A5100779924- Manufacturing Process and Optimization
- Industrial Vision Systems and Defect Detection
- Ferroelectric and Piezoelectric Materials
- Scheduling and Optimization Algorithms
- Machine Fault Diagnosis Techniques
- Advanced Manufacturing and Logistics Optimization
- Neural Networks and Applications
- Neural Networks and Reservoir Computing
- Power System Reliability and Maintenance
- Surface Roughness and Optical Measurements
- Advanced Control Systems Optimization
- Aluminum Alloy Microstructure Properties
- Additive Manufacturing and 3D Printing Technologies
- Optimization and Packing Problems
- Aluminum Alloys Composites Properties
- Advanced machining processes and optimization
- Acoustic Wave Resonator Technologies
- Assembly Line Balancing Optimization
- Ferroelectric and Negative Capacitance Devices
- Microwave Dielectric Ceramics Synthesis
- Microstructure and mechanical properties
- Environmental Quality and Pollution
- Forest, Soil, and Plant Ecology in China
- Digital Transformation in Industry
- Organic Light-Emitting Diodes Research
Dongguan University of Technology
2015-2024
Dalian Maritime University
2023
South China University of Technology
2019
Soil and Fertilizer Institute of Hunan Province
2012
An echo state network (ESN) is a recurrent neural with low computational complexity. However, single ESN cannot extract effective features from complex inputs, especially for dealing low-cost condition signals in machinery fault diagnosis. A novel deep learning model, referred to as the fuzzy (DFESN), was proposed improve feature extraction capability less burden. In present method, output data of previous reservoir were regarded abstract vectors next input. The reinforced each hidden layer...
Fault diagnosis is of importance to guarantee the printing quality and avoid unexpected downtime for 3-D printers. In this paper, a sparse echo autoencoder network (SEAEN) proposed fault delta printers using attitude data. Considering practicality economy diagnosis, data, including three-axial angular velocity signals, vibratory acceleration ones, magnetic field intensity are collected by installing low-cost sensor on moving platform printer. However, will increase chaos To make up...
Under the international background of transformation and promotion manufacturing, Chinese government proposed “Made in China 2025” strategy, which focused on improvement a quality-based innovation ability. Moreover, predicting manufacturing quality is one crucial measures for management. Accurate prediction closely related to feature learning processes. Therefore, two categories intelligent approaches, i.e., shallow deep learning, are investigated compared this paper. Specifically, feed...
An echo state network (ESN) is a type of recurrent neural that good at processing time-series data with dynamic behavior. However, the use ESNs to enhance fault-classification accuracy continues be challenging when condition signals are collected by low-cost sensors. In this paper, deep algorithm, called hybrid (DHSN), proposed for fault diagnosis three-dimensional printers using attitude low measurement precision. DHSN, output sparse auto-encoder regarded as abstract features...
Abstract Organic light-emitting diodes (OLEDs) offer the advantage of flexibility; however, use traditional transparent anode ITO limits further extension their flexible characteristics. In this study, we propose employing an polymer polybenzodifuranedione (PBFDO) as a instead rigid ITO. To address issue encountered during PBFDO solution spin-coating process, introduced n-butanol into conductive to reduce its viscosity and freezing point by modulating intermolecular hydrogen bonding...
All equipment must be maintained during its lifetime to ensure normal operation. Maintenance is one of the critical roles in success manufacturing enterprises. This paper proposed a preventive maintenance period optimization model (PMPOM) find an optimal period. By making use advantages particle swarm (PSO) and cuckoo search (CS) algorithm, hybrid algorithm PSO CS solve PMPOM problem. The test functions show that exhibits more outstanding performance than search. Experiment results has...
Expected product quality is affected by multi-parameter in complex manufacturing processes. Product prediction can offer the possibility of designing better system parameters at early production stage. Many existing approaches fail providing favorable results duo to shallow architecture model that not learn multi-parameter's features insufficiently. To address this issue, a deep neural network (DNN), consisting belief (DBN) bottom and regression layer on top, proposed paper. The DBN uses...
Intelligent fault identification of the mechanical transmission system for multi-joint industrial robots is important to guarantee safe operations. An attitude data-based intelligent approach introduced in this study. Considering that change last joint can be used reflect other connecting rods or joints, an economical data acquisition strategy proposed by only installing one sensor on joint. model subsequently established training a deep sparse auto-encoder network (DSAE) dataset. To test...
Assembly sequence planning (ASP) is a combinatorial optimization problem in which the order for each part and subassembly determined.This then incorporated into an incrementally expanding eventually results final assembly.To address this problem, we propose improved cat swarm (CSO) algorithm redefine some basic CSO concepts operations according to ASP characteristics.The feasibility stability of are verified through assembly experiment.The also compared with particle...
Abstract We report phase transition process during the solid‐state reaction of BaCO 3 ‐TiO 2 system under assistance electric field. Experiments were conducted at a constant heating rate with preset field strength and current limit. Solid‐state was completed upon reactive flash sintering taking place ~1002℃ 200 V/cm. Hexagonal BaTiO phase, which rarely occurs such temperatures, obtained after density 23.5 mA/mm . It is speculated that oxygen deficiency triggered cubic‐to‐hexagonal...
Preventive maintenance is significant for health management of production system, while the plan seldom simultaneously considered in decision-making process scheduling. This paper studies a practical scheduling problem hot metal pretreatment-steelmaking-continuous casting involving preventive consideration (denoted as PSCCPM scheduling). A mathematical programming model designed by considering constraints related to and other requirements. An effective heuristic containing machine allocation...
The integrated determination of the charge batching and casting start time (CBCST) is a combinatorial optimization problem extracted from production operations management steel plants. A hierarchical method based on variable neighborhood search (VNS) proposed in this pape r for CBCST. number casts each continuous caster charges cast are determined encoding. decoding process decomposed into solving two sub-problems. mixed integer programming (MIP) model built first sub-problem by considering...
Reservoir inflow forecast plays a crucial part in programming, development, operation, and management of water resource systems. To better reveal the complex properties daily reservoir inflow, clustered deep fusion (CDF) approach is proposed this paper. First, variational mode decomposition (VMD) used to decompose series into multiple modes, which are different sets by fuzzy c-means according Xie-Beni index view attribute domain. In each cluster, autoencoder model (DAE) developed for...
Assembly sequence planning (ASP) of remote handling maintenance in radioactive environment is a combinatorial optimization problem. This study proposes an improved shuffled frog leaping algorithm (SFLA) for the problem ASP. An ASP experiment conducted to verify feasibility and stability SFLA. Simultaneously, SFLA compared with SFLA, genetic algorithm, particle swarm optimization, adaptive mutation terms efficiency capability locating best global assembly sequence. Experiment results show...
Manufacturing quality prediction performance is influenced by multi-parameter in manufacturing multi-stage processes. To solve this problem, a two-step feature learning approach (TFLR) proposed paper. For the first step learning, (high-dimensional information) learned manifold algorithm (ML), which can enhance representation of inputs and reduce calculation burdens. second features low-dimensional information obtained ML are deep technique, learn sufficient pattern between through layer-wise...
The three-dimensional structures of the primary Al3Sc particles in an Al-2Sc master alloy were studied by synchrotron X-ray microtomography, scanning and transmission electron microscopy. phases found to be a single cube cluster cubes. surface area, equivalent diameter cubes increased with increasing volume, but specific area decreases. Al-matrix have same crystal orientation, indicating that are heterogeneous nucleation sites for Al. experimental results show α-Al2O3 possible
Identifying the most effective features and improving accuracy of health status identification have an important impact on printing reliability 3D printers. In this paper, local support vector machine (LSVM) is used as attitude monitoring method for condition recognition delta To reduce cost experiment, a cheap nine-channel sensor was installed printer's mobile platform to monitor working status. The data 13 fault types 1 normal under different conditions were collected analyze its diagnosis...