Assured learning-enabled autonomy: A metacognitive reinforcement learning framework

Reinforcement learning (RL) agents with pre-specified reward functions cannot provide guaranteed safety across variety of circumstances that an uncertain system might encounter. To guarantee performance while assuring satisfaction of safety constraints across variety of circumstances, an assured autonomous control framework is presented in this article by empowering RL algorithms with metacognitive learning capabilities. More specifically, adapting the reward function parameters of the RL agent is performed in a metacognitive decision-making layer to assure the feasibility of RL agent. That is, to assure that the learned policy by the RL agent satisfies safety constraints specified by signal temporal logic while achieving as much performance as possible. The metacognitive layer monitors any possible future safety violation under the actions of the RL agent and employs a higher-layer Bayesian RL algorithm to proactively adapt the reward function for the lower-layer RL agent. To minimize the higher-layer Bayesian RL intervention, a fitness function is leveraged by the metacognitive layer as a metric to evaluate success of the lower-layer RL agent in satisfaction of safety and liveness specifications, and the higher-layer Bayesian RL intervenes only if there is a risk of lower-layer RL failure. Finally, a simulation example is provided to validate the effectiveness of the proposed approach.     

An experimental study on electrical effect on asphaltene deposition

Asphaltene generally existed in colloidal form in cruds and will precipitate in non-equilibrium conditions. Asphaltene instability may take place in the reservoir leading to permeability damage and contributing to flow restriction issues. It may also occur in production strings and surface facilities causing pipe blockage. Any change in oil composition or pressure and temperature at any stage of production will destabilize crude oil producing asphaltene precipitation. In this paper, the stability of target crude oil under the influence of a direct current and contacting with polar fluid, water, is investigated. The amount of the asphaltene deposit and its electrical charge at various operating conditions are investigated. The fact that deposits form on the anode surface proves that asphaltene particles possess a positive charge. The amounts of asphaltenes precipitation were increased considerably by increasing water as polar component.     

Internet connection with UMTS

Mobile telecommunication new services are based on data networks specially Internet. These services include http, telnet, ftp, Simple Mail Transfer Protocol, etc. Besides, we recognize a mobile network as a multiuser network. Transmission Control Protocol (TCP)/Internet Protocol which is sensitive to link congestion in wireline data links is also used in wireless networks. In order to improve the system performance, the TCP layer uses flow control and congestion control. Besides, radio link control (RLC) and medium access control sublayers have been introduced to compensate the deficiency of TCP layer in wireless environment. RLC has an important role in quality of service enhancement of the Universal Mobile Telecommunications System (UMTS). In this paper, we review the protocol stack of UMTS Terrestrial Radio Access Network which is based on Third-Generation Partnership Project. Then, we evaluate its layer 2 error control mechanisms and verify TCP over automatic repeat request error control mechanism and finally quality of service improvement results from it in fading channels.     

Hyperelastic model analysis of stress‐strain behavior in polybutadiene/ethylene‐propylene diene terpolymer nanocomposites

Micromechanics of elastomer nanocomposite samples based on polybutadiene (BR), ethylene‐propylene diene terpolymer (EPDM) hyperelastic matrixes prepared via melt compounding was investigated using uniaxial tensile analysis. Constitutive hyperelastic models, including Polynomial, Yeoh, Ogden, Arruda‐Boyce, and Van der Waals were used to determine material parameters in incompressible isotropic elastic strain‐energy functions on the basis of a nonlinear least squares optimization method by fitting the data obtained from uniaxial classic experiments. Effect of nanoclay (0, 3, 5, 7, and 10 phr) content on the simulation accuracy was investigated. Simulation results compared with the experimental data suggested that the Ogden model as the most consistent model investigated here. J. VINYL ADDIT. TECHNOL., 23:21–27, 2017. © 2015 Society of Plastics Engineers     

Hydrogen-dependent denitrification in an alternating anoxic-aerobic SBR membrane bioreactor.

A novel hydrogenotrophic denitrification system, which consisted of a sequencing batch membrane bioreactor, was evaluated for simultaneous removal of nitrate and soluble microbial products (SMP) from a synthetic groundwater feed. A hollow fiber membrane diffuser was used for bubble-less diffusion of hydrogen into the bioreactor under anoxic condition followed by aerobic SMP removal and biomass filtration. During the anoxic period, the nitrate loading of 0.328 kg N m(-3) d(-1) was completely denitrified to below detectable levels. A denitrification rate of 0.8 kg N m(-3) d(-1) was obtained at steady state biomass concentrations of 1,162 mg I(-1). During the aerobic period when biomass filtration was performed, 81% of SMP produced within the anoxic phase was retained by the membrane, 9% was biologically removed, 5% was passed through the membrane and 5% was discharged during the wasting of mixed liquor. The aerobic cycle was instrumental as it allowed for effective biomass filtration via membrane scouring and assisted in further reduction of effluent organic matter.     

Determination of a model partial safety factor for unreinforced masonry walls under buckling

The design/verification method in the Eurocodes is based on the partial safety concept. Eurocode 6 suggests a constant partial safety factor for the material γ_M for all design/verification problems, without consideration of model uncertainty in the design/verification formula. In the following, a model partial safety factor is determined for the problem of unreinforced masonry walls mainly subjected to vertical loading. For that purpose, the newly proposed formula for EC 6, annex G will be considered [1–3]. In order to cover all aspects in tests and to use the results for design purposes, several methods have been included in EN 1990 Annex D for design based on test data. In this study, the recommended methods in Annex D of EN 1990 for resistance of the material are used to extract the partial safety factors. A database including more than 119 experimental tests on unreinforced masonry shear walls is used to compare the model prediction and the test results and to determine the model partial safety factor.     

Recognizing faces using Adaptively Weighted Sub-Gabor Array from a single sample image per enrolled subject

In this paper, we propose a new approach for face representation and recognition based on Adaptively Weighted Sub-Gabor Array (AWSGA) when only one sample image per enrolled subject is available. Instead of using holistic representation of face images which is not effective under different facial expressions and partial occlusions, the proposed algorithm utilizes a local Gabor array to represent faces partitioned into sub-patterns. Especially, in order to perform matching in the sense of the richness of identity information rather than the size of a local area and to handle the partial occlusion problem, the proposed method employs an adaptively weighting scheme to weight the Sub-Gabor features extracted from local areas based on the importance of the information they contain and their similarities to the corresponding local areas in the general face image. An extensive experimental investigation is conducted using AR and Yale face databases covering face recognition under controlled/ideal condition, different illumination condition, different facial expression and partial occlusion. The system performance is compared with the performance of four benchmark approaches. The promising experimental results indicate that the proposed method can greatly improve the recognition rates under different conditions.