Line spectra chaotification of the nonlinear vibration isolation system on the flexible foundation based on the open-plus-nonlinear-closed-loop method

Taking single degree of freedom vibration isolation system under simple harmonic excitation as an example, and considering the energy, the vibration isolation performance in different conditions was studied theoretically and numerically. The results shows that when the simple harmonic excitation import energy is definite, the vibration isolation performance at the primary harmonic frequency of the nonlinear vibration isolation system is better than that of the linear system, and the vibration isolation performance of the nonlinear vibration isolation system in chaotic vibration state is much better than that in non-chaotic vibration state. For the same isolated object, if can let the vibration isolation system vibrate chaotically, the system will possess the best isolation performance at the primary frequency.

Bearing is a key component of rotating machinery, thus the accurate and efficient fault diagnosis of bearing is critical to the safety and reliability of rotating machinery systems. In recent years, the intelligent fault diagnosis technology of bearing represented by deep learning has made rapid development. However, the stability and generalization performance of the fault diagnosis model of bearing under complex environmental noise and variable working conditions is not good. Fortunately, self-supervised learning represented by contrastive learning can provide a good solution. The current research on contrastive learning mainly focuses on self-supervised contrastive learning, but the lack of supervised learning of fault label information leads to the reduction of diagnostic accuracy and robustness under variable working conditions. To solve this problem, this paper proposes a supervised contrastive learning combined with multi-scale attention mechanism for fault diagnosis of bearing under variable operating conditions. Firstly, the preprocessing of bearing time-domain vibration data is completed, including data set division, data augmentation, and construction of “positive” or “negative” pairs for comparative learning. Then, a supervised learning network integrating multi-scale attention mechanism is constructed to complete bearing fault feature extraction. Finally, the model transfer fault diagnosis is completed under complex noise and variable working conditions. The performance of the proposed method is evaluated through a large number of noise and variable-condition experiments. The results show that the proposed method has high fault classification accuracy and robustness under the influence of complex noise and variable working conditions.

It is considerable to solve practical fault diagnosis task of gearbox under variable working conditions by introducing sufficient auxiliary data. For this purpose, a new approach called improved deep transfer auto-encoder is proposed for intelligent diagnosis of gearbox faults under variable working conditions with small training samples. First, multi-wavelet is employed as activation function for effectively learning useful features hidden in the non-stationary vibration data. Second, correntropy is used to modify the cost function to enhance the reconstruction quality. Third, pre-train an improved deep auto-encoder using sufficient auxiliary data in the source domain, and transfer its parameters to the target model. Finally, the improved deep transfer should be fine-tuned by small training samples in the target domain to adapt to the characteristics of the rest testing data. The proposed approach is used to analyze two sets of experimental vibration data collected from gearbox under variable working conditions. The results show that the proposed approach can accurately diagnose different faults of gearbox even the working conditions have significant changes, which is superior to the existing methods.

With the development of deep learning methods, the data-driven fault diagnosis methods have attracted a great deal of interest. However, as for the data-driven fault diagnosis methods, technology has to overcome various difficulties in the practical industrial scenarios, such as variable working conditions, insufficient effective samples, and environmental noise interference. Combining with the time–frequency analysis of vibration signals, a domain adaptation fault diagnosis model based on ResNet and Transformer (DAFDMRT) is proposed in this work, aiming to solve the problems encountered by current rotating machinery fault diagnosis methods in the field of application. Firstly, the vibration signal is processed by wavelet packet transform and the time–frequency information grayscale maps is constructed. Next, a deep fusion feature extraction network combining ResNet and Transformer encoder, is designed for the extraction and fusion of the local and global features of multi-scale time–frequency information. Finally, the multi-kernel maximum mean discrepancy is applied to measure and minimize the distribution difference between the deep features of source and target domain, thereby improving the diagnostic performance of the diagnosis model in variable working conditions. In this work, comparative experiments are conducted as for bearing and gearbox datasets under variable working conditions. The results indicate that DAFDMRT can show excellent performances in terms of fault diagnosis and generalization ability.

A robust construction of normalized CNN for online intelligent condition monitoring of rolling bearings considering variable working conditions and sources

A deep learning approach to estimate the state of health of lithium-ion batteries under varied and incomplete working conditions

Many cities around the world are developing over-track buildings above metro depots to achieve efficient and economical land use. However, the vibrations generated by frequent train operations have a significant impact on the over-track buildings. Therefore, the analysis and control of vibrations at metro depots are of great importance. This paper focuses on the train-induced vibration propagation law and the application of vibration isolation trench measures of the over-track building in the metro depot. To this end, a typical metro depot is taken as the research object. The train-track model, used for simulating wheel-rail force, and the track-soil-building model, used for predicting structural response, are established, respectively. Then, the vibration response of the over-track building of the metro depot is explored, and the effects of vibration isolation measures of the open trench and infilled trench in the metro depot are studied. The results show that the train-induced vibration excitation of the metro is mainly concentrated in the range of 1 to 80 Hz, and the predominant frequency range of the floor vibration is 25 to 50 Hz. The vibration response of the floor is mainly affected by the stiffness. The larger the floor area, the smaller the vertical natural frequency, and the wider the range between the train vibration excitation areas, the more prone to resonance. In addition, the vibration isolation effect of the open trench is better than that of the infilled trench. The primary factor affecting the vibration isolation performance of open trenches is their depth; the influence of trench position and width on the vibration isolation performance is weaker compared to the depth. In the predominant frequency range of floor vibration, the overall vibration isolation effect of the flexible infilled trenches is better than that of the rigid infilled trenches. The main factor affecting the vibration isolation effect of the infilled trenches is the impedance ratio of the material. Among the six kinds of filling materials selected in this paper, the barrier effect of gravel is the worst, and the barrier effect of foam is the best. Using the measure of a foam infilled trench, Z-vibration levels of floors can be reduced by 8.6–13.9 dB.

Open-circuit fault under different load and variable working conditions is the most severe issue that affects the robustness of fault diagnosis algorithms. Considering this, this research work proposes a high-frequency fault diagnosis approach that is capable to detect and isolate a single open-switch fault in a single-phase five-level Packed U-Cell (PUC5) inverter, not only under stable conditions but also under different load and variable working conditions. Firstly, a high frequency model of the PUC5 inverter is proposed. Then, the conducted emissions, measured across the electromagnetic interferences filter resistor, are employed as a fault signature. Finally, simulation results are considered to verify the effectiveness of the proposed fault diagnosis method.

Due to the coupling and diversity of compound fault features, it is difficult to decoupling diagnose the compound fault under small datasets and variable working conditions. To enhance the compound fault diagnosis performance, this work proposed a compound fault transfer diagnosis method that uses the multi-label transformer-based transfer learning network. Initially, the 1-D signal is converted into 2-D image using CWT as the input of the constructed deep learning network; Subsequently, a CNN-Transformer network structure is constructed, which simultaneously utilizes convolutional operation and self-attention mechanism to comprehensively capture the multi-scale fault features. In addition, multi-label classification is adopted to identify specific single fault types in the compound fault and output corresponding fault labels. Finally, a transfer learning strategy with fine-tuning mechanism is creatively integrated to achieve the compound fault transfer diagnosis, which can learn fault knowledge from single fault samples in source domain for compound fault diagnosis in required target tasks. The experimental datasets of planetary gearbox and harmonic reducer were collected in our own laboratory to validate the diagnostic performance. The results show that our proposed method achieved 98% average decoupling identification accuracy in insufficient compound fault samples on both datasets, and the accuracy fluctuation was less than 1.5% in the six transfer tasks. It is verified that the proposed method achieved excellent decoupling identification accuracy and generalization ability in small datasets and variable working conditions, which has significant advantage over the traditional local convolution-based multi-label classification methods and domain adaptation -based transfer learning methods.

The manuscript concerns the power flow characterization in a two‐stage nonlinear vibration isolator comprising three springs, which are configured so that each stage of the system has a high‐static‐low‐dynamic stiffness. To demonstrate the distinction of evaluation for vibration isolation using power flow, force transmissibility is used for comparison. The dynamic behavior of the isolator subject to harmonic excitation, however, is of interest here. The harmonic balance method (HBM) could be used to analyze the frequency response curve (FRC) of the strong nonlinear vibration system. A suggested stability analysis to distinguish the stable and the unstable HBM solutions is described. Increasing both upper and lower nonlinear stiffness could bend the first resonant peak to the left. The isolation range in the power and the force transmissibility plot could be extended to the lower frequencies when the nonlinear stiffness is increased, but the rate of roll‐off for the power transmissibility is twice the rate for the force transmissibility at each horizontal stiffness setting. An explanation for this phenomenon is given in the paper.

Buildings are often constructed on a continuous layer of vibration isolation to reduce vibrations and structure-borne noise. There are many vibration isolation products (rubber mats, textiles, polymer foams), but none fulfils all requirements in terms of load-bearing capacity, performance, and cost. In previous work, we have shown that silica aerogel particle beds display a low resonance frequency and correspondingly high vibration isolation performance. Loose silica aerogel granules are not a practical vibration isolation solution. In order to guide the future development of dedicated vibration isolation products based on aerogels, we determine the vibro-acoustic properties of a variety of existing commercial and R&D stage mesoporous composites, that have been developed for thermal insulation applications. The sample set includes monolithic aerogels (polyurethane), silica aerogel composites (aerogel – fibers, aerogel impregnated foams, aerogel bound with polymer foam), glued fumed silica-fiber boards, and nanoporous polymer foam composite boards. The silica aerogel and fumed silica composites display low resonance frequencies (down to 8 Hz for 40 mm, strongly dependent on composite type) and a low dynamic stiffness and loss factor over a wide range of static loads. The polymer aerogel, nanoporous foam and polymer-bound silica aerogel granule board all display higher resonance frequencies, consistent with their higher static stiffness. Although optimized for thermal insulation rather than vibration isolation, many of the tested silica aerogel and fumed silica composites are competitive with the best vibration isolation products on the market in terms of vibration isolation performance. The diversity of materials tested informs on which approaches, materials, and materials combinations are most promising to target vibration isolation.

An isolator with high static low dynamic stiffness (HSLDS) is an efficient instrument to expand its range of low frequency. This paper explains what the nonlinear vibration isolator with the HSLDS and reports several HSLDS passive vibration isolators(PVIs), such as vertical mechanical spring combining with Euler bar in series horizontal mechanical spring, vertical pneumatic isolator in parallel with maglev isolator, maglev isolator connected with rubber ligaments, and so on. The harmonic balance method is efficient to analyze the transmissibility of the HSLDS isolator. In this paper, the stability of the HSLDS is discussed. The increasing damping shrinks the unstable region. We propose a method for the readers how to construct, design, and research the vibration isolation performance(VIP) of an HSLDS isolator.

As a missile flies at high speed, the index and surface shape of the conformal dome will vary under the influence of the aero-optical effect, which will degenerate the performance of the seeker’s imaging detection system. However, many previous studies on aero-optical imaging deterioration of optical domes were usually carried out under fixed operating conditions, which are not in line with the real flight scene of the missile. In addition, the aero-optical imaging degeneration of the dome is diverse as the flight time increases. Therefore, it is of great significance to study the time characteristics of aero-optical imaging degradation of optical domes under variable work conditions. In this study, taking an air-to-air missile as an example, the Zernike polynomials, wavefront aberration, Strehl ratio, and image simulation are applied to evaluate the aero-optical imaging deterioration of an ellipsoidal dome in the flight time range of 0–10 s under variable working conditions. The simulation results show that, as the flight time increases, (1) the dynamic range of tilt, defocus, astigmatism, coma, and wavefront aberration increase; and (2) the Strehl ratio and the peak signal-to-noise ratio (PSNR) of the simulated images decrease. Therefore, the influence of flight time on the aero-optical image degradation of the ellipsoidal dome gradually becomes more serious.

Fault diagnosis plays a very important role in ensuring the safe and reliable operations of machines. Currently, the deep learning-based fault diagnosis is attracting increasing attention. However, fault diagnosis under variable working conditions has been a significant challenge due to the domain discrepancy problem. This problem is also unavoidable in deep learning-based fault diagnosis methods. This paper contributes to the ongoing investigation by proposing a new approach for the fault diagnosis under variable working conditions based on STFT and transfer deep residual network (TDRN). The STFT was employed to convert vibration signal to time-frequency image as the input of the TDRN. To address the domain discrepancy problem, the TDRN was developed in this paper. Unlike traditional deep convolutional neural network (DCNN) methods, by combining with transfer learning, the TDRN can make a bridge between two different working conditions, thereby using the knowledge learned from a working condition to achieve a high classification accuracy in another working condition. Moreover, since the residual learning is introducing, the TDRN can overcome the problems of training difficulty and performance degradation existing in traditional DCNN methods, thus further improving the classification accuracy. Experiments are conducted on the popular CWRU bearing dataset to validate the effectiveness and superiority of the proposed approach. The results show that the developed TDRN outperforms those methods without transfer learning and/or residual learning in terms of the accuracy and feature learning ability for the fault diagnosis under variable working conditions.

A quasi-zero-stiffness (QZS) vibration isolation system presents the characteristics of high-static-low-dynamic, which can effectively isolate ultralow-frequency vibrations. However, most QZS vibration isolation systems are highly susceptible to external loads, causing deviations from the equilibrium position, which degrades their vibration isolation performance and severely restricts the range of QZS. To overcome these limitations, a novel vibration isolator was proposed by integrating a QZS structure with a variable stiffness and variable damping (VSVD) structure. The stiffness and damping of the vibration isolation system can be regulated to resist variable mass loads, and the QZS vibration isolation band can be broadened using the nonlinear positive stiffness of the conical rubber spring to effectively suppress ultralow-frequency vibrations. The dynamic characteristics of the QZS-VSVD vibration isolator were analyzed using the harmonic balance method. The effects of damping coefficients, excitation amplitude, and mass load on the amplitude–frequency characteristics and displacement transmissibility of the vibration isolation system were investigated. A virtual prototype of the QZS-VSVD vibration isolator was established using the ADAMS software, and vibration simulations were conducted. The results indicate that the QZS-VSVD vibration isolator can achieve ultralow-frequency vibration isolation at approximately 1 Hz under variable mass loads ranging from 3300 N to 3700 N. When the mass load was 3500 N, the amplitudes of the isolated object were attenuated by 70.6% and 86.2% under displacement excitation at 3 Hz and 5 Hz, respectively. Compared to linear isolators and a typical QZS structure, the proposed QZS-VSVD vibration isolator exhibited a greater load-bearing capacity, stronger anti-interference capability, and broader vibration isolation bandwidth.