Impact of Climate Change on Future Flood Susceptibility: an Evaluation Based on Deep Learning Algorithms and GCM Model

Floods are common and recurring natural hazards which damages is the destruction for society. Several regions of the world with different climatic conditions face the challenge of floods in different magnitudes. Here we estimate flood susceptibility based on Analytical neural network (ANN), Deep learning neural network (DLNN) and Deep boost (DB) algorithm approach. We also attempt to estimate the future rainfall scenario, using the General circulation model (GCM) with its ensemble. The Representative concentration pathway (RCP) scenario is employed for estimating the future rainfall in more an authentic way. The validation of all models was done with considering different indices and the results show that the DB model is most optimal as compared to the other models. According to the DB model, the spatial coverage of very low, low, moderate, high and very high flood prone region is 68.20%, 9.48%, 5.64%, 7.34% and 9.33% respectively. The approach and results in this research would be beneficial to take the decision in managing this natural hazard in a more efficient way.

     

Generalized Eigenvalue Decomposition Applied to Estimation of Spatial rPPG Distribution of Skin

Journal of Mathematical Imaging and Vision - Remote photoplethysmography (rPPG) has been at the forefront recently, thanks to its capacity in estimating non-contact physiological parameters such as...     

Eco-friendly Geopolymer Composite Based on Non-heat-treated Phosphate Sludge Reinforced With Polypropylene Fibers

Silicon - Geopolymers produced with metakaolin (MK) and thermally untreated phosphate sludge (PS) are beneficial and environmentally advantageous materials, but their fragility limits its...     

Lignin‐based polycondensation resins for wood adhesives

Lignin‐based wood adhesives are obtained that satisfy the requirements of relevant international standards for the manufacture of exterior‐grade wood particleboard. Formulations based on low molecular mass lignin and presenting an increase in the relative proportion of reactive points yield better results than the higher molecular mass lignin used in the past. These lignins allow a higher proportion of hydroxymethylation during preparation of methylolated lignins. These lignin‐based adhesives also yield acceptable results at particleboard pressing times that are sufficiently low to be of industrial significance. Lignin‐based wood adhesives, in which a nonvolatile nontoxic aldehyde (glyoxal) is substituted for formaldehyde in their preparation, are prepared and tested for application to wood panels such as particleboard. The adhesives yield good internal bond strength results for the panels, which are good enough to comfortably pass relevant international standard specifications for exterior‐grade panels. The adhesives also show sufficient reactivity to yield panels in press times comparable to that of formaldehyde‐based commercial adhesives. © 2006 Wiley Periodicals, Inc. J Appl Polym Sci 103: 1690–1699, 2007     

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.

     

Using AR,MA, and ARMA Time Series Models to Improve the Performance of MARS and KNN Approaches in Monthly Precipitation Modeling under Limited Climatic Data

Precipitation is one of the most important components of the hydrologic cycle as it is required for multi-objective applications including flood estimation, drought monitoring, watersheds management, hydrology, agriculture, etc. Therefore, its estimation and modeling via a suitable method is a challenging task for hydrologists. The present study seeks to model monthly precipitation at two stations located in Iran. Two artificial intelligence (AI)-based models consisting of multivariate adaptive regression splines (MARS) and k-nearest neighbors (KNN) were used as the modeling techniques. In doing so, nine single-input scenarios under limited climatic data are implemented using minimum, maximum, and mean air temperatures, dew point temperature, station pressure, vapor pressure, relative humidity, wind speed, and antecedent precipitation data. The attained results illustrate that the performance of single MARS and KNN is relatively poor when modeling the monthly precipitation. Additionally, this study develops hybrid models to enhance the precipitation modeling through combining the MARS and KNN models with three diverse types of the time series (TS) models, namely autoregressive (AR), moving average (MA), and autoregressive moving average (ARMA). The most important justification for integrating the models applied is that the AI and TS-based models are respectively capable of modeling the non-linear and linear terms of the hydrological variables such as precipitation. It is therefore necessary to be considered both of the aforementioned terms in the modeling procedure. A performance comparison of the single and hybrid models denotes the higher accuracy of hybrid models than the single ones. However, the hybrid models generated by combining the KNN and the TS models used are the best-performing models.

     

Electrokinetic power generation of non-Newtonian fluids in a finite length microchannel

In order to examine the effects of the fluid type as the electrolyte solvent on the efficiency of electrokinetic energy conversion, a comparative numerical study among three different fluid types of a transient electrokinetic flow through a single circular finite length microchannel has been conducted. The system was initially at an equilibrium non-flow state, and a step change in flow was applied and the calculation proceeding until steady state was achieved. The analysis was based on non-dimensional transport governing equations that were scaled using Debye length as the characteristic length scale and diffusion time as characteristic time scale. The fluid types considered were shear thinning, Newtonian, and shear thickening, and a power law modeled them with the scaled flow behavior index having values of 0.2, 1.0, and 1.8. In order to isolate the electrokinetic effects of the different relationships between the shear strain rate and shear stress, the flow consistency index was adjusted so that in all the cases the flow rate and total pressure drop matched that of water at 25 °C. All other fluid and interfacial properties were the same for all cases. The key observational difference between the various fluid types was that their different axial velocity profile acted on essential the same free charge density profiles. Consequently, the convection current density (i.e., the radial distribution of charge being advected along the channel) was strongly affected by the fluid type. Integration of this quantity to calculate the convection current showed that for the particular fluid properties chosen the shear thinning fluid was 20 % higher than the Newtonian fluid, while the shear thickening fluid was only 4 % lower than the Newtonian fluid. Combined with the effects, these different currents have on the streaming potential, the shear thinning fluid was 50 % more effective in converting flow work to electrical work than the Newtonian fluid, while the shear thickening fluid was only 16 % lower than the Newtonian fluid.     

A particle swarm optimization for a fuzzy multi-objective unrelated parallel machines scheduling problem

This paper proposes a novel multi-objective model for an unrelated parallel machine scheduling problem considering inherent uncertainty in processing times and due dates. The problem is characterized by non-zero ready times, sequence and machine-dependent setup times, and secondary resource constraints for jobs. Each job can be processed only if its required machine and secondary resource (if any) are available at the same time. Finding optimal solution for this complex problem in a reasonable time using exact optimization tools is prohibitive. This paper presents an effective multi-objective particle swarm optimization (MOPSO) algorithm to find a good approximation of Pareto frontier where total weighted flow time, total weighted tardiness, and total machine load variation are to be minimized simultaneously. The proposed MOPSO exploits new selection regimes for preserving global as well as personal best solutions. Moreover, a generalized dominance concept in a fuzzy environment is employed to find locally Pareto-optimal frontier. Performance of the proposed MOPSO is compared against a conventional multi-objective particle swarm optimization (CMOPSO) algorithm over a number of randomly generated test problems. Statistical analyses based on the effect of each algorithm on each objective space show that the proposed MOPSO outperforms the CMOPSO in terms of quality, diversity and spacing metrics.