No abstract

In this paper, a hybrid model that considers both accuracy and efficiency is proposed to predict photovoltaic (PV) power generation. To achieve this, improved forward feature selection is applied to obtain the optimal feature set, which aims to remove redundant information and obtain related features, resulting in a significant improvement in forecasting accuracy and efficiency. The prediction error is irregularly distributed. Thus, a bias compensation–long short-term memory (BC–LSTM) network is proposed to minimize the prediction error. The experimental results show that the new feature selection method can improve the prediction accuracy by 0.6% and the calculation efficiency by 20% compared to using feature importance identification based on LightGBM. The BC–LSTM network can improve accuracy by 0.3% using about twice the time compared with the LSTM network, and the hybrid model can further improve prediction accuracy and efficiency based on the BC–LSTM network.

Wind power prediction (WPP) of wind farm clusters is important to the safe operation and economic dispatch of the power system, but it faces two challenges: (1) The dimensions of the input parameters for WPP of wind farm clusters are very high so that the input parameters contain irrelevant or redundant features; (2) it is difficult to build a holistic WPP model with high-dimensional input parameters for wind farm clusters. To overcome these challenges, a novel short-term WPP model for wind farm clusters, based on sequential floating forward selection (SFFS) feature selection and bidirectional long short-term memory (BLSTM) deep learning, is proposed in this paper. First, more than 300,000 input features of the wind farm cluster are constructed. Second, the SFFS method is applied to sort the high-dimensional features and analyze the rule that the forecasting accuracy changes with the number of features to obtain the optimal number of features and feature sets. Finally, based on the results of feature selection, BLSTM is applied to build a WPP model for wind farm clusters with a combination of feature selection and deep learning. This case study shows that (1) SFFS is an effective method for selecting the core features for WPP of wind farm clusters; (2) BLSTM shows not only higher WPP accuracy than long short-term memory and backpropagation neural network but also outstanding performance in terms of reducing the phase errors of WPP.

IoT time series data is an important form of big data. How to improve the efficiency of storage system is crucial for IoT time series database to store and manage massive IoT time series data from various IoT devices. Mixing NVM and SSD is an effective method to improve the I/O performance of storage systems. However, there are great differences between HDD and NVM or SSD. As a result, NVM and SSD cannot be directly used in the current time series database effectively. We design an IoT time series database with an embedded engine in storage device drivers for the hybrid solid-state storage system consisting of NVM and SSD. The I/O software stack of storing and managing IoT time series data can be shortened to improve the efficiency. Based upon the intrinsic characteristics of IoT time series data and different features of NVM and SSD, a redundancy elimination and compression fusion strategy, a hierarchical management strategy, and a heterogeneous time series data index are designed to improve the efficiency. Finally, a prototype of embedded IoT time series database named TS-NSM is implemented, and YCSB-TS is used to measure the IOPS. The results show that TS-NSM can improve the write IOPS up to 243.6 times and 174.3 times, respectively, compared with InfluxDB and OpenTSDB, and improve the read IOPS up to 10.1 times and 14.4 times, respectively.

Abstract-WebRTC enables web browsers with real-time communications capabilities via JavaScript APIs. But when the number of the participants increases, the bandwidth and CPU requirements have become a serious issue in a push based mesh network. In this paper, we propose a P2P-MCU approach for multi-party video conferencing that efficiently supports both ordinary smart mobile phones and PCs. In our approach, a MCU module is integrated into the browser to mix and transcode the video & audio streams in real time. And when the browser acts as the MCU node leaves the conference session without notice, another candidate browser can take over the control immediately, and the ongoing WebRTC conference can be seamlessly recovered with our MCU selection algorithm. In addition, our approach works under the 3G symmetric NAT networks by using some UDP hole punching method. Our P2P-MCU solution reduces 64% CPU usages and 35% bandwidth consumptions for each participant compared to the mesh-network solution in our eight-party WebRTC conference experiments. Although the P2P-MCU module may introduce some delay (<500ms), the delay is stable and perceptually almost neglectable.Index Terms-MCU, P2P, video conference, WebRTC. I. INTRODUCTIONDriven by the widespread fixed and mobile broadband networks, there is a trend to have real time multi-party video conferences at any time/place. To meet the emerging requirements, WebRTC [1] (Web Real-Time Communications) received a great interest since the API is inherently supported by many new versions of popular browsers, i.e. Google Chrome and Mozilla Firefox. However, since WebRTC is initially designed for browser to browser communication, even for a small scale group, the multi-party conference model may be either complicated or expensive.In particular, to support N conference participants with a pure Mesh network, there will be N*(N-1)/2 links. The bandwidth/device capability requirements will increase quadratically to the number of the participants in the conference. Accordingly, a MCU [2] (Multi point Control Unit) server is introduced to reduce bandwidth consumption by mixing the media received from users in the conference into a single stream to each participant. However, MCU server, typically based on a fixed and pre-configured hardware, is often quite costly and it consumes significant amount of bandwidth. In this paper, we describe our approach to peer-to-peer MCU (P2P-MCU) that tackles the abovementioned issues. Moreover, in our approach, the MCU is integrated in a browser at the client side, and this specific client is called MCU host. Accordingly, the media flows in the conference run in a P2P manner between the MCU host and web browsers. The proposed approach is implemented and we demonstrate the web applications that we developed for an eight party WebRTC conferencing including mobile clients.The contributions of this paper are: firstly, we design a P2P-MCU architecture working with current WebRTC protocols; secondly, we propose a MCU host determination strategy to dynamically...