Search

Short-term power load has strong randomness and volatility, in order to improve the accuracy of load forecasting, this paper proposes a combined forecasting model based on quadratic decomposition, convolutional neural (CNN) network and long short-term memory (LSTM) neural network. Firstly, the original load series was decomposed into several intrinsic mode components and residuals by complete ensemble empirical mode decomposition with adaptive noise (CEEMDAN). Then, the sample entropy and K-means (SE-K-means, SK) were introduced to reconstruct the sub-sequences obtained by decomposition into three sequences, and the strong non-stationary sequences in the reconstructed components were decomposed twice by using variational mode decomposition. A CNN-LSTM model was established to predict each sub-sequence obtained by decomposition. Finally, the forecasting results were superimposed to achieve effective load forecasting. By using the actual load data for verification, it can be seen that from the perspective of four evaluation indicators: R2, mean absolute error, root mean square error and mean absolute percentage error, the proposed model has higher fitting and prediction accuracy than XGBoost, LSTM, CEEMDAN-LSTM and CEEMDAN-CNN-LSTM models.

Based on a single-phase photovoltaic grid-connected inverter, a control strategy combining traditional proportional–integral–derivative (PID) control and a dynamic optimal control algorithm with a fuzzy neural network was proposed to improve the dynamic characteristics of grid-connected inverter systems effectively. A fuzzy inference rule was established after analyzing the proportional, integral, and differential coefficients of the PID controller. A fuzzy neural network was applied to adjust the parameters of the PID controller automatically. Accordingly, the proposed dynamic optimization algorithm was deduced in theory. The simulation and experimental results showed that the method was effective in making the system more robust to external disruption owing to its excellent steady-state adaptivity and self-learning ability. [ABSTRACT FROM AUTHOR]

This paper is focused on the H control problem for linear systems with interval time-varying delays. By employing a reciprocally convex combination approach and a delay decomposition approach, some new delay-dependent bounded real lemmas (BRLs) are derived such that the closed-loop system is asymptotically stable with a prescribed H level. The BRLs are then used to solve the H controller design by incorporating with the cone complementary approach. Three numerical examples are finally given to show the validity of the proposed method. [ABSTRACT FROM AUTHOR]