This article presents an event-triggered data-driven load frequency control (LFC) method for multiarea interconnected power systems via model-free adaptive control, where the dynamic model of system is assumed to be unknown completely. By introducing linearization technique system, equivalent data relationship between area-control-error (ACE) and input signal established. Then, a LFC scheme developed only relying on output system. Meanwhile, strategy also proposed in design such that...

The problem of waste classification has been a major concern for both the government and society, whether can be effectively classified will affect sustainable development human society. To perform fast efficient detection targets in sorting process, this paper proposes data augmentation + YOLO_EC system. First all, because current shortage multi-objective datasets, heavy workload collection, limited improvement features by traditional methods, DCGAN (deep convolution generative adversarial...

Abstract This brief proposes a model predictive control method using preceding vehicle information within hybrid electric vehicles' (HEVs') cruise system to improve car following performance and reduce fuel consumption. paper adds two original contributions the related literature. First, real‐time optimization approach Pontryagin's minimum principle with analytical methods rather than numerical iteration is proposed. Second, compute desired battery state of charge trajectory as function...

A modern production plant may consist of several parallel-running batch processes, and the monitoring such processes is imperative. This paper proposes a multiset canonical correlation analysis (MCCA)-based joint-individual scheme for which considers individual feature each process joint features shared by all processes. First, fourway data are unfolded into two-way time-slice data. Second, MCCA performed at time instant to extract throughout running Then, process, measurements projected...

This paper addresses the problem of stability and stabilization for a class two-dimensional (2-D) Roesser systems with time-varying delays subject to missing measurements. The phenomenon sensor measurement is governed by stochastic variable satisfying Bernoulli random binary distribution. aim this focused on design state feedback controller such that closed-loop 2-D system asymptotic in mean square sense. A delay-dependent condition derived terms linear matrix inequalities, formulas can be...

Removing the haze in an image is a huge challenge due to difficulty of accurate hazy modeling. Although atmospheric scattering model (ASM) widely used describe formation images, it hard deal with uneven image, once ASM restricted assumption that atmosphere distributed homogeneously. This paper analyzes imaging mechanism, then proposes image-to-image architecture handle dehazing, which heterogeneous twin network (HT-Net) two parallel sub-networks are constructed establish high dimensional...

This paper discusses the issues of finite-time state estimation and switching signal identification for a class switched linear system with both unknown inputs active modes. By some special output transformations, reduced-order is constructed such that observer can be constructed, where original are removed. A cluster observers proposed all subsystems. Then, derived from by choosing any small time-delay parameters, also obtained estimated states based on equivalent transformation. Next, kind...

Traditional text detection mainly relies on manual features, which are only applicable to simple environments and have limited generalisation capabilities. Although deep learning enhances the robustness of detection, complex contexts still face challenges. Current CNN algorithms difficult handle large-scale long-distance due limitation receiving domain spatial information extraction. This chapter proposes GMSTNet model, combines GhostNet V2, MobileNet V3, Swin Transformer enhance efficiency...

The rapid expansion of Internet technology into remote areas has not only broadened network coverage but also facilitated the proliferation terminal devices. This enhancement in infrastructure boosts data resources, which are essential for advancing intelligence and automation linked to Fourth Industrial Revolution. However, a significant challenge remains: many older devices still offline require updates. Technologies such as scene text recognition, crucial applications autonomous driving...

Applying model-based learning for the optimal decision of multi-agent system is not trivial due to expensive price or even impossibility obtaining ground truth training model complex environment. Such as action hydraulic supports in top-coal caving, could accessible corresponding state intricate processes. Regarding latent hidden variable an effective method Markov model. This paper extends and proposes random field (HMRF) with reinforcement optimizing multi-agent. In HMRF model, input...

The quality of WiFi's RSSI have greatly affected the power consumption mobile phone. We analyze impact on smartphone under different network environment. It can be concluded, weak or bad signal strength makes link speed decreases, so will increased. Corresponding, good WiFi energy decreased. Taking this basic concept, we present an optimization strategy based perception, which packet aggregate and delay transmission for intervals. On other hands, reduce transition mode when is bad....

This paper considers the problem of H∞ iterative learning control design for network-based uncertain systems with communication constraints, including data quantization and random dropouts. It is assumed that input update signals are quantized before they transmitted to controller a logarithmic quantizer used decode number levels. Then, dropout modeled by conventional Bernoulli variable describe successful transmission or miss. The 2-D dynamic such ILC process established stochastic Roesser...

In this paper, we consider using finite time switched observer to estimate the states and detect faults for a type of linear system with unknown inputs. We first design by reduced-order which originates from primary system, where inputs are removed help state output transformations. A cluster observers presented all subsystems. Then, choosing any small parameters, can get estimation observers. And is acquired equivalent Next, put forward fault detection method residual. Finally, MATLAB...

Aiming at the situation of lacking automation and informatization for tunnel construction ventilation, this article proposed a monitoring system based on fieldbus technology. The consisted three parts including monitor cabinet, local control cabinet sensors. A program, which was written by WinCC software, ran cabinet. All environmental parameters working could be displayed in program. Meanwhile, Programmable Logical Computer (PLC) used as master station Modbus network, collected all from...

Currently, the widely used methods for direction of arrival (DOA) estimation were constructed based on subspace, such as Multiple Signal Classification (MUSIC) and Estimating Parameter via Rotational Invariance Techniques (ESPRIT), which needed to know number sources in advance. In this paper, a new model Generalized Reference Curve Model (GRCM) DOA was proposed, do not need And comparison performance between proposed MUSIC given demonstrate effectiveness our method. The algorithm...

Mathematical description of a complex signal is very pivotal but nearly impossible in many engineering cases. The emergency the Generative Adversarial Network (GAN) shows possibility to train network be an Arbitrary Signal Generator (ASG), which only controlled by random vector with several elements. This paper designs ASG model based on 1D GAN, suitable for calculation or curve one dimension. Then, two sets sine wave and triangular are generated program as training samples model. Several...

Mathematical description of a complex signal is very important in engineering but nearly impossible many occasions. The emergence the Generative Adversarial Network (GAN) shows possibility to train single neural network be Specific Signal Generator (SSG), which only controlled by random vector with several elements. However, there no explicit criterion for GAN training process stop, and real applications always stops after certain big iteration. In this paper, serious issue was discussed...

Currently, there are many multivariate linear regression algorithms being used for predicting the bioactive capacity of herbal formulae or extracts from their chromatographic fingerprints, such as Principal Component Regression (PCR), Partial Least Squares (PLSR), Orthogonal Projections to Latent Structures (OPLS) and Elastic Net (EN). In this study, performance complexity predictive models developed by PCR, PLSR, OPLS, EN evaluated compared using a set fingerprints Astragali Radix (AR)...

This paper proposes a model named Independent Component Analysis with Reference Curve (ICARC) to extract and remove artifact signal from Electroencephalogram (EEG).Firstly, an additional requirement priori information are introduced directly into the contrast function of traditional ICA model.Then, augmented Lagrangian is formed based on this new model.Finally, iterative solution calculated by using Newton method.The simulations experiments implemented indicate performance our comparing...

After researching deeply on mine electromechanical equipment maintenance management, this paper has developed management system based ASP.NET 2.0 technology and database technology. This includes the information related to such as device information, repair project fitting course review report personnel so on. And functions that realized are warning reminding of with multi-way, query retrieval file fittings' stock, declaration all reports, by short message, certification setting through USB...