When the improved droop control based on virtual impedance is adopted in island microgrid, the problem of inaccurate distribution of reactive power and reactive power circulation will still occur with the changing line impedance due to the fixed value of virtual impedance. To solve this problem, a virtual impedance prediction model based on partial least squares regression (PLSR) is proposed, which uses the line impedance value and the system impedance value before compensation to predict the virtual impedance value and realizes the adaptive virtual impedance, thus overcoming the problem in the improved droop control based on virtual impedance. There is no need to detect the real-time power value and circulation value, and the use of communication network is not required. Furthermore, from a comparison with the prediction results obtained by neural network models, it is proved that the prediction accuracy of the virtual impedance prediction model based on PLSR is better. At last, a simulation system of microgrid is constructed in MATLAB/Simulink to verify the adaptive virtual impedance, and simulation results show the superiority of the proposed model.

To satisfy the low sampling frequency, low computational cost and high accuracy requirements of renewable energy generation systems in the grid voltage detection link, a high-precision discrete-time frequency-locked loop (FLL) which does not need to call trigonometric functions is proposed. First, the open-loop transfer function of discrete-time reduced-order generalized integrator (d-ROGI) is derived according to the expression of voltage based on complex numbers under the static coordinate system. Then, a d-ROGI with a low approximation error is derived according to the relationship between the unknown parameter of the open-loop transfer function and frequency. On this basis, the FLL for estimating the unknown parameter is constructed, the second-order small-signal model of the FLL is established, and the corresponding parameter tuning method is given. Finally, experimental results show that the FLL has a higher detection accuracy at a low sampling frequency than the most commonly used third-order numerical integrator discretization method. At the same time, it has a lower computational cost and requires less storage according to the comparison of computation cost.

For an LCL-type inverter connected to weak grid, the appearance of grid impedance often results in a decrease in the phase margin, serious distortion of grid-connected current and even system instability. To solve this problem, an improved grid-connected current control strategy is proposed, in which a multi-resonance controller is introduced in the voltage feedforward loop to suppress the voltage background harmonics and a phase compensator is added to the current feedforward loop to improve the system's phase margin, so as to avoid the risk that the resonance peak of the multi-resonance link intersects with the -180° line. Theoretical analysis and simulation results show that the proposed strategy can effectively suppress the harmonics of LCL-type grid-connected current, improve the current quality and enhance the stability of the grid-connected system.

Accurately predicting the remaining useful life (RUL) of lithium-ion batteries is of significance for improving the safety of working environment and the reliability of equipment. To improve the stability and accuracy of RUL prediction, a battery RUL prediction method based on the combination of denoising technology and hybrid data-driven model is proposed. First, the original data is decomposed by variational mode decomposition, and the noise components are filtered by the analysis of correlation. The residual error is combined with the components which have a strong correlation to complete the sequence reconstruction process. Second, with the combination of Tent chaotic mapping, sine cosine algorithm and Levy flight strategy, the sparrow search algorithm (SSA) is optimized, and the optimal weight threshold of extreme learning machine (ELM) is obtained. Finally, the improved SSA-ELM model is trained by using the smoothed denoised data, and the RUL prediction is completed. The NASA data sets are used to verify the effectiveness of the proposed method. Experimental results show that the average absolute error and root mean square error of the prediction result obtained using this method are controlled within 1.58% and 2.14%, respectively, indicating that this method has a high robustness and a high prediction accuracy. Therefore, the proposed method can be applied to battery RUL prediction.

At present, the physical parameters of a lithium battery cycle life model are difficult to obtain, and the parameter identification process needs a lot of experimental data and a long test time. In addition, it is difficult and expensive to simulate the cycling effect of lithium-ion batteries. On this basis, in order to explore the electrical stimulation of lithium-ion battery aging (due to cycling) and its effect on the battery capacity and internal resistance, a novel cycle life model of lithium-ion battery is proposed. First, a simple physical equation is established based on the fatigue theory and equivalent cycle counting. The parameter identification process is simple, requiring only a small amount of data in the battery data table and a limited (or short) cycle test. The proposed model is general and can represent the effects of common cycle life factors such as depth-of-discharge, temperature and C rate. Finally, two kinds of lithium-ion batteries (i.e., LFP-LiFePO₄ and NMC-LiNiMnCoO₂) are used to verify the model. The simulation results are close to the actual situation, and the error is within 1.5% compared with the experimental results.

A novel single-switch high-gain converter with no transformers and no coupled inductors is studied in this paper. Since the voltage lifting unit is added to the Boost converter, the voltage gain of the converter is improved, the voltage stresses of the switch and diodes are reduced, and the conduction loss of the switch is reduced under the condition of a small duty cycle. As a result, the efficiency of the converter is improved. To further improve the dynamic performance and anti-disturbance capability of the converter, the immune feedback mechanism is introduced based on the analysis of a single neuron controller. A fuzzy immune-single neuron PID control strategy is studied in this paper, in which the fuzzy immune control is combined with the single neuron smart controller to realize self-tuning of the single neuron proportional coefficient. Finally, a simulation study of the proposed converter and control strategy was carried out, and an prototype with an output of 200 V/0.5 A was designed for experimental verification. Both the simulation and experimental results show that the proposed converter can obtain a higher voltage gain under a smaller duty cycle. Compared with the traditional PID control strategy, the proposed fuzzy immune-single neuron PID control strategy can more effectively suppress system disturbances and improve the dynamic performance of the converter, indicating a stronger adaptive capability and a stronger robustness.

When the grid voltage is unbalanced, the phase-locked loop (PLL) needs to retrack the grid voltage. At this moment, an inaccurate phase lock or an overlong phase lock time will cause the degradation of control performance for the subsequent flexible DC transmission system. Therefore, to avoid the adverse effects caused by PLL, an improved control strategy for converter without PLL is proposed. First, the difference between the traditional PLL scheme and the proposed scheme without PLL is analyzed, and an instantaneous positive-and negative-sequence component extraction scheme is put forward in an environment without PLL. Second, a converter control system without PLL is designed, and the compensation of negative-sequence current is taken as its control target. Finally, the validity and correctness of the proposed strategy was verified by the experimental results of a prototype.

Aimed at the 5th, 7th, 11th, 13th and other low-order harmonics which account for a large proportion in a voltage source converter based high-voltage direct current (VSC-HVDC) AC system, on the basis of proportional integral (PI) control, a selective harmonic current control strategy based on a vector proportional integral (VPI) regulator in dq coordinate system is proposed, in which PI is used to control the DC component of current error while VPI is used to suppress the frequency doubling fluctuation of current error. Different from the proportional integral resonance (PIR) regulator, VPI contains a second-order numerator, which can achieve an ideal 0° phase delay of the closed-loop transfer function of the control system at the resonance frequency point. As a result, its control accuracy of harmonic current is better than that of PIR. A two-terminal VSC-HVDC system is established by using the Simulink software, and the AC current on two sides of VSC in three control modes of traditional PI, PIR and PI parallel VPI is simulated. Through a comparison of harmonic content, the superior harmonic suppression performance of VPI is verified.

Temperature sensitive electrical parameter method has characteristics such as strong online capacity, non-invasiveness, and rapid response, so it has become a research hotspot at present. The on-state voltage drop is taken as a temperature sensitive electrical parameter, and an online monitoring method for IGBT junction temperature is studied based on the on-state voltage drop. First, the data of on-state voltage drop, collector current, and junction temperature of IGBT is obtained through the double-pulse test circuit. Then, based on the measured data, a three-dimensional mapping representation model of IGBT collector current, junction temperature, and on-state voltage drop is constructed. Finally, a novel on-state voltage drop sampling circuit was designed, and an online monitoring experimental of IGBT junction temperature was conducted. Experimental results verified the accuracy and validity of the obtained three-dimensional junction temperature representation model.

In the actual operation of a battery, its temperature will vary with the ambient temperature, which undoubtedly increases the difficulty in estimating its state-of-charge (SOC). To address this problem, the relationship of temperature with the charge and discharge capacities, internal resistance and open circuit voltage of the battery is studied, and an equivalent circuit model considering temperature is established accordingly. The battery SOC is estimated based on this model by combining the extended Kalman filter algorithm, which can update the temperature-dependent variables in real time and adapt to the temperature change of the battery. In addition, this method is validated at variable temperatures. Results show that the proposed method can quickly and accurately estimate the battery SOC with an estimation error within 2%.