Operation optimization for modern subway trains usually requires the speed curve and tracking simultaneously. For optimization, a multi-objective seeking issue should be addressed by considering requirements of energy saving, punctuality, accurate parking, comfortableness at same time. But most traditional searching methods lack in efficiency or tend to fall into local optimum. tracking, widely applied proportional integral differential (PID) fuzzy controllers rely on complicated parameter...

The train plug door is the only way for passengers getting on and off. Its failures will make operation ineffective. Taking developed digital signal processing technologies into consideration, a data-driven diagnosis method doors proposed based sound recognition. First, novel preprocessing empirical mode decomposition hybrid intrinsic functions (IMFs) selection criterion proposed. selected significant IMFs are used to reconstruct signals. Inspired by idea of fractional calculus, entropy...

Contactless fault diagnosis is one of the most important technique for identification equipment. Based on idea contactless diagnosis, this paper presents a sound-based method railway point machines (RPMs). First, sound signals are preprocessed using empirical mode decomposition (EMD). Entropy, time-domain and frequency-domain statistical parameters first 15 intrinsic functions (IMFs) then extracted. Second, two-stage feature selection strategy blending Filter Wrapper proposed, which can...

Conventionally, the transition probabilities in interacting multiple model (IMM) are often fixed based on prior information. However, this conservative setting may result inaccurate state estimations. To solve problem, a Bayesian-based online correction function is proposed paper, which can adaptively adjust probabilities. deal with response lag and short-term peak estimation error problem during respond to jump, jumping threshold defined, so that current information of models be fully...

Considering the advantages of contactless fault diagnosis, a novel sound based diagnosis method for railway point machines (RPMs) is proposed. Firstly, denoising on empirical mode decomposition (EMD) The useful intrinsic functions (IMFs) are selected using kurtosis and energy criteria to reconstruct denoised signal. Then, multi-scale fractional permutation entropy (MFPE) proposed inspired by calculus, which more powerful than classical (MPE). And two-scale algorithm developed avoid...

When a sensor data-based detection method is used to detect the potential defects of industrial products, data are normally imbalanced. This problem affects improvement robustness and accuracy defect system. In this work, welding taken as an example: based on imbalanced radiographic images, using generative adversarial network combined with transfer learning proposed solve imbalance improve detection. First, new model named contrast enhancement conditional proposed, which creatively global...

For deep learning-based soft sensors, the lack of interpretability and consequent unreliability has become one most important problems. In this article, a neural network scheme called multiple attention sensor (DMASS), which consists solely mechanisms, is proposed to develop self-interpretable sensor. DMASS was established ensure self-interpretability data selection modeling try integrate these originally independent phases into single scheme. First, existing mechanisms' core implementation...

Many practical systems in physics, biology, engineering, and information science exhibit impulsive dynamical behaviors due to abrupt changes at certain instants during the processes. This note first introduces behavior into switched linear studies controllability observability of such systems. Necessary sufficient criteria for reachability are established. It is proved that equivalent Then, necessary determinability established by duality. also determinability. Our geometric type, they can...

Optimization of island microgrids should configure the module type and size in such a way that multiple objectives can be balanced. This paper presents bioinspired optimization approach microgrid sizing, with two salient features. First, are categorized into four types: reliability, economy, renewable technology, pollution. We present triangular aggregation model, which is straightforward cost effective to compute fitness. Second, algorithm named Levy-Harmony developed. embed Levy flight...

With the increasing demand for rail transit, wireless communication technologies are playing a growing significant role in train control systems, which enables railway systems to provide higher capacity and more efficient services. However, due nature of radio frequency propagation, quality train-to-ground connections is highly dependent on well-planned deployment wayside access points. To improve both accuracy efficiency network planning, this paper, deep learning technology exploited model...

As one of the most important railway signaling equipment, point machines undertake major task ensuring train operation safety. Thus fault diagnosis for becomes a hot topic. Considering advantage anti-interference characteristics vibration signals, this paper proposes an novel intelligent method based on signals. A feature extraction combining variational mode decomposition (VMD) and multiscale fluctuation-based dispersion entropy is developed, which verified more effective tool selection....

Currently, the fault diagnosis with balanced data and distinct characteristics has received mass concern, related research achievements are remarkable. However, because of weakness scarcity incipient signals, commonly existing in industrial systems is still an intractable problem. In order to solve problem, method based on a sliding-scale resampling strategy improved sparse autoencoder multi-particle noise addition (MpNA-SAE) proposed this paper. Firstly, original time domain signals...

A wireless monitoring network is an effective way to monitor and transmit information about railway infrastructure conditions. Its lifetime significantly affected by the energy usage among all sensors. This paper proposes a novel cluster-based valid maximization protocol (CVLMP) extend of network. In CVLMP, cluster heads (CHs) are selected rotated with selection probability information. Then, clusters determined around CHs based on multi-objective optimization model, which minimizes total...

In order to acquire a high resolution multispectral (HRMS) image with the same spectral as (MS) and spatial panchromatic (PAN) image, pansharpening, typical hot fusion topic, has been well researched. Various pansharpening methods that are based on convolutional neural networks (CNN) different architectures have introduced by prior works. However, scale information of source images is not considered these methods, which may lead loss high-frequency details in fused image. This paper proposes...

Reconfigurable intelligent surfaces (RISs) can shape the wireless environment for enhancing communication performance. In this letter, we propose a cooperative multi-RIS assisted transmission scheme millimeter-wave multi-antenna orthogonal frequency division multiplexing system. We first put forward delay matching based simultaneously estimating multipath channels and delays of distributed RISs, which requires limited training overhead feedback. Based on scheme, obtain closed-form solution...

The failure threshold of a product is not always constant when affected by random shocks. Thus, this paper establishes reliability model that combines dynamic and self-healing characteristics, which are subjected to the competing process. Soft caused degradation exceeding soft threshold, hard occurs because Degradation consists natural increment thresholds rate will also be Moreover, characteristics in process, together with increment, affect whole When shock arrives, change simultaneously....