A5023238883- Granular flow and fluidized beds
- Landslides and related hazards
- Fuzzy Logic and Control Systems
- Particle Dynamics in Fluid Flows
- Robot Manipulation and Learning
- Manufacturing Process and Optimization
- Neural Networks and Applications
- Robotic Mechanisms and Dynamics
- Advanced Surface Polishing Techniques
- Proteins in Food Systems
- Fuzzy Systems and Optimization
- Advanced Control Systems Optimization
- Additive Manufacturing and 3D Printing Technologies
- Pickering emulsions and particle stabilization
- NMR spectroscopy and applications
- Color Science and Applications
- Metallurgy and Material Forming
- Iterative Learning Control Systems
- Textile materials and evaluations
- Reinforcement Learning in Robotics
- Surfactants and Colloidal Systems
- Rheology and Fluid Dynamics Studies
- Diamond and Carbon-based Materials Research
- Robotic Path Planning Algorithms
- Machine Fault Diagnosis Techniques
Sheffield Hallam University
2002-2021
Nottingham Trent University
1995-2003
Real-time acquisition of large amounts machine operating data is now increasingly common due to recent advances in Industry 4.0 technologies. A key benefit factory operators this scale the ability perform real-time condition monitoring and early-stage fault detection diagnosis on industrial machinery-with potential reduce down-time thus costs. The main contribution work development an intelligent method capable these streams provide early developing problems under variable conditions. We...
Demand for alternative sources of protein ingredients due to increasing population growth, environmental sustainability, and consumer demand is on the rise. However, processes used generate these new food must be maximized ensure a circular economy sustainable production system. In this study, structural functional properties faba bean isolate recovered fractions were investigated. Three different with varying purities (35–89 %) using pH conditions fractions. Electrophoresis revealed major...
Shock waves and granular vacua are important phenomena for studying the behavior of materials due to dramatic change in flow properties across shock wave particle-free feature at boundary vacuum. In this paper, we use experiment numerical simulation study free-surface past a cylindrical obstacle an inclined chute, where time-dependent development impacting is analyzed both microscopic macroscopic scales using discrete element method (DEM) depth-averaged model, respectively. Using high-speed...
The research reported in this paper is related to the creation of self-adapting robots that are capable learning manipulative skills online. investigation includes design a novel neural network controller (NNC), which based on adaptive resonance theory (ART) and dynamic knowledge base, whose regulated by specific assembly operations. A force/torque (F/T) sensor was attached robot's wrist only information available NNC during operations, since precise location components unknown. enhanced...
The paper discusses the employment of different sources information to support robotic assembly operations. During component interaction, part wrist sensed force and torque is found be geometrically dependent. This enables real-time sensorial data retrieved from scene combined with on geometry history insertion itself. As a result, an intelligent control architecture developed perform simple peg-hole operations emphasising aspects which relate learning appropriate state-action mapping...
Granular materials are widely encountered in food processing, but understanding their behavior and movement mechanisms remains the early stages of research. In this paper, we present our recent modeling simulation work on chute granular flow using both discrete element method (DEM) continuum method. Based data, apply machine learning techniques such as Random Forest, Linear Regression, Ridge Regression to evaluate effectiveness these models predicting patterns. The study consist soft-sphere...
The successful mating of mechanical components can only be achieved by determining appropriate corrective actions to compensate for initial angular and linear misalignment. automation assembly processes calls systems which provide a flexible nonlinear behaviour due the large uncertainties intrinsic in selection correct action. These issues indicate strongly that an artificial neural network lead effective controllers field robotic assembly. In this paper we describe design intelligent...
The food industries are continuously facing enormous challenges in the formation of stable w/o emulsions. stability emulsions is a key factor when being used as primary for double (DE). Long-term DE still non-solved challenge. Therefore, mixture various surfactants or chemical composition Polyglycerol Polyricinoleate (PGPR) surfactant stabilizing emulsion scientific interest to alleviate this problem. In order stabilize water droplets, two lipophilic surfactants, namely PGPR: Sorbitan...
Robot programming is a difficult, complex and time consuming operation. It consists of three main stages, the definition points/locations, program coding (including error planning) finally proving. Due to problems associated with each these alternative techniques are sought reduce duration, simplify complexity improve success application. Task level (TLP), which has been goal for robotics researchers many years, aims robot ease implementation operation by embedding significant knowledge into...
The work introduces a neural network (NN) algorithm capable of merging the fast and stable learning behaviour offered by adaptive resonance theory (ART) advantageous properties reinforcement agent. result is ART-R particularly suited to state-action mappings in control applications. A real time example addressing typical problem found autonomous robotic assembly discussed highlight achievement unsupervised an optimal behaviour.
The novel idea of an orthogonal fuzzy rule-based system is introduced in this paper. It shown that a where rules are orthogonal, each rule induces its maximum effect on the final decision. Using 18 rules, model for solder paste printing stage SMT created. predicted percentage deposit volume from and non-orthogonal models compared with actual values. shows less error 85% data points.
The new breed of solder paste deposition systems, such as the DEK ProFlow, incorporate a contained chamber enabling independent control machine parameters, pressure, down force and print speed. Previous work by Ekere et al. (1994) Monaghan (1991-2) have modelled conventional squeegee but only part this is applicable to mechanism. Practical experimentation has been employed, both gain an appreciation printing characteristics in general serve reference for purely analytical models. Flow...