Aiming at the problem that a single neural network model has difficulty in accurately predicting trends of the remaining useful life of a rolling bearing, a method of predicting the remaining useful life of rolling bearings using a gated recurrent unit-deep autoregressive model (GRU-DeepAR) with an adaptive failure threshold was proposed. First, time domain and frequency domain features were extracted from the rolling bearing vibration signal. Second, its operation process was divided into a smooth operation stage and degradation stage according to the trend of the accumulated root mean square of maximum. Then, the failure threshold for different bearings were determined adaptively by the maximum of the smooth operation data. The degradation dataset of a rolling bearing was subsequently obtained. In the meantime, a GRU-DeepAR model was built to obtain predictions of the failure time and failure probability. Appropriate model parameters were determined after a large number of tests to assure the effectiveness and prediction accuracy. Finally, the trend of time series and failure times were predicted by inputting the degradation dataset into the GRU-DeepAR model. Experiments showed that the proposed method can effectively improve the accuracy of the remaining useful life prediction of a rolling bearing with good stability.
In light of the problems of a single vibration feature containing limited information on the degradation of rolling bearings, the redundant information in high-dimensional feature sets inaccurately reflecting the reliability of rolling bearings in service, and assessments of the degradation performance being disturbed by outliers and false fluctuations in the signal, this study proposes a method of assessing rolling bearings’ performance in terms of degradation using adaptive sensitive feature selection and multi-strategy optimized support vector data description (SVDD). First, a high-dimensional feature set of vibration signals from rolling bearings was extracted. Second, a method combining the Technique for Order Preference by Similarity to an Ideal Solution (TOPSIS) and K-medoids was used to comprehensively evaluate the features with multiple evaluation indicators and to adaptively select better degradation features to construct the sensitive feature set. Next, multi-strategy optimization of the SVDD model was carried out by introducing the autocorrelation kernel regression (AAKR) and a multi-kernel function to improve the ability of the evaluation model to overcome outliers and false fluctuations. Through validation, it could be seen that the method in this study uses samples of rolling bearings in the healthy early stage to establish the evaluation model, which can adaptively determine the starting point of the bearing’s degradation. The stability and accuracy of the model were effectively improved.
Reasonable techniques and methods in biogas slurry application are significant for the promotion of biogas slurry and the improvement of crop quality in agricultural production. To investigate the impacts of different biogas slurry application techniques on the water use efficiency, growth, yield, and quality of tomatoes, three irrigation techniques, and two application methods were considered in this study. The three irrigation techniques are alternate partial root-zone irrigation (APRI), fixed partial root-zone irrigation (FPRI), and two sides root-zone irrigation (TSRI). Two application methods refer to applying the biogas slurry with hole irrigation and surface irrigation. In addition, principal component analysis (PCA) and technique for order preference by similarity to ideal solution (TOPSIS) methods were adopted to evaluate the comprehensive quality and comprehensive indicators of tomatoes among different treatments. There are three hole irrigation treatments, T1 (APRI), T2 (TSRI), T3 (FPRI), and three surface irrigation treatments, T4 (APRI), T5 (TSRI), and T6 (FPRI) were set in two-season pot experiments. The results show that the plant height, dry matter accumulation, fruit yield, and water use efficiency present a similar descending trend for APRI, TSRI, and FPRI under the same methane irrigation method, yet show that the hole irrigation treatment was higher than the surface irrigation treatment for the same irrigation technique. These indicate that the coupling of APRI technique and hole irrigation is more conducive to the increase of plant production and water use efficiency. Meanwhile, T1 treatment can significantly improve the soluble sugar, sugar-acid ratio, VC content, soluble protein, soluble solid content, and firmness of tomato fruits, which are better for the taste, storage, and transportation of tomato fruit. The titratable acid content in tomato fruit is the highest in T2 treatment, followed by T5 treatment, indicating that TSRI technique may result in an accumulation of titratable acid and is not conducive to the taste of the tomato. The comprehensive nutritional quality and index evaluation results show that T1 treatment ranks the highest among all treatments, and can be used as an optimal irrigation method for the implementation of integrated water/biogas slurry.
One basis for the real-time management of furrow irrigation is to estimate the infiltration parameters and Manning roughness, which can represent the furrow condition as early as possible. Closed-end furrow irrigation experiments were conducted on 45 fields on the Guanzhong Plain in Shaanxi province, China. The infiltration parameters of the Kostiakov equation and Manning roughness were estimated according to multipoint advance trajectory and upstream end water depth at four advance distances (50%, 60%, 70%, and 100% of furrow length; i.e., 0.5 L, 0.6 L, 0.7 L, and 1.0 L, respectively). The estimated infiltration parameter and Manning roughness were then input into WinSRFR to evaluate their effects on furrow irrigation and determine a reasonable distance for collecting multipoint advance trajectory and upstream end water depth. The results indicated that the estimated infiltration parameters and Manning roughness by multipoint advance trajectory and upstream end water depth at 0.6 L or later were highly consistent with the measured values. Except for the upstream end water depth, the mean absolute percentage errors between the simulated advance trajectory, irrigation performance, and applied water volume with measured values were lower than 10% for all furrow irrigation experiments. Therefore, the reasonable distance for collecting irrigation data should not be less than 0.6 L.
Accurately estimating the soil wetting pattern that closely reflects the measured value can improve the water use efficiency for drip irrigation. Ignoring the effect of the initial soil water content on the soil wetting pattern affects the accuracy of the estimated results to a certain extent. This research aimed to develop a soil wetting pattern estimation model for drip irrigation that included four easily measurable parameters (i.e., initial soil water content, saturated hydraulic conductivity, total volume of applied water, and emitter discharge rate) based on dimensional analysis theory. In this study, the wetting front advance data of 12 typical soil textures were obtained in Hydras-2D/3D. The estimated values were then compared with measured or simulated wetting front advance values. For different experiments, the mean absolute error, root mean square error, and mean relative error varied from 2.77 to 4.69 cm, 6.20 to 10.61 cm, and 5.61% to 10.51%, respectively. Compared with the existing models, the proposed model was more consistent between the measured and simulated values. Therefore, the proposed model of this study is efficient and simple, which can help accurately estimate the soil wetting pattern of drip irrigation with a variety of soil textures.
scite is a Brooklyn-based organization that helps researchers better discover and understand research articles through Smart Citations–citations that display the context of the citation and describe whether the article provides supporting or contrasting evidence. scite is used by students and researchers from around the world and is funded in part by the National Science Foundation and the National Institute on Drug Abuse of the National Institutes of Health.
hi@scite.ai
10624 S. Eastern Ave., Ste. A-614
Henderson, NV 89052, USA
Copyright © 2024 scite LLC. All rights reserved.
Made with 💙 for researchers
Part of the Research Solutions Family.