A neural network-based multi-agent classifier system with a Bayesian formalism for trust measurement

In this paper, a neural network (NN)-based multi-agent classifier system (MACS) utilising the trust-negotiation-communication (TNC) reasoning model is proposed. A novel trust measurement method, based on the combination of Bayesian belief functions, is incorporated into the TNC model. The Fuzzy Min-Max (FMM) NN is used as learning agents in the MACS, and useful modifications of FMM are proposed so that it can be adopted for trust measurement. Besides, an auctioning procedure, based on the sealed bid method, is applied for the negotiation phase of the TNC model. Two benchmark data sets are used to evaluate the effectiveness of the proposed MACS. The results obtained compare favourably with those from a number of machine learning methods. The applicability of the proposed MACS to two industrial sensor data fusion and classification tasks is also demonstrated, with the implications analysed and discussed.

Effects of load modeling in power distribution system with distributed wind generation

This paper presents the impact of different types of load models in distribution network with distributed wind generation. The analysis is carried out for a test distribution system representative of the Kumamoto area in Japan. Firstly, this paper provides static analysis showing the impact of static load on distribution system. Then, it investigates the effects of static as well as composite load based on the load composition of IEEE task force report [1] through an accurate time-domain analysis. The analysis shows that modeling of loads has a significant impact on the voltage dynamics of the distribution system with distributed generation.

Designing a responsive material system with physical computing

This paper focuses on an investigation to explore architectural design potentials with a responsive material system and physical computing. Contemporary architects and designers are seeking to integrate physical computing in responsive architectural designs; however, they have largely borrowed from engineering technology's mechanical devices and components. There is the opportunity to investigate an unexplored design approach to exploit the responsive capacity of material properties as alternatives to the current focus on mechanical components and discrete sensing devices. This opportunity creates a different design paradigm for responsive architecture that investigates the potential to integrate physical computing with responsive materials as one integrated material system. Instead of adopting highly intricate and expensive materials, this approach is explored through accessible and off-the-shelf materials to form a responsive material system, called Lumina. Lumina is implemented as an architectural installation called Cloud that serves as a morphing architectural skin. Cloud is a proof of concept to embody a responsive material system with physical computing to create a reciprocal and luminous architectural intervention for a selected dark corridor. It represents a different design paradigm for responsive architecture through alternative exploitation of contemporary materials and parametric design tools. © 2014, The Association for Computer-Aided Architectural Design Research in Asia (CAADRIA), Hong Kong.

Autonomous capillary microfluidic system with embedded optics for improved troponin I cardiac biomarker detection

 Cardiovascular diseases are the most prevalent medical conditions affecting the modern world, reducing the quality of life for those affected and causing an ever increasing burden on clinical resources. Cardiac biomarkers are crucial in the diagnosis and management of patient outcomes. In that respect, such proteins are desirable to be measured at the point of care, overcoming the shortcomings of current instrumentation. We present a CO2 laser engraving technique for the rapid prototyping of a polymeric autonomous capillary system with embedded on-chip planar lenses and biosensing elements, the first step towards a fully miniaturised and integrated cardiac biosensing platform. The system has been applied to the detection of cardiac Troponin I, the gold standard biomarker for the diagnosis of acute myocardial infarction. The devised lab-on-a-chip device was demonstrated to have 24 pg/ml limit of detection, which is well within the minimum threshold for clinically applicable concentrations. Assays were completed within approximately 7–9 min. Initial results suggest that, given the portability, low power consumption and high sensitivity of the device, this technology could be developed further into point of care instrumentation useful in the diagnosis of various forms of cardiovascular diseases. 2014 Elsevier B.V. All rights reserved.

Applying a multi-agent classifier system with a novel trust measurement method to classifying medical data

In this paper, we present the application of a Multi-Agent Classifier System (MACS) to medical data classification tasks. The MACS model comprises a number of Fuzzy Min-Max (FMM) neural network classifiers as its agents. A trust measurement method is used to integrate the predictions from multiple agents, in order to improve the overall performance of the MACS model. An auction procedure based on the sealed bid is adopted for the MACS model in determining the winning agent. The effectiveness of the MACS model is evaluated using the Wisconsin Breast Cancer (WBC) benchmark problem and a real-world heart disease diagnosis problem. The results demonstrate that stable results are produced by the MACS model in undertaking medical data classification tasks. © 2014 Springer Science+Business Media Singapore.

Fuzzy system with tabu search learning for classification of motor imagery data

This paper introduces an approach to classify EEG signals using wavelet transform and a fuzzy standard additive model (FSAM) with tabu search learning mechanism. Wavelet coefficients are ranked based on statistics of the Wilcoxon test. The most informative coefficients are assembled to form a feature set that serves as inputs to the tabu-FSAM. Two benchmark datasets, named Ia and Ib, downloaded from the brain-computer interface (BCI) competition II are employed for the experiments. Classification performance is evaluated using accuracy, mutual information, Gini coefficient and F-measure. Widely-used classifiers, including feedforward neural network, support vector machine, k-nearest neighbours, ensemble learning Adaboost and adaptive neuro-fuzzy inference system, are also implemented for comparisons. The proposed tabu-FSAM method considerably dominates the competitive classifiers, and outperforms the best performance on the Ia and Ib datasets reported in the BCI competition II.

Control and charge management of a grid-connected photovoltaic system with plug-in hybrid vehicle as energy storage

This paper presents the control and charge management strategy of a photovoltaic system (PV) with plug-in hybrid electric vehicle (PHEV) as energy storage. The hybrid energy storage system (HESS) of PHEV consists of battery and supercapacitor. A simulation model for the PV system with PHEV energy storage has been developed using Matlab/SimpowerSystems. The system consists of PV arrays, SEPIC dc-dc converter with maximum power point tracking (MPPT), hybrid battery-supercapacitor energy storage with bidirectional dc-dc converter and inverter for grid connection. A charge management algorithm for the hybrid energy storage system is proposed to control the power flows among the PV system, energy storage and the grid. Results show that the proposed power management algorithm can control the power flows in an efficient manner.

Rainfall-runoff modeling using dynamic evolving neural fuzzy inference system with online learning

This is an open access article under the CC BY-NC-ND license.Neuro-Fuzzy Systems (NFS) are computational intelligence tools that have recently been employed in hydrological modeling. In many of the common NFS the learning algorithms used are based on batch learning where all the parameters of the fuzzy system are optimized off-line. Although these models have frequently been used, there is a criticism on such learning process as the number of rules are needed to be predefined by the user. This will reduce the flexibility of the NFS architecture while dealing with different data with different level of complexity. On the other hand, online or local learning evolves through local adjustments in the model as new data is introduced in sequence. In this study, dynamic evolving neural fuzzy inference system (DENFIS) is used in which an evolving, online clustering algorithm called the Evolving Clustering Method (ECM) is implemented. ECM is an online, maximum distance-based clustering method which is able to estimate the number of clusters in a data set and find their current centers in the input space through its fast, one-pass algorithm. The 10-minutes rainfall-runoff time series from a small (23.22 km2) tropical catchment named Sungai Kayu Ara in Selangor, Malaysia, was used in this study. Out of the 40 major events, 12 were used for training and 28 for testing. Results obtained by DENFIS were then compared with the ones obtained by physically-based rainfall-runoff model HEC-HMS and a regression model ARX. It was concluded that DENFIS results were comparable to HEC-HMS and superior to ARX model. This indicates a strong potential for DENFIS to be used in rainfall-runoff modeling.