Abstract-Adaptive Neuro-fuzzy systems constitute an intelligent systems hybrid technique that combines fuzzy logic with neural networks in order to have better results. A study is presented to forecast the relative magnetic permeability using ANFIS. The global electromagnetic parameter, namely, the magnetic induction has been used as input to estimate the relative magnetic permeability. In this exceptional research, finite element simulations are carried out to build up a database which will be used to train ANFIS network. The ANFIS approach learns the rules and membership functions from training data. The hybrid system is tested by the use of the validation data. Performance of the trained ANFIS network was compared with the multilayer feed forward network model and experimental results. The results show the effectiveness of the proposed approach in solving inverse electromagnetic problem.

Wireless power transfer (WPT) via coupled magnetic resonance is an encouraging technology to be applied in many fields. In this paper, a method using a circular coil spiral inductor structure to wirelessly transfer energy is proposed. It represents the characteristic of six parallel air core inductors mutually coupled in the free space for wireless power transfer system. Based on the analytical model and circuit theory, the relationship between the coil design parameters and the system performance is deduced, and the effects of the outer radius, inner radius, channel width, and coil turns are thoroughly studied to improve the system performance at different axial distances and in lateral misalignment. Also, an elimination method for transmission efficiency dead-zone (TEDZ) is proposed. The proposed method utilizes angular rotation of the receiver (P x) to eliminate the zero-coupling point which causes TEDZ and boosts the coupling coefficient such that the TEDZ is eliminated, and the high efficiency region is extended.

New accurate approximation is proposed using integral expressions for evaluating the magnetic force between cylindrical permanent magnet arrays. The magnetic field distribution is calculated analytically by using Coulombian model. In this paper, every cylindrical magnet is divided into elementary cuboidal magnets. The accuracy can be controlled by regulating the value of elementary cuboidal permanent magnets "N ". The approximation can also be used to calculate the force interaction in the cylindrical linear single-axis-actuator. We confirm the validity of magnetic force calculation by comparing it with other methods and measurements. The calculation results are in very good agreement with measured values, which indicates the feasibility of our approximation.

Abstract-This paper presents a new model based on simulated annealing algorithm (ASA) and adaptive neuro-fuzzy inference system (ANFIS) for shape optimization and its applications to electromagnetic devices. The proposed model uses ANFIS system to evaluate the electromagnetic performance of the device. Both the ANFIS and ASA method are applied to the design/optimization of the electromagnetic actuator. The results of the proposed approach are compared with other techniques such as: method of moving asymptotes, penalty method, augmented lagrangian genetic algorithm and simulated annealing method (SA). Among the algorithms, the proposed ANFIS-ASA approach significantly outperforms the other methods.

No abstract