Rasoul Heydari Dastjerdi; Ghasem Ahmadi; Mahmood Dadkhah; Ayatollah Yari
Abstract
This paper presents a novel approach using artificial neural networks to solve the SEIR (Susceptible, Exposed, Infected, and Recovered) model of infectious diseases based on dynamical systems. Optimal control techniques are employed to determine a vaccination schedule ...
Read More
This paper presents a novel approach using artificial neural networks to solve the SEIR (Susceptible, Exposed, Infected, and Recovered) model of infectious diseases based on dynamical systems. Optimal control techniques are employed to determine a vaccination schedule for a standard SEIR epidemic model. The multilayer perceptron is utilized to approximate the state and co-state functions of the SEIR model and to solve the optimal control problem by utilizing a nonlinear programming approach. By constructing a loss function and using Pontryagin's Minimum Principle (PMP) for the SEIR model, a minimization problem is defined, a minimization problem is defined, and the approximate solution of the Hamiltonian system is computed. This method is compared with the fourth-order Runge-Kutta method. The proposed approach's effectiveness is demonstrated through illustrative examples.
Mohammad Dehghandar; Ghasem Ahmadi; Heydar Aghebatbeen Monfared
Abstract
The purpose of this study was to provide a fuzzy system for predicting and diagnosing metabolic syndrome (MetS) in preschoolers, children, and adolescents. In this study, previous research on the factors affecting metabolic syndrome, especially in children, ...
Read More
The purpose of this study was to provide a fuzzy system for predicting and diagnosing metabolic syndrome (MetS) in preschoolers, children, and adolescents. In this study, previous research on the factors affecting metabolic syndrome, especially in children, and adolescents, has been considered. After integrating the initial variables, a fuzzy system has been designed with 8 data on age, waist size, systole blood pressure, diastole blood pressure, body mass index (BMI), waist-to-height ratio, nutrition, and abdominal obesity. Ultimately, the system gives us an output that diagnoses the health status of a child or adolescent with MetS or predicts the possibility of a person contracting the disease in the future. The system is designed based on the data of 1300 persons participating in the fifth study of the program for monitoring and prevention of non-communicable diseases of children, and Adolescents in Tehran and Isfahan provinces that 1050 data were used as training data and 250 data as test data that used to test the rules and output of the system. After reviewing the rules and eliminating similar or contradictory rules using their degree calculation, finally, the system was designed with 45 rules, a multiplication inference engine, a single fuzzifier, and a centroid defuzzifier. Then the system was evaluated using the confusion matrix accuracy, sensitivity, and specificity. Our analysis shows that this method, with an error rate of less than 4 percent more accurate than other methods, can predict and diagnose MetS in children.
Ghasem Ahmadi
Abstract
Rough extreme learning machines (RELMs) are rough-neural networks with one hidden layer where the parameters between the inputs and hidden neurons are arbitrarily chosen and never updated. In this paper, we propose RELMs with a stable online learning algorithm for the identification ...
Read More
Rough extreme learning machines (RELMs) are rough-neural networks with one hidden layer where the parameters between the inputs and hidden neurons are arbitrarily chosen and never updated. In this paper, we propose RELMs with a stable online learning algorithm for the identification of continuous-time nonlinear systems in the presence of noises and uncertainties, and we prove the global asymptotically convergence of the proposed learning algorithm using the Lyapunov stability theory. Then, we use the proposed methodology to identify the chaotic systems of Duffing's oscillator and Lorentz system. Simulation results show the efficiency of the proposed model.
Control and Optimization
Ghasem Ahmadi; Mohammad Teshnehlab; Fahimeh Soltanian
Abstract
o enhance the performances of rough-neural networks (R-NNs) in the system identification, on the base of emotional learning, a new stable learning algorithm is developed for them. This algorithm facilitates the error convergence by increasing the memory depth of R-NNs. ...
Read More
o enhance the performances of rough-neural networks (R-NNs) in the system identification, on the base of emotional learning, a new stable learning algorithm is developed for them. This algorithm facilitates the error convergence by increasing the memory depth of R-NNs. To this end, an emotional signal as a linear combination of identification error and its differences is used to achieve the learning laws. In addition, the error convergence and the boundedness of predictions and parameters of the model are proved. To illustrate the efficiency of proposed algorithm, some nonlinear systems including the cement rotary kiln are identified using this method and the results are compared with some other models.
Control and Optimization
Alaeddin Malek; Ghasem Ahmadi; Seyyed Mehdi Mirhoseini Alizamini
Volume 1, Issue 1 , April 2016, , Pages 55-67
Abstract
Linear semi-infinite programming problem is an important class of optimization problems which deals with infinite constraints. In this paper, to solve this problem, we combine a discretization method and a neural network method. By a simple discretization ...
Read More
Linear semi-infinite programming problem is an important class of optimization problems which deals with infinite constraints. In this paper, to solve this problem, we combine a discretization method and a neural network method. By a simple discretization of the infinite constraints,we convert the linear semi-infinite programming problem into linear programming problem. Then, we use a recurrent neural network model, with a simple structure based on a dynamical system to solve this problem. The portfolio selection problem and some other numerical examples are solved to evaluate the effectiveness of the presented model.
