Kemal S. Arsava

Nonlinear system identification of smart structures under high impact loads

The main purpose of this paper is to develop numerical models for the prediction and analysis of the highly nonlinear behavior of integrated structure control systems subjected to high impact loading. A time-delayed adaptive neuro-fuzzy inference system (TANFIS) is proposed for modeling of the complex nonlinear behavior of smart structures equipped with magnetorheological (MR) dampers under high impact forces. Experimental studies are performed to generate sets of input and output data for training and validation of the TANFIS models. The high impact load and current signals are used as the input disturbance and control signals while the displacement and acceleration responses from the structure–MR damper system are used as the output signals. The benchmark adaptive neuro-fuzzy inference system (ANFIS) is used as a baseline. Comparisons of the trained TANFIS models with experimental results demonstrate that the TANFIS modeling framework is an effective way to capture nonlinear behavior of integrated structure–MR damper systems under high impact loading. In addition, the performance of the TANFIS model is much better than that of ANFIS in both the training and the validation processes.

Fuzzy Control for Impact Mitigation of Coastal Infrastructure Equipped with Magnetorheological Dampers

ABSTRACT Arsava, K.S., Kim, Y., and Kim, K.H., 2016. Automatic control for hazard mitigation of coastal infrastructures. In: Vila-Concejo, A.; Bruce, E.; Kennedy, D.M., and McCarroll, R.J. (eds.), Proceedings of the 14th International Coastal Symposium (Sydney, Australia). Journal of Coastal Research, Special Issue, No. 75, pp. 1037 - 1041. Coconut Creek (Florida), ISSN 0749-0208. This paper proposes to develop an acceleration feedback-based smart fuzzy controller for coastal bridge piers employing magnetorheological (MR) dampers. The proposed system reduces nonlinear structural responses to collision forces so that the risk of collapse is minimized. Based on the American Association of State Highway and Transportation Officials (AASHTO, 2012) specifications, a scaled coastal bridge pier was manufactured to mimic a full scale coastal bridge pier that has three columns and a pier cap. The scaled coastal bridge pier is equipped with an MR damper, accelerometers, an LVDT, strain gages, load cells, data acquisition systems, and a voltage-current converter. A variety of experimental studies were conducted on the smart coastal bridge pier under a variety of impact loads and control signals to generate sets of input data to train the proposed smart fuzzy controller. Passive and the proportional integral derivative controller (PID) were used as baselines. The performances of controllers were compared with those of the uncontrolled system in order to determine which system effectively reduces the collision responses of the coastal bridge pier. Comparisons of the smart fuzzy controller with the benchmark controllers demonstrated that the smart fuzzy controller is the most effective way to mitigate the structural collision response of integrated coastal bridge pier-MR damper systems subjected to various impact loadings.

High Impact Force Attenuation Of ReinforcedConcrete Systems

This paper presents the application of smart control technology to reinforced concrete systems for high impact force attenuation. The smart reinforced control system is developed through the integration of reinforced concrete beams, smart control devices, sensors, and signal processors. Various combinations of input and output signals are applied to the smart reinforced concrete beams: impact forces and current signals are used as input signals while accelerations and displacements are used as output signals. Based on the set of input-output signals, a nonparametric model is developed to predict the nonlinear behaviour of smart concrete beam systems. It is demonstrated from the experimental studies that the proposed model is effective in predicting the nonlinear impact responses of the reinforced concrete systems.

Nonlinear Analysis of Bridge Piers under Barge Collision Load

This paper proposes nonlinear analyses of bridge pier-barge collision. A reinforced concrete bridge pier, a precast concrete deck and a Jumbo Hopper (JH) barge are modeled using LS-DYNA. The impact forces and structural responses of the reinforced concrete pier subjected to various barge impact loads are investigated. The American Association of State Highway and Transportation Officials (AASHTO) specifications and the previous studies on barge-pier collisions are used as baselines. It is demonstrated that the proposed finite element analysis (FEA) is effective to estimate the nonlinear behavior of the reinforced concrete bridge pier under barge impact loading.

System Identification Of Smart Structures

This paper proposes a fuzzy model for predicting nonlinear behaviour of smart structures. The parameters of the fuzzy model are trained using the backpropagation neural network and least squares algorithms. To demonstrate the effectiveness of the proposed Takagi-Sugeno (TS) fuzzy model, a structure equipped with magnetorheological (MR) dampers is constructed and investigated. Various levels of high impact loads and current signals are used as disturbances and control signals, respectively. It is demonstrated from the experimental studies that the proposed TS fuzzy model is effective in estimating the high impact responses of the smart structural systems subjected to a variety of high impact loads.