جستجوی مقالات مرتبط با کلیدواژه « radial basis function » در نشریات گروه « مکانیک »

In this paper, a fairly comprehensive comparative study has been done on online identification methods for the dynamic model of an aircraft system. To this aim, first, the existing algorithms in this field are introduced. Then, some of these approaches including the recursive least square, the recursive extended least square, the recursive instrumental variable, the extended matrix, the radial basis function neural network, and the multilayer perceptron neural network are utilized to identify the aircraft model. To carry out the simulations, and train the neural networks, the linearized model and online data of the Boeing 747 aircraft controlled by a sliding mode controller on an arbitrary reference trajectory are employed. Finally, the efficiency of each of the above-mentioned methods is evaluated and compared to the other approaches. According to the obtained results of this research, the radial basis function neural network method has a significantly superior performance over the other algorithms due to dynamic noise estimation, independence from the system model, rejecting the linear model of the system, and higher accuracy while maintaining the appropriate speed.

Continuously varying cross-section members have found wide applications in engineering for cost and resistance optimization. Since steel structures generally have more slender members compared to concrete structures, buckling analysis of steel members is of more importance. Determining the critical load of functionally varying cross-section columns using the analytical solution is a time-consuming process. In this paper, buckling analysis of non-prismatic steel columns is conducted using the meshless local Petrov-Galerkin (MLPG) method. In meshless methods, the scattered nodes are used rather than the elements to model the problem domain and its boundaries. The change of the inertia moment within the length of a column is characterized by introducing a power function with variable taper ratio and exponent. The radial basis function is used to discretize the differential equation governing the buckling. The penalty method is used for the imposition of the boundary conditions. Numerical examples of the critical buckling load for prismatic and non-prismatic columns using the proposed method are compared with the analytical solution, and the effectiveness of the MLPG method for buckling analysis of non-prismatic columns is validated. Also, buckling analysis of muscle column members subjected to non-uniform axial load is carried out to show the efficiency of the proposed method. The effect of several parameters such as non-uniformity of the load and variation of the cross-section on the buckling load of the column is discussed in details.