Fast evolving neuro-fuzzy model and its application in online classification and time series prediction

A strategy for improving speed of the previously proposed evolving neuro-fuzzy model (ENFM) is presented in this paper to make it more appropriate for online applications. By considering a recursive extension of Gath?CGeva clustering, the ENFM takes advantage of elliptical clusters for defining validity region of its neurons which leads to better modeling with less number of neurons. But this necessitates the computing of reverse and determinant of the covariance matrices which are time consuming in online applications with large number of input variables. In this paper a strategy for recursive estimation of singular value decomposition components of covariance matrices is proposed which converts the burdensome computations to calculating reverse and determinant of a diagonal matrix while keeping the advantages of elliptical clusters. The proposed method is applied to online detection of epileptic seizures in addition to prediction of Mackey?CGlass time series and modeling a time varying heat exchanger. Simulation results show that required time for training and test of fast ENFM is far less than its basic model. Moreover its modeling ability is similar to the ENFM which is superior to other online modeling approaches.     

Investigating the effect of nano-silica on efficiency of the foam in enhanced oil recovery

Due to the vast production of crude oil and consequent pressure drops through the reservoirs, secondary and tertiary oil recovery processes are highly necessary to recover the trapped oil. Among the different tertiary oil recovery processes, foam injection is one of the most newly proposed methods. In this regard, in the current investigation, foam solution is prepared using formation brine, C₁₉TAB surfactant and air concomitant with nano-silica (SiO₂) as foam stabilizer and mobility controller. The measurements revealed that using the surfactant-nano SiO₂ foam solution not only leads to formation of stable foam, but also can reduce the interfacial tension mostly considered as an effective parameter for higher oil recovery. Finally, the results demonstrate that there is a good chance of reducing the mobility ratio from 1.12 for formation brine and reservoir oil to 0.845 for foam solution prepared by nanoparticles.     

Automatic navigation of mobile robots in unknown environments

Online navigation with known target and unknown obstacles is an interesting problem in mobile robotics. This article presents a technique based on utilization of neural networks and reinforcement learning to enable a mobile robot to learn constructed environments on its own. The robot learns to generate efficient navigation rules automatically without initial settings of rules by experts. This is regarded as the main contribution of this work compared to traditional fuzzy models based on notion of artificial potential fields. The ability for generalization of rules has also been examined. The initial results qualitatively confirmed the efficiency of the model. More experiments showed at least 32 % of improvement in path planning from the first till the third path planning trial in a sample environment. Analysis of the results, limitations, and recommendations is included for future work.     

Vacuum desorption of gases from inorganic solvents with various boiling points

The vacuum desorption of gases from inorganic solvents with low and high boiling points is reported. A thin-film evaporator consisting of regular tubular packing elements is considered. A laminar downward cocurrent flow of liquid and vapor is analyzed. General theoretical propositions are illustrated by the example of the desorption of simple gases from lower alcohols (methanol, ethanol) and liquid naphthalene, as well as by the example of removing the products of ester decomposition from esters at high temperatures.     

Solving a bi-objective cell formation problem with stochastic production quantities by a two-phase fuzzy linear programming approach

We propose a bi-objective cell formation problem with demand of products expressed in a number of probabilistic scenarios. To deal with the uncertain demand of products, a framework of two-stage stochastic programming model is presented. The proposed model considers minimizing the sum of the miscellaneous costs (machine constant cost, expected machine variable cost, cell fixed-charge cost, and expected intercell movement cost) and expected total cell loading variation. Because of conflicting objectives, we develop a two-phase fuzzy linear programming approach for solving bi-objective cell formation problem. To show the effectiveness of the proposed approach, numerical examples are solved and the results are compared with the two existing approaches in the literature. The computational results show that the proposed fuzzy method achieves lower objective functions as well as higher satisfaction degrees.     

Spark plasma sintering of Al-doped ZrB2–SiC composite

This research presents the influence of Al addition on microstructure and mechanical behavior of ZrB₂–SiC ultra-high temperature ceramic matrix composite (UHTCMC) fabricated by spark plasma sintering (SPS). A 2.5?wt% Al-doped ZrB₂–20?vol% SiC UHTCMC was produced by SPS method at 1900?°C under a pressure of 40?MPa for 7?min. The microstructural and phase analysis of the composite showed that aluminum-containing compounds were formed in-situ during the SPS as a result of chemical reactions between Al and surface oxide films of the raw materials (i.e. ZrO₂ and SiO₂ on the surfaces of ZrB₂ and SiC particles, respectively). The Al dopant was completely consumed and converted to the intermetallic Al₃Zr and Al₄Si compounds as well as Al₂O₃ and Al₂SiO₅. A relative density of 99.8%, a hardness (HV5) of 21.5?GPa and a fracture toughness (indentation method) of 6.3?MPa?m^1/2 were estimated for the Al-doped ZrB₂–SiC composite. Crack bridging, branching, and deflection were identified as the main toughening mechanisms.     

Sweet lemon mechanical damage detection using image processing technique and UV radiation

Non-destructive measurement of qualitative parameters of agricultural produce is quite beneficial in the postharvest operations. The consequential effect of mechanical damage in citrus fruits is rarely visible in their appearance compared to other commodities. The purpose of this study was to propose a fast, non-destructive method for sweet lemon mechanical damage detection using image processing technique and UV radiation. For this purpose, 135 sweet lemons were tested based on a completely randomized factorial design. In order to examine mechanical damage, the independent variables included drop height, fruit diameter and tempering period (holding time at room temperature after treatment). Fruits were dropped from heights of 2, 2.5 and 3 m onto the ground. Then images were captured under UV light having a wavelength of 365 nm, 1, 3 and 6 days after treatment. The images were sent to a PC and analyzed using MATLAB software. “Green Spot Index” or GSI was defined to show the extent of mechanical damage. Results of the analysis of variance showed that the percentage of green spots on fruit skin is significant at the 1% level considering the main and double interaction effects of drop height and fruit diameter. Green spot index significantly increases with the level of mechanical damage. Accuracy of the developed method in differentiating the bruised and undamaged fruits was found to be 100%.