Variational mode decomposition (VMD) provides a feasible approach to decompose vibration signals obtained from complex machinery for further applications. The frequency bandwidth control parameter and the total number of modes are critical parameters VMD. Thus, optimally automatically setting two is an essential issue VMD various practical signal sources. To this end, work proposes automatic signals. First, we use evaluation criterion mean mode-location distance evaluate sparsity modes;...

Rotating machinery is a key piece of equipment for tremendous engineering operations. Vibration analysis powerful tool monitoring the condition rotating machinery. Furthermore, vibration signals have characteristics time series. Hence, it necessary to monitor signal series avoid any catastrophic failure. To this end, paper proposes an effective strategy under hybrid method framework. First, we add variational mode decomposition (VMD) preprocess data points listed in order into subseries,...

In traditional fault diagnosis strategies, massive and disordered data cannot be utilized effectively. Furthermore, just a single parameter is used for of weapons fire control system, which might lead to uncertainty in the results. This paper proposes an information fusion method rough set theory (RST) combined with improved Dempster–Shafer (DS) evidence identify various system operation states. First, feature different faults extracted from original data, then this as state object. By...

Engine vibration is the consequence of a combination internal forces and external forces, signals contain variety information, also it easy measurement, low cost strong robustness. Hence, one most common promising methods in fault detection area analysis which effectively used diesel engine diagnosis. In this paper, new diagnosis method for with feature extraction based on ensemble empirical mode decomposition combined support vector machine was proposed. Firstly, are first preprocessed...

Partial differential equations (PDEs) are a model candidate for soft sensors in industrial processes with spatiotemporal dependence. Although physics-informed neural networks (PINNs) promising machine learning method solving PDEs, they infeasible the nonhomogeneous PDEs unmeasurable source terms. To this end, coupled PINN (CPINN) recurrent prediction (RP) strategy (CPINN- RP) is proposed. First, CPINN composed of NetU and NetG approximating solutions regularizing training NetU. The two...

Distortion of accounting information is a situation where an enterprise's financial reports are untrue or deliberately distorted, which may lead to misleading and losses for investors stakeholders. In this study, random forest algorithm used identify distortion. First, real data set containing normal distorted samples was collected. Then, according the principle feature engineering, series features related distortion extracted. Next, divided into training test set, train optimize model....

The vibration signal of diesel engine bearing has the characteristics nonlinear and is easily affected by coupling noise vibration. In this paper, we use variational modal decomposition Gray Wolf algorithm to optimize support vector machines for fault diagnosis. Firstly, VMD used decompose preprocess signals, intrinsic functions (IMF) different scales are obtained. Then, energy value entropy decomposed signals extracted construct feature vectors. Finally, SVM GWO optimized identification...

In order to efficiently diagnose the mechanical wear failure of aero-engine lubricating oil systems, a base KPCA-ABC-SVM fault diagnosis model is established based on number metal abrasive particles considering multiple indicators such as viscosity, temperature, moisture and density. Firstly, detection results obtained by feature extraction multi-parameters kernel principal component analysis (KPCA) method are used reference, then extracted values classified support vector machine (SVM);...

Partial differential equations (PDEs) are the model candidates for soft sensors in industrial processes with spatiotemporal dependence. However, gaps often exist between idealized PDEs and practical situations. Discovering proper PDE structures can remedy gaps. To this end, a coupled physics-informed neural network Akaike's information criterion (CPINN-AIC) is proposed to discover sensors. First, CPINN trained through hierarchical training strategy approximating solutions source terms...

In order to save the high cost of tank maintenance, reduce redundant input manpower and material resources for improve reliability performance, a fault diagnosis method based on NRS WOA-SVM is proposed. Taking fire control computer sensor subsystem certain type system as research object, algorithm used properties performance parameters computer, most important index selected. Then, novel meta-heuristic algorithm, WOA, optimize SVM, data classification model constructed according global best...

The characteristic information of fire control system is miscellaneous, which reflects the operation status from different levels. However, due to inaccuracy measurement process, evaluation criteria are not uniform.This will cause great ambiguity and uncertainty for real-time health state system.The traditional fuzzy comprehensive model only considers fuzziness index but ignores randomness. cloud theory both randomness things. applied system. analytic hierarchy process (AHP) obtain...

Soft sensors have been extensively used to monitor key variables using easy-to-measure and mathematical models. Partial differential equations (PDEs) are model candidates for soft in industrial processes with spatiotemporal dependence. However, gaps often exist between idealized PDEs practical situations. Discovering proper structures of PDEs, including the operators source terms, can remedy gaps. To this end, a coupled physics-informed neural network Akaike's criterion information...

Maxwell's equations are a collection of coupled partial differential (PDEs) that, together with the Lorentz force law, constitute basis classical electromagnetism and electric circuits. Effectively solving is crucial in various fields, like electromagnetic scattering antenna design optimization. Physics-informed neural networks (PINNs) have shown powerful ability PDEs. However, PINNs still struggle to solve heterogeneous media. To this end, we propose domain-adaptive PINN (da-PINN) inverse...

This paper focuses on the study of sensitivities microcantilever chemical sensors based SOI POLYMUMPS process. Through changing geometry beams and analyzing resonance frequency shift in a dynamic mode by using FEA (finite element analysis) method, most sensitive structure, which is triangle, selected out from various kinds beam designs. The relation between sensitivity parameters such as length L, width W thickness t obtained with commercial software Intellisuite. research provides primary...

In order to solve the problem of effective use massive and disorderly data in integrated detection system fire control reduce uncertainty fault diagnosis result weapon by using a single parameter traditional method, this paper proposes D-S evidence theory, combining rough set information fusion method identify type state operation. The concepts equivalence attribute decision coefficient are defined separately: reduction extracted is used, synthesis based on trust degree defined, final...

The use and maintenance support requirements of modern weapons equipment are constantly improving, the mode armored is changing from "passive maintenance" to "active maintenance". core foundation active master health status operation comprehensively accurately. This paper presents a research method strategy fire control system based on hierarchical state evaluation. Firstly, we evaluate current layer by layer; then performance degradation failure trend model established stochastic process....

In recent years, the complex equipment system presents rapid development of integration and intelligence, it is easy to break down when works in bad environment for a long time. The maintenance support ability comprehensive economic benefit play an important role defense ability. Therefore, order improve combat readiness equipment, solve current situation loss low operation caused by insufficient or excessive maintenance. this paper, dynamic strategy proposed. goal minimize average cost...

Preventing and resolving major financial risks is an important part of the three battle. Under new normal economic situation affected by COVID-19, China's environment complex changeable, problem constantly highlighted. There are many reasons for enterprise bankruptcy □ However, domestic foreign enterprises closely related to crisis. A systematic perfect management system can promote growth at beginning establishment enterprises, detect prevent possible crisis enterprises. Therefore, it...