A hybrid forecasting model for depth-averaged current velocities of underwater gliders

In this paper,we propose a hybrid forecasting model to improve the forecasting accuracy for depth-averaged current velocities(DACVs) of underwater gliders.The hybrid model is based on a discrete wavelet transform(DWT),a deep belief network(DBN),and a least squares support vector machine(LSSVM).The original DACV series are first decomposed into several high-and one low-frequency subseries by DWT.Then,DBN is used for high-frequency component forecasting,and the LSSVM model is adopted for low-frequency subseries.The effectiveness of the proposed model is verified by two groups of DACV data from sea trials in the South China Sea.Based on four general error criteria,the forecast performance of the proposed model is demonstrated.The comparison models include some well-recognized single models and some related hybrid models.The performance of the proposed model outperformed those of the other methods indicated above.

A NEW HYBRID FORECASTING ALGORITHM AND ITS APPLICATION IN ECONOMIC ANALYSIS

There exists a great deal of periodic non-stationary processes in natural,social and eco- nomical phenomenon.It is very important to realize the dynamic analysis and real-time forecast within a period.In this letter,a wavelet-Kalman hybrid estimation and forecasting algorithm based on step-by-step filtering with the real-time and recursion property is put forward.It combines the advantages of Kalman filter and wavelet transform.Utilizing the information provided by multi- sensor effectively,this algorithm can realize not only real-time tracking and dynamic multi-step fore- casting within a period,but also the dynamic forecasting between periods,and it has a great value to the system decision-making.Simulation results show that this algorithm is valuable.

A hybrid model for short-term rainstorm forecasting based on a back-propagation neural network and synoptic diagnosis

Rainstorms are one of the most important types of natural disaster in China.In order to enhance the ability to forecast rainstorms in the short term,this paper explores how to combine a back-propagation neural network(BPNN)with synoptic diagnosis for predicting rainstorms,and analyzes the hit rates of rainstorms for the above two methods using the county of Tianquan as a case study.Results showed that the traditional synoptic diagnosis method still has an important referential meaning for most rainstorm types through synoptic typing and statistics of physical quantities based on historical cases,and the threat score(TS)of rainstorms was more than 0.75.However,the accuracy for two rainstorm types influenced by low-level easterly inverted troughs was less than 40%.The BPNN method efficiently forecasted these two rainstorm types;the TS and equitable threat score(ETS)of rainstorms were 0.80 and 0.79,respectively.The TS and ETS of the hybrid model that combined the BPNN and synoptic diagnosis methods exceeded the forecast score of multi-numerical simulations over the Sichuan Basin without exception.This kind of hybrid model enhanced the forecasting accuracy of rainstorms.The findings of this study provide certain reference value for the future development of refined forecast models with local features.

A hybrid model for numerical wave forecasting and its implementation-Ⅰ.The wind wave model

The authors make an endeavor to explain why a new hybrid wave model is here proposed when several such models have already been in operation and the so- called third generation wave modej is proving attractive. This part of the paper is devoted to the wind wave model. Both deep and shallow water models have been developed, the former being actually a special case of the latter when water depth is great. The deep water model is exceptionally simple in form. Significant wave height is the only prognostic variable. In comparison with the usual methods to compute the energy input and dissipations empirically or by 'tuning', the proposed model has the merit that the effects of all source terms are combined into one term which is computed through empirical growth relations for significant waves, these relations being, relatively speaking, easier and more reliable to obtain than those for the source terms in the spectral energy balance equation. The discrete part of the model and the implementation of the model as a whole will be discussed in the second part of the present paper.

A hybrid model for numerical wave forecasting and its implementation-Ⅱ .The discrete part and implementation of the model

In the first part of the present paper we have explained why we manage to formulate another wave prediction model when so many of them, including the so-called third generation model, have already been in use. The wind-wave part of the proposed model has also been given. Now we proceed to discuss the swell part,the implementation of the model as a prediction method,mumerical experiments done with ideal wind fields and hindcasts made in the Bohai Sea,in the neighboring seas adjacent to China and in the Northwest Pacific.

Hybrid partial least squares and neural network approach for short-term electrical load forecasting

Intelligent systems and methods such as the neural network （NN） are usually used in electric power systems for short-term electrical load forecasting. However, a vast amount of electrical load data is often redundant, and linearly or nonlinearly correlated with each other. Highly correlated input data can result in erroneous prediction results given out by an NN model. Besides this, the determination of the topological structure of an NN model has always been a problem for designers. This paper presents a new artificial intelligence hybrid procedure for next day electric load forecasting based on partial least squares （PLS） and NN. PLS is used for the compression of data input space, and helps to determine the structure of the NN model. The hybrid PLS-NN model can be used to predict hourly electric load on weekdays and weekends. The advantage of this methodology is that the hybrid model can provide faster convergence and more precise prediction results in comparison with abductive networks algorithm. Extensive testing on the electrical load data of the Puget power utility in the USA confirms the validity of the proposed approach.

草地贪夜蛾迁飞气象预报方法探索及应用

为了准确预报害虫的迁飞轨迹,将天气预报技术应用于迁飞性害虫的预测预报,基于天气预报平台,利用欧洲中心中短期气象数值预报产品以及大气环流形势及低层风动力对草地贪夜蛾(Spodoptera frugiperda)的迁飞路径、迁入时间(包括首见日及高峰日)、落区等进行预报,并基于2021年草地贪夜蛾迁飞的2次典型预报案,分析2021年草地贪夜蛾春季北迁至湖北(首见日)以及秋季南迁回湖北(高峰日)的典型天气过程以及迁飞层气象要素场,运用HYSPLIT轨迹模型模拟迁飞后向轨迹,再利用草地贪夜蛾田间监测数据、测报灯监测数据以及迁飞轨迹对预报结论进行验证。结果显示,2次典型预报案例的预报结论与草地贪夜蛾田间监测数据及测报灯监测数据以及轨迹模拟的情景吻合度较好,草地贪夜蛾迁入时间、落区及路径预报基本正确。研究表明,将天气预报技术应用于迁飞性害虫的预报具有实际可行性。

A Novel Hybrid FA-Based LSSVR Learning Paradigm for Hydropower Consumption Forecasting

Due to the nonlinearity and nonstationary of hydropower market data, a novel hybrid learning paradigm is proposed to predict hydropower consumption, by incorporating firefly algorithm （FA） into least square support vector regression （LSSVR）, i.e., FA-based LSSVR model. In the novel model, the powerful and effective artificial intelligence （AI） technique, i.e., LSSVR, is employed to forecast hydropower consumption. Furthermore, a promising AI optimization tool, i.e., FA, is espe- cially introduced to address the crucial but difficult task of parameters determination in LSSVR （e.g., hyper and kernel function parameters）. With the Chinese hydropower consumption as sample data, the empirical study has statistically confirmed the superiority of the novel FA-based LSSVR model to other benchmark models （including existing popular traditional econometric models, AI models and similar hybrid LSSVRs with other popular parameter searching tools）~ in terms of level and direc- tional accuracy. The empirical results also imply that the hybrid FA-based LSSVR learning paradigm with powerful forecasting tool and parameters optimization method can be employed as an effective forecasting tool for not only hydropower consumption but also other complex data.

基于CNN-BiGRU-Attention的短期电力负荷预测

针对目前电力负荷数据随机性强,影响因素复杂,传统单一预测模型精度低的问题,结合卷积神经网络(Convolutional neural network,CNN)、双向门控循环单元(Bi-directional gated recurrent unit,BiGRU)以及注意力机制(Attention)在短期电力负荷预测上的不同优点,提出一种基于CNN-BiGRU-Attention的混合预测模型。该方法首先通过CNN对历史负荷和气象数据进行初步特征提取,然后利用BiGRU进一步挖掘特征数据间时序关联,再引入注意力机制,对BiGRU输出状态给与不同权重,强化关键特征,最后完成负荷预测。试验结果表明,该模型的平均绝对百分比误差(Mean absolute percentage error,MAPE)、均方根误差(Root mean square error,RMSE)、判定系数(R-square,R~2)分别为0.167%、0.057%、0.993,三项指标明显优于其他模型,具有更高的预测精度和稳定性,验证了模型在短期负荷预测中的优势。

基于注意力机制的CNN-LSTM-XGBoost台风暴雨电力气象混合预测模型

极端台风暴雨灾害具有非线性、极差大以及多峰值等特点。为使电网及时获取预警信息,提出一种基于注意力机制的CNN-LSTM-XGBoost台风暴雨电力气象混合预测模型。首先,利用基于注意力机制的卷积神经网络(CNN)辨识关键台风暴雨灾害特征;然后,利用长短期记忆网络(LSTM)训练时间序列预测模型以挖掘台风暴雨时序特征,使用极限梯度提升算法替换模型输出层以缓解过拟合问题;最后,以2023年台风泰利为例验证所提方法的有效性。算例分析表明,所提模型具有较高的准确性,对预测精度的提升可达40.84%以上。

考虑光伏随机性的交直流混合配电网鲁棒机会约束安全域模型

为实现高比例光伏接入下交直流混合配电网的安全评估,构建了交直流混合配电网的鲁棒机会约束安全域模型。首先,提出一种基于区间划分的自适应带宽核密度估计方法,以获取光伏预测误差概率分布,构建光伏出力机会约束不确定集合;其次,采用功率与节点电压幅值比值为状态变量,在已有交流线性化潮流模型的基础上推导一种交直流混合配电网线性化潮流模型;然后,提出交直流混合配电网鲁棒机会约束安全域的概念和模型,结合列与约束生成算法和径向迭代搜索算法生成可观测的鲁棒机会约束安全域空间。最后,算例分析验证了所提模型能够在鲁棒机会约束安全域上准确、直观的反映光伏出力的随机性,进而为交直流混合配电网的准确安全评估提供支撑。

模态分解及混合模型在比特币价格预测中的应用

独特的生产、发行和交易机制等多种因素的影响下,比特币价格表现出极端的波动性,导致了预测任务的复杂性.为此提出了基于改进的自适应噪声完备集合经验模态分解(ICEEMDAN)的混合预测模型,将复杂的原始序列分解成多个简单固有模态函数(IMFs),并通过重构算法将IMFs集成为不同频率的分量.根据各分量的不同数据模式,选取不同机器学习模型分别进行预测,叠加各分量预测结果得到最终比特币价格预测结果.对比结果表明,该模型在各评价指标上均优于单一预测模型,混合模型可以优化预测结果,较好地减小预测误差.

基于数据分解与斑马算法优化的混合核极限学习机月径流预测

为提高月径流预测精度,改进混合核极限学习机(HKELM)预测性能,提出小波包分解(WPT)-斑马优化算法(ZOA)-HKELM组合模型。利用WPT处理月径流时序数据,构建局部高斯径向基核函数和全局多项式核函数相混合的HKELM;通过ZOA优化HKELM超参数(正则化参数、核参数、权重系数),建立WPT-ZOA-HKELM组合模型,并构建WPT-遗传算法(GA)-HKELM、WPT-灰狼优化(GWO)算法-HKELM、WPT-鲸鱼优化算法(WOA)-HKELM、WPT-ZOA-极限学习机(ELM)、WPT-ZOA-最小二乘支持向量机(LSSVM)、ZOA-HKELM作对比模型,通过黑河流域莺落峡、讨赖河水文站月径流时间序列预测实例对各模型进行检验。结果表明:(1)莺落峡、讨赖河水文站月径流时间序列WPT-ZOA-HKELM模型预测的平均绝对百分比误差分别为1.054%、0.761%,决定系数均达0.999 9,优于其他对比模型,具有更高的预测精度,预测效果更好。(2)利用ZOA优化HKELM超参数,可提高HKELM预测性能,优化效果优于GWO、WOA、GA。(3)预测模型能充分发挥WPT、ZOA和HKELM优势,提高月径流预测精度;在相同分解和优化情形下,HKELM的预测性能优于ELM、LSSVM。

基于动态关联图注意力网络的虚拟电厂居民短期负荷预测

居民短期负荷预测能够为虚拟电厂提供实时、灵活的电力需求信息,有助于虚拟电厂实现能源高效利用与优化电力市场交易。由于居民负荷相关性的日益凸显,传统预测方法仅基于单个居民历史负荷进行时序预测,无法满足规模化虚拟电厂对居民负荷关联性的综合需求。基于此,文中提出一种基于动态关联图注意力网络的虚拟电厂居民短期负荷预测方法。首先,提出了混合相关性分析方法来刻画居民负荷之间的线性和非线性关系,并进一步提出了权重剪枝阈值机制得到居民负荷混合相关性矩阵;然后,基于该矩阵构建动态关联图结构,进而提出时间图注意力网络机制以深入学习居民负荷的时空关联特性,并实现居民短期负荷预测目标;最后,以某地区实际居民负荷数据为例,验证了所提方法的有效性。

基于混合卷积神经网络的多特征负荷预测方法研究

针对负荷预测任务中准确性、稳定性和环境因素适应性的挑战,提出了一种基于混合卷积神经网络的电力负荷短期预测方法。提出了基于一维卷积神经网络(1D convolutional neural network,1D-CNN)的多尺度特征融合方法,通过融合不同尺度的特征来捕捉负荷变化的趋势,提高了对负荷突变和复杂模式的识别能力;针对多种环境特征因素对电负荷影响的问题,设计了基于2D-CNN的多特征因素学习方法,提高了模型对环境因素与负荷间复杂关系的建模能力;构建了混合网络模型,通过对1D-CNN和2D-CNN的特征信息进行深度特征融合和信息传播,实现了有效关联时空特征的综合性负荷预测方法。开展了具体算例分析研究,通过分析参数优化和融合学习对模型精度和效率的影响,并与经典模型进行对比,结果显示所提模型的均方根误差(root mean squared error,RMSE)为36.3,平均绝对误差(mean absolute error,MAE)为5.34,平均绝对百分比误差(mean absolute percentage error,MAPE)为1.02%,有效提高了负荷预测的准确性和鲁棒性。

基于Coulomb-ETAS混合模型的强余震时空发生率预测及效能评估

为考察震后早期阶段混合模型的优势及实际预测效能,提升震后早期强余震时空预测的准确性,构建可操作的地震预测研究工作模型基础,选择能够较好反映震后空间应力分布的库仑应力变化和反映余震序列衰减且拟合效果较好的ETAS模型,构建Coulomb-ETAS混合模型。此模型依据强余震大多发生在应力加载区的实际情况,利用混合模型学习期间加载区事件的发生比例,将预测率从抑制区重新分配到加载区。以2021—2022年发生的云南漾濞6.5级、青海玛多7.4级、青海门源6.9级和四川泸定6.8级4次6.0级以上强震为例,对构建的混合模型进行检验和评估,并与单一的ETAS模型、C-RS模型进行对比。研究结果表明,3个模型在震后早期阶段对强余震均表现出相对较好的预测效果,呈现出与实际结果类似的衰减特性,仅有较少的预测失效现象。在频次滑动预测上,ETAS和Coulomb-ETAS模型优于C-RS模型,统计模型优势相对比较显著;在空间发生率预测上,Coulomb-ETAS模型优于ETAS模型,认为库仑应力分布的混合模型能够降低虚报率,进而提高空间预测的准确性。因此,从4次震例上看,该混合模型优于单一的统计模型和物理模型,能够较好地适用于主震后的强余震时空预测,也可为开展多个模型混合及强余震预测之外的地震预测业务应用场景提供参考。

基于改进Transformer和超图模型的股票趋势预测方法研究

股票预测是一项令人痴迷又极具挑战的任务。近年来,融合关系信息的股票时序预测方法取得一些进展,但仍存在如下问题:首先,基于图神经网络的方法仅考虑股票之间简单的成对关系,而未考虑股票间的高阶协同关系。其次,现有方法采用预定义图的方式直接给出股票间的静态关系,无法建模股票间潜在的动态变化关系。为了解决上述问题,提出一种端到端的动态超图卷积神经网络股票趋势预测框架。该框架基于改进的Transformer提取股票的时序信息,通过静态超图和动态超图将股票间的协同关系信息引入到时序建模中。在中国A股和美股市场数据集上的实验结果表明,与当前先进模型相比,本文模型的预测性能具有显著优势。

基于GRO优化的VMD-HKELM月蒸发量预测方法研究

水面蒸发预测对于水库水量预测、区域水量平衡分析和水资源量核算等具有重要意义。水面蒸发量预测影响因素众多,并最终体现在随时间变化的蒸发量监测数据中。为此,基于淘金热(GRO)算法优化变分模态分解(VMD)-混合核极限学习机(HKELM)提出两种方案。方案Ⅰ先对月蒸发量时间序列分解,后划分训练集、测试集;方案Ⅱ先对月蒸发量划分训练集、测试集,再进行时间序列分解。通过一种新型元启发式算法对分解技术VMD、预测器HKELM超参数进行目标寻优并建立多种模型,采用云南省龙潭寨、西洋街水文站月蒸发量预测实例对方案Ⅰ、方案Ⅱ各模型进行检验。结果表明:方案Ⅰ各模型性能优于方案Ⅱ,各模型的拟合精度和预测精度总体上随分解分量数的增加而提高,但方案Ⅰ使用了测试集信息,导致预测精度虚高;方案Ⅱ各模型具有较好的预测精度和稳健性能,其用于月蒸发量时间序列预测是可行的,反映出客观真实的预测效果,具有较好的实用价值和意义。

基于QMD-HBi GRU的短期光伏功率预测方法

为了解决光伏功率数据固有的强不确定性导致单一预测模型预测精度不高的问题,提出一种基于二次模态分解和混合双向门控循环单元模型(hybrid bi-directional gated recurrent unit, HBiGRU)的短期光伏功率预测方法。首先,为应对光伏功率数据的不确定性,基于自适应噪声完备集合经验模态分解、样本熵和变分模态分解对光伏功率数据进行处理,得到一系列较为平稳的本征模函数分量;其次,构建HBi GRU模型以充分挖掘各分量与光伏功率影响因素之间的特征关系,得到各分量预测结果;最后,将各分量预测结果叠加得到短期光伏功率预测结果。以澳大利亚某地光伏电站数据进行测试,仿真结果表明:所提集成预测模型能够有效提高短期光伏功率预测精度,与其他预测模型相比,其归一化平均绝对误差和均方根误差分别降低了3.21%和5.04%,决定系数提高了22.7%。

浅谈PHEV的未来发展趋势及企业规划建议

插电混合动力汽车,简称PHEV,作为新能源汽车的重要组成部分,自2021年以来,PHEV开始进入市场化发展阶段,规模呈爆发式增长,其中2021年产销60.4万辆,2022年更是达到了146.1万辆。文章拟从市场趋势、技术动向以及企业布局三个方面重点分析PHEV未来的发展趋势,并基于以上分析对PHEV未来的市场规模进行预测,为企业的产品规划以及电动化转型提供参考。