Wavelet Packet Decomposition-Based Multiscale CNN for Fault Diagnosis of Wind Turbine Gearbox.

This article presents an intelligent fault diagnosis method for wind turbine (WT) gearbox by using wavelet packet decomposition (WPD) and deep learning. Specifically, the vibration signals from the gearbox are decomposed using WPD and the decomposed signal components are fed into a hierarchical convolutional neural network (CNN) to extract multiscale features adaptively and classify faults effectively. The presented method combines the multiscale characteristic of WPD with the strong classification capacity of CNNs, and it does not need complex manual feature extraction steps as usually adopted in existing results. The presented CNN with multiple characteristic scales based on WPD (WPD-MSCNN) has three advantages: 1) the added WPD layer can legitimately process the nonstationary vibration data to obtain components at multiple characteristic scales adaptively, it takes full advantage of WPD and, thus, enables the CNN to extract multiscale features; 2) the WPD layer directly sends multiscale components to the hierarchical CNN to extract rich fault information effectively, and it avoids the loss of useful information due to hand-crafted feature extraction; and 3) even if the scale changes, the lengths of components remain the same, which shows that the proposed method is robust to scale uncertainties in the vibration signals. Experiments with vibration data from a production wind farm provided by a company using condition monitoring system (CMS) show that the presented WPD-MSCNN method is superior to traditional CNN and multiscale CNN (MSCNN) for fault diagnosis.

Training Deep Neural Network for Optimal Power Allocation in Islanded Microgrid Systems: A Distributed Learning-Based Approach.

Currently, numerical optimization methods are used to solve distributed optimal power allocation (OPA) problems for islanded microgrid (MG) systems. Most of them are developed based on rigorous mathematical derivation. However, the complexity of such optimization algorithms inevitably creates a gap between theoretical analysis and real-time implementation. In order to bridge such a gap, in this article we provide a new distributed learning-based framework to solve the real-time OPA problem. Specifically, inspired by the human-thinking scheme, distributed deep neural networks (DNNs) together with a dynamic average consensus algorithm are first employed to obtain an approximate OPA solution in a distributed manner. Then a distributed balance generation and demand algorithm is designed to fine-tune it to obtain the final optimal feasible solution. In addition, it is theoretically proved that the proposed DNN can well approximate one existing OPA algorithm (Guo et al. 2018), where quantitative numbers of at most how many hidden layers and neurons are provided. Several experimental case studies show that our proposed distributed learning framework can achieve similar optimal results to those obtained by using typical existing distributed numerical optimization methods while it is superior in terms of simplicity and real-time capability.

False data injection against state estimation in power systems with multiple cooperative attackers.

Given the strong cyber-physical interactions in today's smart grid, false data injection (FDI) attack can readily mislead the state estimation and influence the system operation by manipulating meter measurements. In this paper, a new FDI attack strategy is considered where multiple attackers cooperatively launch an unobservable attack. Firstly, the entire transmission system is partitioned into several subsystems, with each attacker only acquiring and manipulating the measurements in its local area. With limited communications among neighboring attackers, all of them can successfully modify the estimated states without being detected. In addition, by taking practical constraints into account, a least-effort attack problem is formulated and subsequently solved by a distributed alternating direction method of multipliers (ADMM)-based approach. Several case studies implemented on a 4-bus and IEEE 118-bus power systems have finally demonstrated the effectiveness of the proposed approach in the scenario of multiple attackers.

Analysis of plasma proteome from cases of the different traditional Chinese medicine syndromes in patients with chronic hepatitis B.

A proteomic analysis method, two dimensional gel electrophoresis (2-DE) followed by matrix-assisted laser desorption/ionization time-of-flight MS (MALDI-TOF-MS), was used to explore the link between plasma proteome and the different syndromes of traditional Chinese medicine (TCM) in patients with chronic hepatitis B (CHB). In compared with the plasma proteomes from health donors, the alterations in protein expression from cases of the five TCM syndromes, including damp heat stasis in the middle-Jiao syndrome, liver Qi stagnation and spleen deficiency syndrome, spleen and kidney Yang deficiency syndrome, liver and kidney Yin deficiency syndrome, and blood stasis into collateral syndrome with CHB were identified (P<0.05). In the cases of the five TCM syndromes with CHB, immunoglobulin J-chains (IGJ) and C-reactive protein (CRP) were up-regulated, while haptoglobin (HPT), retinol binding protein (RBP) and vitronectin were down-regulated. To further confirm these results, four proteins, including CRP, IGJ, HPT and RBP, from more plasma samples were quantified by ELISA. The results showed that the changes of protein levels were consistent with those from the 2-DE experiment. Importantly, the upregulation tendency of IGJ level in plasma is related with the different TCM syndromes with CHB (P<0.05). Our results show that IGJ may serve as a novel potential biomarker for diagnosis of the different TCM syndromes in patients with CHB.

[Plasma proteomic analysis of patients with chronic hepatitis B of damp-heat retention in the middle-jiao syndrome].

OBJECTIVE: To study the essence of chronic viral hepatitis B (CHB) of damp-heat retention in the middle-jiao syndrome (DRMS) from plasma proteomic angle. METHODS: Plasma proteomic analyses of plasma whole protein of patients in the group with CHB of DRMS (20 cases) and subjects in the health control group (5 cases) were compared using two-dimensional gel electrophoresis (2-DE), mass spectrography, and other bioinformatics analyses methods. RESULTS: Eight protein dots with obvious regularity changes of differential expression were obtained by 2-DE. Seven protein dots were obtained by mass spectrography (One protein dot with undetected results): apolipoprotein C2 (APO-C2), vitronectin (VN), haptoglobin (HPT), transthyretin (TTHY), APO-A1, serum amyloid P-component (SAMP), and APO-A4. Compared with the health control group, the expressions of APO-A1 and APO-A4 were somewhat higher and the expressions of the expressions of the rest five protein dots were obviously down-regulated. CONCLUSION: APO-Al and APO-A4 were of potential significance in the diagnosis of CHB patients of DRMS, prognostic markers, or treatment targets.