The constant-current power supply system is suitable for remote seabed power supply in a harsh environment owing to its strong anti-failure capability. As all the seabed equipment adopts constant-voltage power supply, a constant-current to constant-voltage conversion device is needed to convert constant-current input into constant-voltage output to provide electric energy for the seabed equipment. To solve the problem that an efficient conversion from constant-current to constant-voltage in a wide load range as well as a high-pressure isolation control, a constant-current to constant-voltage converter topology with cascade of a shunt regulator circuit and a DC transformer is proposed to achieve the efficient conversion from constant-current to constant-voltage in a wide range. Aimed at the output control problem under high-pressure isolation, an indirect control strategy for output voltage based on input-side detection is studied to achieve an accurate control of output voltage without the need of high-cost and large-volume output isolation detection devices. Finally, an experimental prototype with input of 1 A and rated power of 500 W was built to verify the feasibility of the power conversion technology of constant-current to constant-voltage converter.

The magnetic integrated coupling technology is introduced into a high-gain converter, and a magnetic integrated Boost converter based on diode clamping is proposed. Through the theoretical analysis of the working principle and performance characteristics of the proposed converter, it is shown that this converter can reduce its volume and inductance current ripple based on the advantages of the original converter, such as a high voltage gain and a simple control strategy. In addition, the voltage gain of the novel converter is further improved by using the switched capacitor technology, the voltage stress of the switch and diode is further reduced, and the energy conversion efficiency of the converter is effectively improved. Finally, the findings were verified by simulation and experimental results.

The traditional fault diagnosis methods for power transformers cannot detect the power faults accurately or ensure their normal operation. Therefore, a fault diagnosis method for power transformers based on wavelet packet transform and support vector machine (SVM) is proposed. For the power signal collected from a power transformer, the improved minimum noise fraction (MNF) transform denoising is used to denoise, and the noise matrix is estimated by the weighted neighborhood mean method. After the estimation, the improved MNF transform is used to effectively realize image dimensionality reduction and denoising, extract the signal characteristics, and divide the signal into low- and high-frequency part by means of wavelet packet transform to obtain the wavelet packet energy feature vector. The obtained wavelet packet energy feature vector is input into an SVM classifier, and the output results from the SVM classifier are used to realize the state recognition and fault diagnosis of power transformer. Experimental results show that the proposed method can effectively diagnose the faults in the power transformer, such as iron core short-circuit, coil interlayer short-circuit, bushing-to-ground breakdown, coil insulation resistance drop and bushing-to-bushing discharge, and the fault diagnosis accuracy was higher than 98.5%.

With the large-scale integration of wind power and other renewable energy sources, the frequency regulation capacity and effect of traditional frequency regulation power sources are difficult to meet the requirements of power grid. To solve this problem, a comprehensive control strategy based on the frequency regulation signal optimization of a battery energy storage system which assists the thermal power unit to participate in secondary frequency regulation is proposed. First, a simulation model of energy storage that meets the power grid’s frequency regulation requirements is established. Based on this model, the allocation mode of area control error signal and area frequency regulation requirement signal is analyzed in the complex frequency domain, and the switching criterion for frequency regulation signal is determined by combining the advantages of the two control signals. Then, considering the economy and efficiency of energy storage frequency regulation, the allocation coefficient is optimized by a decomposed multi-objective evolutionary algorithm to reduce the frequency offset and optimize the cost of frequency regulation. Finally, the effectiveness of frequency regulation signal switching criterion and multi-objective evolutionary algorithm in optimizing the energy storage allocation coefficient is verified by step disturbance simulation. The comprehensive control strategy is verified by continuous disturbance simulation, and results show that it can not only reduce the system frequency offset effectively, but also lower the operating cost of energy storage.

The bi-directional power transmission with a high efficiency and a high power density can be achieved by employing CLLC resonant converters. However, the traditional parameter design method is cumbersome and requires multiple iterations to obtain appropriate circuit parameters. To solve this problem, the working principle and characteristics of a bi-directional CLLC converter are analyzed, and a novel parameter design method is proposed. By considering the full range of soft switching, design index constraints and high-efficiency optimization conditions, the range of design parameters is narrowed and the design steps are optimized, thus effectively reducing the complexity of the converter parameter design process. Based on the demand for a 48~380 V/kW bi-directional DC-DC converter in industrial applications, specific parameter design steps and results were given, and a prototype was developed. The correctness and effectiveness of the proposed parameter design method was verified through experimental testing.

With the development of wide band gap devices, SiC MOSFET has been widely applied, and the research on its short-circuit protection has become an important topic to ensure the reliability of power electronic equipment. In view of the short short-circuit withstand time of SiC MOSFET and the difficulty in short-circuit fault protection, a short-circuit detection method for SiC MOSFET based on a planar differential Rogowski coil is proposed, which realizes a rapid identification of short-circuit fault by measuring the drain source current of the circuit and has advantages such as a fast response speed, a strong anti-interference capability and complete isolation from the main circuit. First, the working process of the SiC MOSFET short-circuit detection method based on the planar Rogowski coil is introduced. The partial element equivalent circuit (PEEC) modeling method for planar Rogowski coil is studied in detail, and an equivalent model which can reflect the coil’s high-frequency characteristics is obtained. At the same time, the influence of the geometric structure of the planar Rogowski coil on its performance is analyzed, and an optimal design scheme considering both the high gain and high bandwidth is proposed. Aimed at the problem of low measurement accuracy of the Rogowski coil in an environment with strong electromagnetic interference, a scheme of using the differential coil is put forward to improve the anti-interference performance. Finally, the anti-interference per-formance of the designed planar differential Rogowski coil and the reliability of short-circuit protection method based on this coil were verified by experimental results.

The power supply load of a high voltage direct current (HVDC) power supply system is prone to be affected by signal interruption, which will result in the unstable operation of power supply system. To effectively improve the management efficiency of power supply system, an optimization method for the management efficiency of HVDC power supply system based on genetic algorithm is proposed. The optimization of management efficiency is transformed into the problem of load distribution, and a load distribution control strategy of efficiency optimization is adopted to reasonably distribute the load current in the HVDC power supply system. The genetic algorithm is used to optimize the management efficiency of the HVDC power supply system, so as to optimize its management efficiency. Experimental results show that the system management efficiency can be adjusted through the independent operation of dual power supplies, thereby improving the efficiency of power supply system. In addition, the running time of this method was maintained at 90-120 s, the optimization time was short, and the efficiency was high.

To avoid the dynamic problems caused by dynamic and static loads in the operation of a power electronic power system, which affect the system stability, a modeling and analysis method for the voltage power angle dynamic stability is proposed. The continuation method is used to track the equilibrium solution manifold of the power system, and the small disturbance analysis method is used to calculate the voltage power angle state matrix to judge the dynamic stability of voltage power angle. By means of the power system stability mode discrimination method, the correlation between voltage and power angle state variables in the instability state is analyzed, the system instability category is determined, and the construction and analysis of voltage and power angle dynamic stability model is realized. Experimental results show that the load model is directly related to the category of dynamic instability of power system in an electronic environment. The static load model is easy to cause power angle instability, while the dynamic load model is easy to cause voltage instability. The time-domain simulation results are consistent with the discrimination results of state variable participation factor. Through the tests of power angle instability and voltage instability, the generator angle and node voltage are analyzed. It is proved that the proposed method has a good effect in the analysis of voltage and power angle dynamic stability, providing certain reference value for researches in the field of power system engineering.

Harmonic and electromagnetic interference (EMI) filters are two important output filters used to sup-press the harmonic distortion and EMI noise in grid-connected inverters. Harmonic and EMI filters are combined by planar magnetic integration to reduce the volume and weight. Through the selection of an appropriate magnetic core, the common mode and differential mode inductors are integrated into the same core by drawing PCB planar coil. To integrate the discrete capacitors and further realize the planar magnetic integration of EMI filter, the dielectric is inserted into the PCB and the layer connection mode is reasonably planned. A symmetric LCL filter is used to replace the traditional asymmetric structure of magnetic integration. Furthermore, by designing the air gap in the center pillar of the magnetic core and reasonably arranging the planar windings, the inductors of LCL harmonic filter are also integrated into the same magnetic core unit to form an LCL-EMI planar magnetic integrated filter. A gallium nitride single-phase inverter platform was built, and the LCL-EMI filter with planar magnetic integration was experimentally analyzed to verify the feasibility of the planar magnetic integration method.

To solve the problems of constant-voltage output instability and low efficiency caused by load resistance and coupling coefficient fluctuations in the dynamic charging process of an electric vehicle, a novel dual-side control scheme is proposed. In this scheme, the constant-voltage control output is realized by adjusting the pulse width angle θ of a high-frequency inverter at the transmitter, and the maximum efficiency tracking (MET) control of the system is realized by adjusting the pulse width angle φ of a controllable rectifier at the receiver. Through theoretical analysis, it is proved that when the derivative ∂[sin(θ/2)]/∂[sin(φ/2)] is a specific constant, the system can always work under the operating condition of maximum efficiency. Compared with the same type of MET control scheme, the proposed scheme does not need to install expensive current or power sensors on the transmitter, which reduces the system’s development cost to a certain extent. To verify the rationality of the proposed scheme, an experimental verification device with a rated power of 360 W was built, and experimental results fully proved the rationality and effectiveness of this scheme.