Parameter selection for kernel functions is important to the robust classification performance of a support vector machine (SVM). This paper introduces a parameter selection method for kernel functions in SVM. The proposed method tries to estimate the class separability by cosine similarity in the kernel space. The optimal parameter is defined as the one that can maximize the between-class separability and minimize the within-class separability. The experiments for several kernel functions are conducted on eight benchmark datasets. The results demonstrate that our method is much faster than grid search with comparable classification accuracy. We also found that the proposed method is an extension of a reported method in reference [2].

To overcome the performance degradation in the presence of steering vector mismatches, strict restrictions on the number of available snapshots, and numerous interferences, a novel beamforming approach based on nonlinear least-square support vector regression machine (LS-SVR) is derived in this paper. In this approach, the conventional linearly constrained minimum variance cost function used by minimum variance distortionless response (MVDR) beamformer is replaced by a squared-loss function to increase robustness in complex scenarios and provide additional control over the sidelobe level. Gaussian kernels are also used to obtain better generalization capacity. This novel approach has two highlights, one is a recursive regression procedure to estimate the weight vectors on real-time, the other is a sparse model with novelty criterion to reduce the final size of the beamformer. The analysis and simulation tests show that the proposed approach offers better noise suppression capability and achieve near optimal signal-to-interference-and-noise ratio (SINR) with a low computational burden, as compared to other recently proposed robust beamforming techniques.

Transformerless inverter for grid‐tied photovoltaic (PV) system has been widely used due to lower cost, higher efficiency and lighter weight. Various transformerless inverter topologies have been proposed to meet the safety requirement of low leakage current and obtain the reactive power capability. To get better performance, a novel transformerless hybrid‐H6 inverter with an improved modulation technique is proposed in this study. By adopting the improved modulation technique, two symmetry paths are realised to share the current during the freewheeling mode. Thus, without paralleling any additional capacitors to the switch, the inverter can reduce the influence of junction capacitance on common mode voltage naturally which results in mitigating the leakage current issue. Thanks to the dead time reduction through the improved modulation, the qualities of output waveforms are improved. Moreover, reactive power control is achieved without any modification of the inverter structure. Finally, a 1 kW prototype is simulated and tested to verify the theoretical analysis of this study. Not only the reactive power capability is obtained for the proposed inverter, but also the small common mode voltage fluctuation is achieved at the same time. In addition, the total harmonic distortion of the current is decreased by more than 1.7%.

Due to its low conduction loss, hence high current ratings, as well as low cost, Silicon Insulated Gate Bipolar Transistor (Si IGBT) is widely used in high power applications. However, its switching frequency is generally low because of relatively large switching losses. Silicon carbide Metal-Oxide-Semiconductor Field-Effect Transistor (SiC MOSFET) is much more superior due to their fast switching speed, which is determined by the internal parasitic capacitance instead of the stored charges, like the IGBT. By the combination of SiC MOSFET and Si IGBT, this paper presents a novel series hybrid switching method to achieve IGBT’s dynamic switching loss reduction by switching under Zero Voltage Hard Current (ZVHC) turn-on and Zero Current Hard Voltage (ZCHV) turn-off conditions. Both simulation and experimental results of IGBT are carried out, which shows that the soft switching of IGBT has been achieved both in turn-on and turn-off period. Thus 90% turn-on loss and 57% turn-off loss are reduced. Two different IGBTs’ test results are also provided to study the modulation parameter’s effect on the turn-off switching loss. Furthermore, with the consideration of voltage and current transient states, a new soft switching classification is proposed. At last, another improved modulation and Highly Efficient and Reliable Inverter Concept (HERIC) inverter are given to validate the effectiveness of the device level hybrid soft switching method application.