NN-based Prediction Interval for Nonlinear Processes Controller

Neural networks (NNs) are extensively used in modelling, optimization, and control of nonlinear plants. NN-based inverse type point prediction models are commonly used for nonlinear process control. However, prediction errors (root mean square error (RMSE), mean absolute percentage error (MAPE) etc.) significantly increase in the presence of disturbances and uncertainties. In contrast to point forecast, prediction interval (PI)-based forecast bears extra information such as the prediction accuracy. The PI provides tighter upper and lower bounds with considering uncertainties due to the model mismatch and time dependent or time independent noises for a given confidence level. The use of PIs in the NN controller (NNC) as additional inputs can improve the controller performance. In the present work, the PIs are utilized in control applications, in particular PIs are integrated in the NN internal model-based control framework. A PI-based model that developed using lower upper bound estimation method (LUBE) is used as an online estimator of PIs for the proposed PI-based controller (PIC). PIs along with other inputs for a traditional NN are used to train the PIC to predict the control signal. The proposed controller is tested for two case studies. These include, a chemical reactor, which is a continuous stirred tank reactor (case 1) and a numerical nonlinear plant model (case 2). Simulation results reveal that the tracking performance of the proposed controller is superior to the traditional NNC in terms of setpoint tracking and disturbance rejections. More precisely, 36% and 15% improvements can be achieved using the proposed PIC over the NNC in terms of IAE for case 1 and case 2, respectively for setpoint tracking with step changes.

     

FQL‐RED: an adaptive scalable schema for active queue management

In a potentially congested network, random early detection (RED) active queue management (AQM) proved effective in improving throughput and average queuing delay. The main disadvantage of RED is its sensitive parameters that are impossible to estimate perfectly and adjust manually because of the dynamic nature of the network. For this reason, RED performs differently during different phases of a scenario and there is no guarantee that it will have optimal performance. Giving adaptability to RED has been the subject of broad research studies ever since RED was proposed. After a substantial study of AQM schemes and presenting a novel categorization for so‐called modern approaches utilizing artificial intelligence tools to improve AQM, this paper proposes an algorithm enhancing RED as an add‐on patch that makes minimal changes to the original RED. Being built on the basis of AQM schemes like ARED and Fuzzy‐RED, this algorithm inherits adaptability and is able to adjust RED inaccurate parameters regarding network traffic status, trying to optimize throughput and average queuing delay in a scenario. This algorithm is a Q‐learning method enhanced with a fuzzy inference system to provide RED with self‐adaptation and improved performance as a result. Given the name of FQL‐RED, this algorithm outperformed RED, ARED, and Fuzzy‐RED, as the OPNET simulations show. Copyright © 2010 John Wiley & Sons, Ltd.     

Arbitrary dual-band unequal Wilkinson power divider based on negative refractive index transmission lines

This paper focused on the application of negative refractive index transmission line (NRI-TL) in dual-band unequal Wilkinson power divider (WPD) with controllable frequency and power dividing ratio. Theory and design procedure of the dual-band NRI-TL are presented in details. For demonstration, two dual-band unequal Wilkinson power dividers (WPDs) with power dividing ratio of 2 : 1 and operating frequencies of 0.9 and 1.8 GHz are designed, fabricated and tested. The first unequal divider is based on 2-stage NRI-TLs and the second one is based on 4-stage NRI-TLs. In addition, these two types of NRI-TLs are presented to demonstrate that by increasing the number of NRI-TL unit-cells the phase response of the NRI-TLs converge to the desired characteristic. The good agreement between measured and simulated results confirmed the design concept and derived closed-form design equations. Measurements show that the first divider has 18.37% and 21.86% relative bandwidths and the second one has 33.52% and 29.12% relative bandwidths at 0.9 and 1.8 GHz, respectively. The design concept of this paper can be extended to equal dual-band power dividers with arbitrary frequency ratio.     

Cooperative caching for content dissemination in vehicular networks

The explosive growth of mobile data traffic has made cellular operators to seek low‐cost alternatives for cellular traffic off‐loading. In this paper, we consider a content delivery network where a vehicular communication network composed of roadside units (RSUs) is integrated into a cellular network to serve as an off‐loading platform. Each RSU subjecting to its storage capacity caches a subset of the contents of the central content server. Allocating the suitable subset of contents in each RSU cache such that maximizes the hit ratio of vehicles requests is a problem of paramount value that is targeted in this study. First, we propose a centralized solution in which, we model the cache content placement problem as a submodular maximization problem and show that it is NP‐hard. Second, we propose a distributed cooperative caching scheme, in which RSUs in an area periodically share information about their contents locally and thus update their cache. To this end, we model the distributed caching problem as a strategic resource allocation game that achieves at least 50% of the optimal solution. Finally, we evaluate our scheme using simulation for urban mobility simulator under realistic conditions. On average, the results show an improvement of 8% in the hit ratio of the proposed method compared with other well‐known cache content placement approaches.     

Non-Intrusive Online Quality of Experience Assessment for Voice Communications

Recently user quality of experience (QoE) is employed in evaluating end user satisfaction in communications systems. Generally, current approaches for QoE assessment are obtrusive, laboratory based and offline. Estimation of user satisfaction in static manner based on mean opinion score is not directly related to instantaneous individual end user contentment. In this paper, based on correlations between user’s physiological signals and her/his feelings about the service quality, a non-intrusive and user centric QoE assessment system for voice communications is developed. The findings of this study indicate that the emotional patterns in response to the changes in channel quality can be adapted to estimate the level of satisfaction in a QoE assessment system in a live manner. Based on experimental results, two categories of users are identified: sensitive and insensitive towards quality degradations. The results indicate that for the sensitive users, our non-intrusive subjective quality assessment method outperforms ITU-T P.563 standard with respect to root mean square error; while, the results are much better among the insensitive users.     

Blind separation of image sources via adaptive dictionary learning

Sparsity has been shown to be very useful in source separation of multichannel observations. However, in most cases, the sources of interest are not sparse in their current domain and one needs to sparsify them using a known transform or dictionary. If such a priori about the underlying sparse domain of the sources is not available, then the current algorithms will fail to successfully recover the sources. In this paper, we address this problem and attempt to give a solution via fusing the dictionary learning into the source separation. We first define a cost function based on this idea and propose an extension of the denoising method in the work of Elad and Aharon to minimize it. Due to impracticality of such direct extension, we then propose a feasible approach. In the proposed hierarchical method, a local dictionary is adaptively learned for each source along with separation. This process improves the quality of source separation even in noisy situations. In another part of this paper, we explore the possibility of adding global priors to the proposed method. The results of our experiments are promising and confirm the strength of the proposed approach.     

A Novel Approach for Prediction of Monthly Ground Water Level Using a Hybrid Wavelet and Non-Tuned Self-Adaptive Machine Learning Model

Water Resources Management - In recent decades, due to groundwater withdrawal in the Kabodarahang region, Iran, Hamadan, hazardous events such as sinkholes, droughts, water scarcity, etc., have...     

Parametric study of a 2-d model of the nip region in a counter-rotating,non-intermeshing twin screw extruder

Within the polymer processing industry, the demand for more efficient mixing within the existing mixer geometries and keeping a tighter control over increasingly difficult applications calls for a quantitative understanding of the role of the mixer geometry on the mixing performance. An important extruder geometry to consider is the Counter-Rotating, Non-Intermeshing Twin Screw Extruder (CRTSE). Recent studies suggest that the greater than linear mixing performance in this geometry may be attributed to the flow in the nip region. Solution of the fluid dynamic equations for a 2-D model of the nip region was found in order to quantify mixing using a FEM software called FIDAP which was available to us on the CRAY X-MP supercomputer at San Diego Supercomputer Center. In order to study the effects of design parameters in the nip region, an arbitrary interface was placed in the entrance of the region. A program called “LINE” tracks the path and growth of the interface through the region. Measures of mixing are computed and compared. Effect of several design parameters on mixing performance are studied. Conclusions are presented as to the relative effect of various design parameters. Discussions regarding the measures of mixing and mixing efficiency are presented.     

Improvement in volume resistivity and morphology of a blend of polyolefin elastomer with linear low-density polyethylene

A silane moisture-cured polyolefin elastomer/linear low-density polyethylene (LLDPE) blend was prepared through a two-step silane-grafting method (Sioplas Process) in an industrial scale twin-screw extruder. The silane-grafted compound was used to make wire and cable coatings. In this work, the effect of some interactive parameters on quality of the products prepared by the above method has been studied, while so far, there have been less experimental investigations. The volume resistivity of cross-linked compound was changed from 2.96 × 10¹⁴ to 7.41 × 10¹⁴ Ω cm with increasing LLDPE component by maximum 10 wt%. Surface morphology of the product was corrected with reduction in benzoyl peroxide (BPO) concentration from 0.2 wt% to 0.13 wt%. BPO at this level acted as an initiator in grafting reaction of vinyl trimethoxysilane. The curing condition and specimen preparation method by injection molding and/or extrusion were factors which influenced the hot-set test results at 200 °C. The results of tensile and elongation studies showed a maximum value of 9 MPa and 397% for the tests, after 6 h curing. With increases in curing time at a specified temperature, the gel content of the cross-linked compound was increased and reached its maximum value. The maximum gel content values were found to be approximately 60%, 80%, and 82% at temperatures of 25, 60, and 85 °C, respectively. The hardness, density, and tear strength of the samples did not vary significantly with the curing temperature.

     

A study on the effect of compositing silver oxide nanoparticles by carbon on the ‎electrochemical behavior and electronic properties of zinc‐silver oxide batteries ‎

In this study, the effect of compositing silver oxide nanoparticles by carbon on the electrochemical behavior and electronic properties of zinc‐silver oxide batteries have been investigated. For this purpose, firstly four silver oxide electrodes containing 5, 10, 15, and 20 wt% carbon powder were produced by powder metallurgy method. For the next step, all four silver oxide electrodes were sintered at 500°C for 10 minutes. Afterward and in order to investigate the microstructure, phase and elemental analysis of the electrodes were carried out using Transmission Electron Microscopy (TEM), Scanning Electron Microscopy (SEM), X‐ray Diffraction (XRD), and Energy Dispersive Spectroscopy (EDS), respectively. Moreover, in order to investigate the effect of compositing silver oxide nanoparticles by carbon on the electrochemical behavior and electronic properties of zinc‐silver oxide, electrochemical tests (potentiodynamic polarization and electrochemical impedance spectroscopy) and electric discharge test in 1.4 wt%KOH electrolyte were carried out respectively. The microstructural observations revealed that increasing carbon content in the silver oxide electrodes results in increasing the apparent porosities in these electrodes. Investigating the phase and elemental analysis results showed that by increasing the content of carbon in the silver oxide electrode, the amount of Ag₂O and AgO phases in this electrode reduces and also the extent of pure silver formation increases. Investigations on the results of electrochemical tests showed that increasing carbon content results in the reduction of corrosion resistance in silver oxide electrodes. Moreover, the results of electric discharge test revealed that the silver oxide electrode containing 10wt% carbon yields the highest energy efficiency in the zinc‐silver oxide batteries.