components of vibration can be extracted. Additionally, the maximum eccentricity amount, maximum eccentricity phase angle, and eccentricity waveform value can be calculated from the rotation trajectory and used for diagnosing abnormal conditions.

Claims (1)

A sensor input terminal that receives acceleration signals in directions perpendicular to each other and perpendicular to the rotation axis, an A/D conversion unit that converts one rotation's worth of acceleration signals into digital signals, and a bandpass filter and bandpass that allow only the frequency band of interest to pass through the digital signal. A first DC offset removal unit that subtracts the average value from the output of the filter, an FFT unit that extracts a frequency component from the output of the second integrator, a displacement calculation unit that reconstructs the rotation trajectory from harmonic components of the rotation speed of the rotation shaft, and the rotation trajectory. A rotating machinery equipment diagnostic device including a communication unit that transmits the status information to a computer or mobile terminal by wire or wirelessly.