基于特征选择和组合模型的短期电力负荷预测

提出基于特征选择和组合模型的短期电力负荷预测方法。首先将特征向量按特点分为2类,分别使用斯皮尔曼相关系数、最大相关最小冗余算法进行选择,依据贝叶斯信息量准则确定最优特征向量维度。然后使用3个不同的核函数建立单核递归支持向量回归模型并完成预测。最后构建神经网络,进行实验分析。仿真结果表明所提方法具有较高的预测精度与鲁棒性。     

基于集成经验模态分解和最小二乘双支持向量回归机的风速预测算法研究

为有效解决风电大规模并网过程中面临的并网难和弃风等问题,实现可再生能源大规模平滑并网并保证大电网的安全稳定运行,采用集成经验模态分解(ensemble empirical mode decomposition,EEMD)和最小二乘双支持向量回归机(least square twin support vector regression,LSTSVR)算法进行风电场风速预测。分别介绍了LSTSVR、EEMD及自适应变异粒子群算法原理。给出基于EEMD和LSTSVR的风速预测流程,以安徽女儿岭风电场测风声雷达30、70 m处风速采样数据为例,开展基于EEMD和LSTSVR的风速预测算法验证,预测结果误差分析表明:基于EEMD+LSTSVR+自适应变异粒子群算法可以实现风电场风速的高精度预测。     

双碳及能源安全背景下中国电煤月度需求组合预测

双碳发展背景下,准确及时的电煤需求预测有利于国家制定电煤供需计划及能源安全供给。现有研究对电煤月度需求预测的较少,且精度欠佳,难以保证电煤月度需求感知的及时性和准确性。提出由回归加权融合反向传播神经网络、门控循环单元网络和长短期记忆网络的组合预测模型预测我国电煤月度需求。首先采用格兰杰因果检验从月度经济及能源生产因素中筛选显著影响电煤需求的变量。然后构建单一及组合模型进行预测,其中,组合模型权重由回归计算得到。结果表明,相较单一模型,组合预测模型在电煤需求预测中性能更好。     

Stock-Price Forecasting Based on XGBoost and LSTM

Using time-series data analysis for stock-price forecasting (SPF) is complex and challenging because many factors can influence stock prices (e.g., inflation, seasonality, economic policy, societal behaviors). Such factors can be analyzed over time for SPF. Machine learning and deep learning have been shown to obtain better forecasts of stock prices than traditional approaches. This study, therefore, proposed a method to enhance the performance of an SPF system based on advanced machine learning and deep learning approaches. First, we applied extreme gradient boosting as a feature-selection technique to extract important features from high-dimensional time-series data and remove redundant features. Then, we fed selected features into a deep long short-term memory (LSTM) network to forecast stock prices. The deep LSTM network was used to reflect the temporal nature of the input time series and fully exploit future contextual information. The complex structure enables this network to capture more stochasticity within the stock price. The method does not change when applied to stock data or Forex data. Experimental results based on a Forex dataset covering 2008–2018 showed that our approach outperformed the baseline autoregressive integrated moving average approach with regard to mean absolute error, mean squared error, and root-mean-square error.     

基于双重注意力机制和GRU网络的短期负荷预测模型

电力负荷预测对电力系统的部署、规划和运行影响重大,但目前各输入特征对电网负荷情况影响的程度不稳定,且递归神经网络捕获负荷数据的长期记忆能力差,导致预测精度下降。提出一种基于双重注意力机制和GRU网络的预测新模型,利用特征注意力机制自主分析历史信息与输入特征间的关联关系,提取重要特征,并通过时序注意力机制自主选取GRU网络中关键时间点的历史信息,提升较长时间段预测效果的稳定性。在3个公开数据集上的实验结果表明,该模型在预测精度指标上表现良好,对比SVR、KPCA-ELM、DBN、GRU、Attention-GRU、CNN-LSTM、Attention-CNN-GRU模型预测精度分别提高了2.47、1.14、1.93、1.37、1.04、0.74、0.41个百分点。     

Development and evaluation of a rolling horizon purchasing policy for cores

A number of companies utilise end-of-use products (i.e. cores) for remanufacturing or recycling. An adequate supply of cores is needed for such activities. Establishing a purchasing policy for cores, over a finite planning horizon, requires multi-step ahead forecasts. Such forecasts are complicated by the fact that the number of cores in any future period depends upon previous sales and recent returns of the product. Distributed lag models have been used to capture this dependency for single-period ahead forecasts. We develop an approach to use distributed lag models to make multi-period ahead forecasts of net demand (i.e. demand minus returns), and investigate the cost implications, at a prescribed service level, of using such forecasts to purchase cores on a rolling horizon basis. Our results indicate that the effects of errors in the sales forecasts are negligible if sales follow an autoregressive pattern but are substantial when sales are more random. Dynamic estimation of the parameters in a rolling horizon environment yielded the most cost savings at high prescribed service levels (i.e. >0.95). Collectively, our results demonstrate the conditions in which companies can best leverage the dynamic nature of distributed lag models to reduce the acquisition costs over a finite horizon.     

霍泉灌区作物需水量实时预报

采用联合国粮农组织最新推荐的修正Ｐｅｎｍａｎ公式分析计算了霍泉灌区长系列参考作物的腾发量及其变化规律,在此基础上提出了该灌区冬小麦、夏玉米需水量预报模型．分析了预报模型中参数Ａ０的变化规律,并在预报过程中进行调整．运用实测资料检验表明,修正后的模型在逐日作物需水量预报中更符合实际,具有比较高的精度．     

基于本体的组合预测预警系统模型

在分析现有预测预警系统不足的基础上,提出一种基于本体的组合预测预警模型(OCFWSM),解决组合预测预警系统内各单一预测模型和指标体系间的语义异构问题,通过建立本体知识库,实现系统内数据、模型、知识的一致化表示,实现各模型和指标体系间的知识共享和重用.具体分析了本体知识库的组成及其实现原理和方法,根据本体建模理论,建立...     

SA-SVR在移动通信话务量预测中的应用

根据移动通信话务量的时间序列,采用基于模拟退火(SA)算法对超参数选择的支持向量回归机(SVR)进行建模预测。比较ARIMA、人工神经网络和SVR 3种模型的预测效果,并对比研究网格法、遗传算法和SA 3种SVR超参数选择方法对预测效果的影响。实验结果表明,SA-SVR预测精度高、耗时少,是一种预测移动通信话务量的有效方法。     

智能式动态路线诱导系统诱导单元决策方法

诱导单元决策方法的研究是实现智能交通系统的一个关键环节,其实现方法的好坏对诱导效果有直接的影响.为此,提出一种智能动态路线诱导系统诱导信息的模式,对诱导信息所在路段的多目标终点交通流构成比例以及到下游路段的转弯率的预测方法进行研究,并在此基础上提出了智能动态路线诱导系统诱导单元的综合决策方法.实验结果及分析表明了所提出的诱导单元决策方法的有效性.