Lele Yao

In this paper, an uncertainty and disturbance estimator (UDE) method is proposed for dual-buck grid-connected inverter to improve the robustness of deadbeat predictive current control with respect to the variation of model parameters. Although deadbeat control has fast dynamic response, it has the delay problem in digital control and relies on accurate model parameters. Therefore, a two-step prediction method is adopted to compensate the digital control delay. Then, the prediction current error caused by parameters mismatch is analyzed. Based on the idea of regarding the variation of model parameters as an uncertain disturbance to the PV system, the deadbeat predictive control strategy based on UDE is proposed to reduce the sensitivity to parameters. Comparative simulation results verify that the proposed control method can effectively improve the tracking ability of the grid-connected current to the reference value when the system parameters change.

When the full-bridge converter works under the light load condition, the power efficiency obtained by the theoretical model is much different from that of the actual converter. Facing with this situation, an improved power loss model based on the typical power loss model is proposed. In this paper, the typical power loss model is called typical model for short and the improved power loss model is called proposed model for short. Firstly, the antiparallel freewheeling diodes at the arms of full-bridge circuit are taken into account. Every barrier junction capacitance of Schottky diode in the rectifier circuit is neglected. Then, the turning-off loss of full-bridge and the core loss of inductive components (the transformer and the filter inductor) in the typical model are compensated and modified by combining the theoretical values with the measured input current under the minimum and the maximum output current. In addition, it also corrects the equivalent resistance related to the conduction loss of converter. Eventually, the proposed model is established. The rise time and fall time of the midpoint voltage of two arms, and the fluctuation degree of the reverse bias voltage related to the Schottky rectifier diodes are regarded as the local indexes. The conduction time of the metal oxide semiconductor field effect transistor (MOSFET) in each switching period, and the power efficiency of converter are regarded as the global indexes. Based on the analyses, the local indexes are compared qualitatively, while the global indexes are compared quantitatively. It is found that the differences of the local indexes between the proposed model and the experiment are smaller than those between the typical model and the experiment. Meanwhile, the global indexes of the proposed model are closer to the experimental results. Therefore, it can be further demonstrated that the proposed model is more approximate to the actual converter than the typical model.

For the full-bridge LLC converter, it is necessary to establish a large signal model with low-complexity and high-precision when the current of filter inductor works under the continuous conduction mode (CCM) situation. In terms of low complexity, peak values of resonant state variables and average values of slow state variables are taken as the measured indexes between models and actual converter. Models based on the first harmonic approximation (FHA) are studied in depth because they have lower complexity than the existing models. From prospective of high precision, deficiencies of typical FHA model are explained in the light of harmonic generation mechanism of primary current and influence of Fourier series on the typical equivalent circuit. Therefore, transient process of each metallic oxide semiconductor field effect transistor (MOSFET) and its related fast recovery diode, transient process of Schottky rectifier diodes, piecewise equivalent circuit, appropriate selection of variables, and simplified modified dynamic equations are all considered. Furthermore, the unified proposed model is achieved according to the equivalent principle of indexes between aforementioned improved analyses and proposed model. Numerical results of typical FHA model and proposed model are given in accordance with the key indexes. Then, corresponding experimental results are also presented. Differences of indexes between the two models and the actual converter are compared respectively. Though complexity of proposed model is the same as that of typical FHA model, precision of proposed model is higher than that of typical FHA model.

At present, in terms of the influences of temperature and holidays on the power load, an improved T-S fuzzy-neural network is proposed to forecast the short-term power load. Considering the learning rate and smoothing factor should be adjusted dynamically for the higher performance of T-S fuzzy-neural network, the two optimal parameters must be found automatically when the parameters are changed. Thus a T-S fuzzy-neural network based on the Particle Swarm Optimization(PSO) to adjust parameters dynamically is proposed. Firstly, the stable intervals of the parameters must be found, so that parameters can be adjusted dynamically in the intervals. Secondly, the PSO is introduced to find the optimal parameters, therefore the system can converge to a stable state rapidly, and achieve an optimal control precision. Finally, combining with data of power load, temperature and holidays provided by the UNITE(the European Network of Excellence on Intelligent Technologies for Smart Adaptive Systems)web, and the MATLAB simulation results analysis, the improved T-S network has been proved a higher control precision in power load forecasting.