基于VMD和改进TCN的短期光伏发电功率预测

光伏发电功率存在波动性,且光伏出力易受各种气象特征影响,传统TCN网络容易过度强化空间特性而弱化个体特性。针对上述问题,文中提出一种基于VMD和改进TCN的短期光伏发电功率预测模型。通过VMD将原始光伏发电功率时间序列分解为若干不同频率的模态分量,将各个模态分量以及相对应的气象数据输入至改进TCN网络进行建模学习。利用中心频率法确定VMD的最优分解模态分解个数。在传统TCN预测模型的基础上,使用DropBlock正则化取代Dropout正则化以达到抑制卷积层中信息协同的效果,并引入注意力机制自主挖掘并突出关键气象输入特征的影响,量化各气象因素对光伏发电的影响,从而提高预测精度。以江苏省某光伏电站真实数据为例进行仿真实验,结果表明所提预测方法的RMSE为0.62 MW,MAPE为2.03%。     

Measuring quality of government in EU regions across space and time

A wave of recent cross-national research has pointed to the positive consequences for countries with high levels of “quality of government” (QoG), broadly defined, such as corruption, impartiality, and quality of public services. Yet the question of how QoG varies at the sub-national level is still widely overlooked, in particular with measures that are available over time. To address it, we present the third round of data from the regional European Quality of Government Index (EQI) survey corruption (D73), Europe (N44) governance (H11); sub-national (R50), time series (C22), collected in 2017 and built upon the opinions of 78,000 respondents in 193 regions from 21 European countries. The data provides several contributions to the literature. First, while the majority of QoG-type indices rely on expert assessments, the EQI relies on the assessments of citizens, who are the on-the-ground consumers of public services. Second, the data begins to show trends on QoG variation over time, as well as across European regions. Consequently, this data is the most comprehensive sub-national data to date; mapping of QoG within and across EU countries over the past decade. Building on previous rounds of data collected in 2010 and 2013, the 2017 EQI, which is published free for scholarly use, builds on both perceptions and experiences of citizens in public service areas such as health care, education, and law enforcement. This paper presents the results of the latest survey, improved with respect to the previous ones, discussion of trends across space and over time, as well as interesting avenues for future research that we detect across European regions.     

An automatic tool for reconstructing monthly time-series of hydro-climatic variables at ungauged basins

Integrative information models for filling/reconstructing hydro-climatic time-series are required for a variety of practical applications. A GIS-based model for a rapid and reliable assessment of monthly time-series of several key hydro-climatic variables at the basin scale, is here developed as plug-in and applied to the entire region of Sicily (Italy). The plug-in, once the desired basin outlet section and time-window are selected, uses appropriate spatial techniques and algorithms to identify its drainage area and estimate the corresponding mean areal rainfall and temperatures time-series. A recent regional regressive rainfall-runoff model is successively applied for the assessment of the runoff time-series. Finally, a consolidated temperature-based method is applied to estimate monthly potential evapotranspiration time-series, while, actual evapotranspiration and soil moisture storage time-series are derived through a classical water balance model. The tool, supported by a preliminarily developed database, includes automatic procedures for data retrieving and processing and a user friendly interface.     

Sensory Data Prediction Using Spatiotemporal Correlation and LSTM Recurrent Neural Network

The Wireless Sensor Networks (WSNs) are widely utilized in various industrial and environmental monitoring applications. The process of data gathering within the WSN is significant in terms of reporting the environmental data. However, it might occur that certain sensor node malfunctions due to the energy draining out or unexpected damage. Therefore, the collected data may become inaccurate or incomplete. Focusing on the spatiotemporal correlation among sensor nodes, this paper proposes a novel algorithm to predict the value of the missing or inaccurate data and predict the future data in replacement of certain nonfunctional sensor nodes. The Long-Short-Term-Memory Recurrent Neural Network (LSTM RNN) helps to more accurately derive the time-series data corresponding to the sets of past collected data, making the prediction results more reliable. It is observed from the simulation results that the proposed algorithm provides an outstanding data gathering efficiency while ensuring the data accuracy.     

Artificial intelligence and healthcare: Forecasting of medical bookings through multi-source time-series fusion

Nowadays, Artificial intelligence (AI), combined with the digitalization of healthcare, can lead to substantial improvements in Patient Care, Disease Management, Hospital Administration, and supply chain effectiveness. Among predictive analytics tools, time series forecasting represents a central task to support healthcare management in terms of bookings and medical services predictions. In this context, the development of flexible frameworks to provide robust and reliable predictions became a central point in this healthcare innovation process. This paper presents and discusses a multi-source time series fusion and forecasting framework relying on Deep Learning. By combining weather, air-quality and medical bookings time series through a feature compression stage which preserves temporal patterns, the prediction is provided through a flexible ensemble technique based on machine learning models and a hybrid neural network. The proposed system is able to predict the number of bookings related to a specific medical examination for a 7-days horizon period. To assess the proposed approach’s effectiveness, we rely on time series extracted from a real dataset of administrative e-health records provided by the Campania Region health department, in Italy.     

时间序列分析方法在干旱区地下水位动态预测中的应用

地下水动力学模型与时间序列模型都可用于分析、预测干旱区地下水位埋深动态变化。但比较而言,时间序列模型无须进行专门的试验来获取有关地质参数,方法易于掌握,计算量小,因而更便于应用推广。本文采用时间序列分析方法分析新疆南部平原区地下水位变化,通过HP（Hodrick-Prescott）滤波方法与ARMA（p,q）模型相结合...     

按时序特征优化模型后在线选配的超短期风电预测

讨论超短期风电功率预测(USTWPP)模型的适用性。提出的USTWPP方法,从历史数据的风电功率时间序列(WPTS)中筛选特征量,选择门限值,并将短窗口内的WPTS划分为不同形态的子集,以及一个囊括所有不具有排他性分类特征的"非形态子集"。然后在离线环境下,分别按对应的训练样本优化各子集专用的预测模型及参数。在线应用时,将当前时刻前一个短窗口的WPTS与各子集的分类判据比对,以归入上述子集之一,然后调用相应的预测模型完成USTWPP。最后,以实际算例验证了该方法的有效性。     

时序犹豫模糊软集及其在决策中的应用

针对犹豫模糊软集的信息随着时间动态变化的情形,引入时间参数,将犹豫模糊软集推广为时序犹豫模糊软集。基于时序犹豫模糊软集的概念,定义了其基本的运算法则,分析对应的运算结果并讨论其运算性质。给出了时序犹豫模糊软集的一种决策方法,并通过实例表明了该方法的有效性与可行性。     

状态估计在最小方差控制中的应用

通过引入状态变量,将普通的控制系统改写为状态空间模型,应用时域上的现代时间序列分析方法和射影理论,基于ARMA新息模型和白噪声估值器,给出状态变量的滤波估值,并用状态变量的滤波代替控制系统中的观测,从而得到含有新息补偿的最小方差控制律,它比直接由控制系统得到的最小方差控制律更加准确,也更具应用价值。     

Extracting viewer interests for automated bookmarking in video-on-demand services

Video-on-demand (VoD) services have become popular on the Internet in recent years. In VoD, it is challenging to support the VCR functionality, especially the jumps, while maintaining a smooth streaming quality. Previous studies propose to solve this problem by predicting the jump target locations and prefetching the contents. However, through our analysis on traces from a real-world VoD service, we find that it would be fundamentally difficult to improve a viewer’s VCR experience by simply predicting his future jumps, while ignoring the intentions behind these jumps. Instead of the prediction-based approach, in this paper, we seek to support the VCR functionality by bookmarking the videos. There are two key techniques in our proposed methodology. First, we infer and differentiate viewers’ intentions in VCR jumps by decomposing the interseek times, using an expectation-maximization (EM) algorithm, and combine the decomposed inter-seek times with the VCR jumps to compute a numerical interest score for each video segment. Second, based on the interest scores, we propose an automated video bookmarking algorithm. The algorithm employs the time-series change detection techniques of CUSUMandMB-GT, and bookmarks videos by detecting the abrupt changes on their interest score sequences. We evaluate our proposed techniques using real-world VoD traces from dozens of videos. Experimental results suggest that with our methods, viewers’ interests within a video can be precisely extracted, and we can position bookmarks on the video’s highlight events accurately. Our proposed video bookmarking methodology does not require any knowledge on video type, contents, and semantics, and can be applied on various types of videos.