Electromagnetic devices containing permanent magnet (PM) are common and typical electromagnetic devices that have advantages of low-power consumption, small size, high sensitivity, and its optimization problem has aroused widespread concern. Calculation accuracy of electromagnetic properties and efficiency of optimization processes are two important factors that affect the optimization effect of electromagnetic devices. A fast calculation method was proposed in this paper that aimed at electromagnetic devices with PM characteristics of strong non-linearity and low solving precision. The response surface method was used to derive the basis function, and the error between actual measured values and calculated values was modified by the Kriging method. On the basis of the calculation model, the distributed collaborative strategy was adopted to fulfill the parallel execution of the fast computation. Simultaneously, the updating strategy of particles was improved, and the robustness of the multi-objective particle swarm optimization algorithm in solving the optimization problem of electromagnetic devices with PM was enhanced. Effectiveness of the method proposed in this paper was verified by the case study of a specific type of electromagnetic relay with PM.

The magnetic properties of soft magnetic materials, including their saturation magnetic induction strength and permeability, significantly affect the dynamic characteristics of electromagnetic relays. However, the soft materials most commonly used for relays in the magnetic conductive components of electromagnetic systems, such as electrical pure iron, limit further relay design improvement and optimization to a certain extent. Thus, this paper proposes the use of amorphous and nanocrystalline soft magnetic materials with good high-frequency magnetic properties in magnetic circuits. A wavelet analysis was conducted on the high-frequency components of the coil current while the relay operated, and the corresponding magnetic materials were selected. Considering the challenges in processing amorphous and nanocrystalline materials and collecting test data for the accuracy verification of simulation methods, we prepared a scaled-up prototype for use in dynamic characteristic tests. The simulation method was improved, yielding more accurate simulation results regarding the relay’s dynamic characteristics. On this basis, six replacement schemes using amorphous and nanocrystalline materials were considered. The test results proved that this application could improve the relay’s dynamic characteristics. Finally, a full-size sample with an iron core consisting of nanocrystalline alloy 1K107B was prepared, and the conclusions were verified in tests.

Fault diagnosis and bounce reduction of electromagnetic actuators require one to obtain the electromechanical transient characteristics (ETCs) in real-time. ETCs can be measured by sensors, but due to their large size and intrusiveness, the approach has its limitations. Acquiring ETCs through finite element method (FEM) could be an alternative, except that FEM is known to be time consuming. In contrast, real-time measurement of coil current is not subjected to these restrictions, yet, it has not been properly utilized for calculating ETCs. In this regard, this paper analyzes the source of the armature kinetic energy, and thus proposes a model based on energy conservation (ECM), describing the relationship between the mechanical and electrical characteristics. Based on the model, ETCs can be calculated from coil current, moving part mass, and the static counterforce. To exemplify the effectiveness of the procedure, an electromagnetic contactor is considered, and the results obtained using the ECM are found fairly consistent with that obtained using sensors. The merit of the proposed ECM lies in its non-intrusive nature and its ability to circumvent the time-consuming FEM in calculating ETCs.

Various intelligent algorithms are applied in optimization design, and the differential evolution (DE) algorithm is widely applied with its excellent convergence speed and convergence precision. This study analyzed the advantages and disadvantages of the existing multi-objective differential evolution (MODE) algorithm, and developed a MODE algorithm based on the adaptive weight and the multi-population strategy (MODE/AWMS). The proposed algorithm was verified using test functions. MODE/AWMS exhibited certain advantages compared with several other multi-objective optimization algorithms. Taking a polarized magnetic relay as an example, MODE/AWMS was used to optimize its key parameters by establishing a rapid calculation model of its electromagnetic mechanism. The electromagnetic force (EMF) of the release position was improved, which verified the validity of MODE/AWMS.