In the background of a high proportion of new energy source connected to grid，the phenomenon of new energy interface inverters being connected to the grid through long-distance AC lines is becoming increasingly significant. The traditional vector controlled weak grid connected voltage source converter （VSC） based on phase locked loop （PLL） synchronization is prone to PLL synchronization instability. To improve the PLL synchronization stability of weakly connected VSC，an analytical model to reveal the PLL synchronization instability mechanism of weakly connected VSC was established. Through theoretical analysis，the influence of the dynamic coupling characteristics of PLL and different time scale control links on the PLL synchronization instability mechanism was revealed. Based on the theoretical analysis results，a PLL dynamic compensation control strategy suitable for multiple time scales was designed to improve the PLL synchronization stability of weakly connected VSC. The simulation results based on Matlab/Simulink verify the correctness of the theoretical analysis and the effectiveness of the compensation control strategy.

In order to further improve the output performance and efficiency of the converter，a grid-connected converter based on heterogeneous device mixing was proposed，referred to as heterogeneous grid-connected converter（HGCC）. The HGCC consists of two half-bridge modules based on SiC MOSFET devices，which are cross-connected by the commutation bridge arm based on Si IGBT devices. Furthermore，the HGCC modulation principle was given. The SiC MOSFET device works in the high-frequency switching state，while the Si IGBT device works in the low-frequency switching state，giving full play to the advantages of low switching loss of SiC devices and low on-state loss of Si devices. Then，the working mode of HGCC was analyzed in detail，and the control block diagram of HGCC and the internal voltage balancing strategy of capacitor were given. Finally，the effectiveness and feasibility of the proposed topology and control strategy was verified by simulation.

In recent years，the use of microinverters has received a lot of attention as one of the viable methods for residential and commercial PV power conversion systems. Flyback inverters operating in current continuous conduction mode （CCM） have been widely studied for their low output ripple current，high efficiency，and low cost. However，its control duty cycle to output current transfer function has right-half-plane （RHP） zeros，which may cause system instability when the inverter is combined with a high-gain feedback controller. In addition，flyback CCM inverters are subject to grid time-varying voltage disturbances. As a result，the conventional control scheme leads to inaccurate output tracking. An interleaved flyback CCM microinverter iterative learning control （ILC）method was proposed to help the system output converge to the reference trajectory by utilizing the prediction term and the current learning term. Compared with the traditional PI control scheme，the system output globally converges to the reference trajectory without state disturbance，output noise and initial state error.

The common-mode equivalent circuit model of the motor is the basis for analyzing the electric erosion of bearings，especially for analyzing the technical problems and experimental phenomena of shaft voltage and shaft current，among which the centralized equivalent model is an important calculation basis for the key index of bearing electric erosion that is bearing voltage ratio （BVR），which can be evaluated in the early stage of motor design. The accurate simulation calculation of the distributed capacitance parameters of the centralized equivalent model were studied in depth. Firstly，the simulation calculation and comparison of the two-dimensional single-slot and full-model motors were carried out. Secondly，the influence of the stator winding end and the rotor end ring on the stray capacitance was considered，and the three-dimensional motor model was simulated and calculated，and the more accurate distributed capacitance parameters were obtained through comparative study. Finally，the parameters of YQ190-14 type traction motor were taken as an example，and the simulation calculation results and test results were compared and analyzed，and the accuracy of the simulation calculation results was verified.

The electricity power price setting mechanism will be fundamentally reformed with the continuous advancement of a new round of electricity market-oriented reform. Presently，some provincial and regional power grid enterprises are facing the problem of large fluctuations in electricity prices，which is not conducive to the in-depth promotion of electricity marketization reform. Firstly，based on the partial least squares method，the agency power purchase price formation model was established. Then，the main factors affecting the fluctuation of power purchase price were analyzed，and the mechanism of the influence of power supply and consumption balance on the price change of different types of electricity consumption was clarified. In order to reduce the fluctuation of electricity purchase price，a multi-objective optimization model of agent electricity purchase was established with the constraint of power supply and consumption balance. The actual example results show that the model can realize the reasonable distribution of power consumption among different types of power sources in the process of agent power purchase，and effectively reduce the fluctuation of agent power purchase price.

Voltage source converter （VSC） is prone to low-frequency oscillations of 10⁰~10¹ Hz when connected to weak AC power grids. The mechanism of improving the control strategy of VSC on the traditional inverter side to suppress low-frequency oscillations（LFO） in the AC power grid is to transfer the oscillations from the AC side to the DC side，which poses a risk of inducing instability in the DC system. An adaptive voltage power damping control strategy was designed that can suppress low-frequency oscillations in AC systems based on the adjustable output power of the interface converter of energy storage devices containing new energy storage stations. This strategy was based on a damping control loop embedded in the rectifier side VSC control system，which suppresses DC voltage oscillation by controlling the output power of the rectifier energy storage system，effectively solving the shortcomings of traditional inverter side VSC control strategies in transferring oscillations. Based on the analysis of small signal stability，the parameter design process of the virtual damping control link was provided. The simulation results in Matlab/Simulink environment verified the effectiveness of the proposed control strategy and the correctness of the parameter design process.

In order to improve the compensation effect of LCL-type shunt active power filter，a frequency adaptive decimal delay repetitive control based on IIR filter was proposed. The resonance suppression loop of LCL was designed through root locus and the design result was used as a new control object for repetitive control. The IIR filter used could approximate the decimal delay caused by frequency changes，thereby matching the resonant frequency of repetitive control with the fundamental and harmonic frequencies of the power grid. The adaptive implementation diagram，the phase characteristics of the approximation part，and the design method of the compensator in sequence were presented. Finally，the stabilities of the system were analyzed. Simulation and experimental verification show that the proposed method can improve the steady-state tracking performance and THD of the system.

Since the compressed air energy storage system grid-connected inverter voltage is always influenced by voltage fluctuations of the power grid with little problem，for the sake of ameliorating the dynamic and stable performance of the DC bus voltage of the inverter on the side of the grid and the actual output power of the system in the compressed air energy storage expansion power generation system and reducing the fluctuations，second-order linear active disturbance rejection control （LADRC） technique was used in the voltage outer loop control，a new-style double closed loop control construction was formed. The mathematical model of the grid-connected converter of expansion power generation system was established，the operating principle of LADRC was researched，and the conventional PI control was contrasted with the control methods studied. The simulation results indicate that when make a contrast with the conventional control methods，the control methods designed can make the DC voltage enter the stable state more quickly，the total harmonic distortion （THD） of the grid-connected current becomes smaller，and has better disturbance immunity.

In order to solve the problem that the time-of-use price is not taken into account in the calculation of the peaking cost in the case of a large number of renewable energy sources，a method of calculating the peaking cost of power system with time-of-use price was proposed. First，the key scenarios of power grid peak regulation were obtained，and the technical cost characteristic boundaries of various peak regulation resources were determined. Second，according to typical daily new energy and load characteristics，taking the minimum total peaking cost of the power system as the objective function，taking into account the technical output of thermal power units，charging and discharging characteristics of energy storage，capacity of pumped storage power station and other constraints，and taking into account the time-of-sale price factor，a quantitative model of the peaking cost of the power system was established. Finally，through solving the model，the peak load balancing cost and unit output under the optimal scheme were obtained. The actual data of Ningxia Power Grid，the first provincial power grid whose output of new energy generation exceeds the entire regional power load，was taken as an example to verify the model. The results show that the model can effectively evaluate the peaking cost and unit output in the key scenarios of power grid peak regulation.

To address the complexities of transformer condition assessment indicators and the overreliance on expert judgment in the evaluation process，a novel method for fuzzy comprehensive assessment of transformer condition was introduced，which is based on gaming theory and an improved version of the D-S evidence theory. First，the analytic hierarchy process （AHP） and the Critic method was adopted to determine both the subjective and objective weights of the evaluation indicators. These weights are then combined using game theory，reducing the dependence on expert opinions in traditional weight determination methods. Second，instead of traditional membership functions，cloud models was employed to preserve the inherent uncertainty in fuzzy assessments. Finally，membership information at the project level was fused by using improved D-S evidence，eliminating paradoxical results that can occur when combining high-conflict evidence. Through the results of case studies，the method effectively assesses the transition probabilities of various transformer states.