A5053016322- Speech and Audio Processing
- Manufacturing Process and Optimization
- Product Development and Customization
- Advanced Manufacturing and Logistics Optimization
- Digital Transformation in Industry
- Geological and Geophysical Studies
- Advanced Adaptive Filtering Techniques
- Target Tracking and Data Fusion in Sensor Networks
- Seismic Imaging and Inversion Techniques
- Indoor and Outdoor Localization Technologies
- Direction-of-Arrival Estimation Techniques
- Hydrocarbon exploration and reservoir analysis
- Vehicle Routing Optimization Methods
- Radar Systems and Signal Processing
- AI and Big Data Applications
- Cloud Computing and Resource Management
- Blockchain Technology Applications and Security
- Service and Product Innovation
- Artificial Intelligence in Healthcare
- Technology Assessment and Management
- Advanced Technologies in Various Fields
- Optimization and Packing Problems
- Advanced Multi-Objective Optimization Algorithms
- Security in Wireless Sensor Networks
- Speech Recognition and Synthesis
Nanjing University of Aeronautics and Astronautics
2020-2022
Chengdu University of Technology
2021
GNS Science
2009-2015
Huazhong University of Science and Technology
2011-2014
Shanghai University
2013
PLA Navy General Hospital
2011
The number of service offerings in cloud manufacturing continues to grow, so that level agreement (SLA) within is increasingly used ensure the coordination cross-organisational business processes between providers (CSP) and consumers (CSC). definition SLA monitoring SLA's performance are significant should be highly valued effectiveness efficiency services. At present, third-party methods based on credit mainly applied environments. There no effective supervision mechanism monitor third...
With the rapid growth of manufacturing data, it is becoming an important issue to collect, manage, analyze and utilize big data generated in cloud services mode. In order solve above problems, characteristics difference from traditional were analyzed first. Then, core technologies service platform grouped into a roadmap which was guidance for research hierarchy. The architecture also discussed give overall solution each part based on this kind application scenario proposed. Thereby novel...
Abstract This paper describes a multidisciplinary integrated workflow for mapping Late Cretaceous to Recent subsurface structure and stratigraphy, modeling petroleum systems of the Taranaki Basin (74,209 km2), New Zealand. The basin-scale assessment geology fluid flow is aimed improve previous, mostly local, studies. size study area integration over nine-year timeframe large datasets, multiple software disciplines present technical challenges. consists following six main tasks: data...
Aiming at the problems of less study sample, large network scale and long training time existing in current fault diagnosis field, we develop a new method based on KPCA selective neural ensemble. First, reducing data size by using to extract sample features. Then achieving ensemble improved binary particle swarm optimization algorithm (IBPSOSEN), combining two methods for diagnosis. In algorithm, bagging is used take number different sets samples solve problem samples. Finally, simulation...
A wireless sensor network (WSN) is a large collection of nodes with limited power supply and constrained computational capability. Clustering routing method in networks has been considered as an important field research recently to prolong the lifetime WSNs. We present novel clustering that can balance energy consumption extend WSN. Network be divided into densely connected subgroups through algorithm detecting community structure complex networks. Moreover, role cluster-head scheduled among...
The purpose of this paper is to improve the performance traditional optimum space-time adaptive processing (STAP). signal model and basic idea STAP are given, then a robust algorithm propounded. In algorithm, linear estimation used estimate direction arrival. Theoretical analysis simulation results show that Robust can SINR robustness algorithm.
Due to the fluctuation of RSSI (Received Signal Strength Indicator) value, positioning accuracy indoor localization algorithm that based on value is not high and unstable, filtering reference distance used in this paper. Fluctuation amplitude reduced by mean filtering, a known between anchor nodes introduced when calculating unknown blind node node. According result experiment, new can reach 1.45 meter, it meet demand positioning.
The paper proposed a new tolerance analysis method: unified and deviation model is adopted to synthesize different kinds of tolerances; an integrated propagation designed construct the functional relationship between tolerances quality requirements. method simpler more widely applicable than traditional linearization method. A case study provided illustrate
Information-theory has been used to research the manufacturing system complexity for decade years. In order describe more accurately, assembly is aimed in this paper. The hierarchy structure of first analyzed. classified as complexity, process and control complexity. improve model, furthermore static dynamic which can ideal actual production respectively. Then an integrated model proposed based on classification.
Aiming at general constraints in vehicle routing problem workshop, this paper improve and simplify some relative to make it more line with real conditions workshop. Different situations are discussed experimentally computed respectively. The results show that tabu search is effective get satisfactory algorithmic solutions, can be expanded meet additional constraints.
This paper designed a moving average sampling method for small samples, further (MA) control chart and cumulative sum (MACS) respectively, calculated the in-control out-of-control run length both charts. The charts are robust, which can monitor process state effectively without knowing distribution. Through analyzing costs quality loss that is related to production lot size, parameters reasonably optimized. By comparing lengths some numerical examples, finds MACS has good performance on...
Aiming at the quality management issues in manufacturing enterprise, this paper establishes product lifecycle (PLQM) system based on architecture of SOA and knowledge computer integrated manufacturing. This assembly-centered PLQM collects analyzes data by integrating different information systems. The system-crossing process, mechanism problem prevention are also included to realize tracing run-to-zero quickly. Therefore, provides decision making supports for plan, inspection analysis...
Relying on the background of constructing digital workshop, an improved RSSI (Received Signal Strength Indicator) location algorithm is introduced in this paper. At first, application requirements are discussed from cargo tracking, instrument management and automatic guided vehicles three aspects; then focuses improvements new algorithm. The test results analysis show that positioning accuracy up to 0.89 meter by applying algorithm, it can meet workshop.
The paper proposed a new systematic method to construct the functional relationship between product quality and tolerances. In method, unified user-defined tolerance model is designed synthesize different kinds of tolerances using 3-D state space model; deviation propagation analyze quality-tolerance function. Particularly, suitable for any requirements. successfully applied turbo-generator stator-core lamination auto-assembly project, Monte Carlo used simulate performance optimize target
This study provides the first New Zealand-wide assessment of hydrocarbon seal rock potential twenty sedimentary basins and is designed to help evaluate pre-drill risk assessment. Seal quality has been assessed using key petroleum wells seismic lines. Deep-sea drilling research holes provide glimpses stratigraphy in some more distal basins. The Taranaki Basin currently Zealand's only producing basin — all others are frontier many have no or few exploration rather sparse coverage. We produced...
The purpose of this paper is to estimate the interference subspace quickly. fundamental Multistage Wiener Filter(MWF) given. shortage MWF pointed out and then a new real-time tracking algorithm propounded. Theoretical analysis simulation results show that could be tracked quickly by utilized algorithm.
The purpose of this paper is to estimate the interference subspace quickly. We show that solution proposed constrained optimization problem results signal subspace. Then a novel tracking algorithm propounded. Theoretical analysis and simulation could be tracked quickly by utilized algorithm.
Hydrocarbon exploration on the East Coast of North Island has not yet yielded significant commercial reserves, even though elements a working petroleum system are all present (Field et al, 1997). Exploration focussed shallow, Neogene part succession, built up during plate margin convergence over last ∼24 million years. Convergent margins generally characterised by low-total organic carbon (TOC) source rocks and poor clastic reservoir quality due to sorting labile grains. However,...
Suffering from limited singing voice corpus, existing synthesis (SVS) methods that build encoder-decoder neural networks to directly generate spectrogram could lead out-of-tune issues during the inference phase. To attenuate these issues, this paper presents a novel acoustic model with independent pitch encoder and phoneme encoder, which disentangles information music score fully utilize corpus. Specifically, according equal temperament theory, is constrained by metric loss maps distances...