Search

To quickly respond to variations in the state of network load demand, a solution using data-driven techniques to predict optimal power flow (OPF) has emerged in recent years. However, most of the existing methods are highly dependent on large data volumes. This limits their application on the newly established or expanded systems. In this regard, this work proposes a sample-efficient OPF learning method to maximize the utilization of limited samples. By decomposing the OPF task before knowledge distillation, deep learning complexity is reduced. Thereafter, knowledge distillation is used to integrate decoupled tasks and improve accuracy in low-data setups. Unsupervised pre-training is introduced to alleviate the demand for labeled data. Additionally, the focal loss function and teacher annealing strategy are adopted to achieve higher accuracy without extra samples. Numerical tests on different systems corroborate the advanced accuracy and training speed over other training methods, especially on fewer-sample occasions.

Different types of outliers have existed in the monitoring data of wind turbines, which are not conducive to the follow-up data mining. However, the complex inner characteristics of the monitoring data pose major challenges to detect the outliers. To address this problem, an unsupervised outlier detection approach combining stacked denoising autoencoder (SDAE) and density-grid-based clustering method is proposed. First, the characteristics of the outliers in supervisory control and data acquisition data caused by different reasons are analyzed. Then, the SDAE is utilized to extract features by training the original data. Furthermore, the density-grid-based clustering method is applied to achieve the clustering results. Window width is added to classify the outliers as isolated outliers, missing data, and fault data according to the duration of abnormal data. The monitoring data of four wind turbines are sampled as the training data to demonstrate the effectiveness of the proposed method. The results show that the proposed model can effectively identify the isolated outliers, missing data, and fault information in the high dimensional data set by unsupervised learning.

Compressive sensing is a desirable technique to acquire and reconstruct signals at sub-Nyquist rates. Recently, several deep learning-based studies on solving the compressive sensing problem have been carried out, which dramatically reduce the intensive computational complexity of the traditional greedy or convex recovery algorithms and even improve the signal recovery performance. However, as the signal size increases, most of these methods recover signals block by block due to the large computational complexity and memory consumption, which usually imposes block effect on the recovered signals. To deal with this issue, in this paper, we apply a tensor decomposition method named Tensor-Train (TT) on the neural network, by which the number of parameters is reduced by a tremendous factor and the computational complexity is further decreased so that the large signals can be recovered as a whole. In particular, the TT-decomposition is jointly applied on a stacked denoising autoencoder (SDA) network called TT-SDA in this paper. The experiments indicate that the proposed TT-SDA network can preserve the reconstruction performance of the conventional SDA network and outperform the traditional methods, especially with low measurement rates. Meanwhile, it can also significantly reduce the computational complexity and occupied memory space, which becomes a time and memory efficient method in compressive sensing problem.

This paper presents an automatic wearable electrocardiogram (ECG) classification and monitoring system with stacked denoising autoencoder (SDAE). We use a wearable device with wireless sensors to obtain the ECG data, and send these ECG data to a computer with Bluetooth 4.2. Then, these ECG data are classified by the automatic cardiac arrhythmia classification system. First, the ECG feature representation is learned by the SDAE with sparsity constraint. Then, the softmax regression is used to classify the ECG beats. In the fine-tuning phase, an active learning is added to improve the performance. In the active learning phase, we use the method that relies on the deep neural networks posterior probabilities to associate confidence measures to select the most informative samples. Breaking-ties and modified breaking-ties methods are used to select the most informative samples. We validate the proposed method on the well-known MIT-BIH arrhythmia database and ECG data obtained from the wearable device. We follow the recommendations of the Association for the Advancement of Medical Instrumentation for class labeling and results presentation. The results show that the classification performance of our proposed approach outperforms the most of the state-of-the-art methods.

Accurate transient stability assessment (TSA) is a fundamental requirement for ensuring secure and stable operation of power systems. Tremendous efforts have been made to apply artificial intelligence approaches for TSA with phasor measurement unit data. However, many previous approaches may be failed to provide favorable accuracy due to the shallow architectures and error-prone hand-crafting features. This paper proposed a model for TSA, which is termed multi-branch stacked denoising autoencoder (MSDAE). This model is a unified framework integrating multiple stacked denoising autoencoders (SDAEs), one fusion layer, and one logistic regression (LR) layer. Initially, the SDAEs at the bottom of MSDAE extract features from multiple kinds of measurements respectively. Then, the extracted features are encoded into unified fusion features by the fusion layer. Finally, the LR layer performs TSA by using the fusion features. The depth of the architecture contributes to the remarkable ability for feature learning, while the width of the architecture (i.e., the multiple branches) enables MSDAE to deal with different kinds of measurements by a reasonable mechanism. In this way, MSDAE achieves feature extraction and classification intrinsically and simultaneously, namely, achieves TSA in an end-to-end manner. The results of experiments on IEEE 50-machine system demonstrate the superiority of the proposed model over the prior methods.