9 December 2021 in IEEE Transactions on Power Electronics
Fast and accurate optimization of the dynamic response is very important for an LCC resonant converter. The trajectory control is an effective way to achieve these goals. However, the model of the LCC resonant converter is generally too complex to get the target steady-state operation frequency and trajectory online and in real time with the common controllers. In this article, a fast solution can be obtained by adopting the proposed two-point solution method. A simplified steady-state trajectory model is also established, which is the solution object of the two-point solution method. When the necessary parameters are acquired, the target state trajectory and operation parameters, e.g., frequency, can be obtained quickly and accurately. Furthermore, the fast and accurate control strategy based on the simplified state trajectory and the two-point solution method is proposed to realize control of the LCC converter. The experiments are performed to verify the proposed control strategy. During the load transient, compared with the conventional proportional–integral control, the voltage variation and the response time are improved by more than 80%, respectively, and the two-point solution method can be finished within 2.3 μs by the controller. The fast response with a very small voltage variation is realized by the proposed control strategy.
25 January 2021 in IEEE Access
Extreme learning machine (ELM) has become popular in recent years, due to its robust approximation capacity and fast learning speed. It is common to add a $\ell _{2}$ penalty term in basic ELM to avoid over-fitting. However, in $\ell _{2}$ -regularized extreme learning machine ( $\ell _{2}$ -RELM), choosing a suitable regularization factor is random and time consuming. In order to select a satisfactory regularization factor automatically, we proposed an adaptive regularized extreme learning machine (A-RELM) by replacing the regularization factor with a function. The function is defined in terms of the output weights named regularization function. And an iterative algorithm is proposed for obtaining the output weights, therefore, allowing for deriving their values simultaneously. Besides, the constructed regularization function ensures the convexity of the model, which contributes to a globally optimal solution. The convergence analysis of the iterative algorithm guarantees the effectiveness of the model training. Experimental results on some UCI benchmarks and the Yale face database B indicate the superiority of our proposed algorithm.
9 September 2022 in International Journal of Machine Learning and Cybernetics
3 February 2024 in Applied Soft Computing
10 April 2023 in Neural Processing Letters
21 February 2025 in Knowledge-Based Systems