基于XGBOOST-DNN的中期电力负荷预测

精准的负荷预测是电力工作者重要的工作之一,而负荷预测以预测周期的不同,一般可以划分为短期电力负荷预测与中长期电力负荷预测.其中中长期电力负荷预测相较短期电力负荷预测而言,该领域缺乏大量前沿工作者的探索.因此本文提出一种可应用于中期电力负荷预测领域且基于XGBoost-DNN的算法.该算法将树模型和深度神经网络相结合,并将短期电力负荷预测引入到了中期电力负荷预测的工作中,基于树模型自身特点,将数据特征加工成高阶的交叉特征,同时结合原有数据利用深度神经网络可学习到丰富的特征信息.这里是以2017全球能源预测竞赛的数据进行算法分析,其中实验表明,在中期电力负荷预测领域,该方法提出的XGBoost-DNN模型相较于DNN,LSTM而言,其具备更加精准的准确性.     

包容性检验和PCA相融合的物流需求预测

模型选择以及如何进行组合是物流需求组合预测的关键,为了提高物流需求的预测精度,提出一种包容性检验和主成分分析相融合的物流需求预测模型（ET-PCA）。采用多个单一模型对物流需求进行预测,采用包容性检验选择最合理的单一模型,利用PCA对选择的单一模型预测结果进行组合,采用仿真实验对组合模型性能进行测试。结果表明,相对于传统组合模型,ET-PCA较好地解决了物流需求单一预测模型选择及组合问题,更加全面、准确描述了物流需求复杂的变化趋势,提高了物流需求的预测精度和效率,具有一定应用价值。     

A hybrid statistical genetic-based demand forecasting expert system

Demand forecasting is considered a key factor for balancing risk of over-stocking and out-of-stock. It is the main input to supply chain processes affecting their performance. Even with much effort and funds spent to improve supply chain processes, they still lack reliability and efficiency if the demand forecast accuracy is poor. This paper presents a proposal of an integrated model of statistical methods and improved genetic algorithm to generate better demand forecast accuracy. An improved genetic algorithm is used to choose the best weights among the statistical methods and to optimize the forecasted activities combinations that maximize profit. A case study is presented using different product types. And, a comparison is conducted between results obtained from the proposed model and from traditional statistical methods, which demonstrates improved forecast accuracy using the proposed model for all time series types.     

A Bayesian approach to demand forecasting for new equipment programs

Demand forecasting is a fundamental component in a range of industrial problems (e.g., inventory management, equipment maintenance). Forecasts are crucial to accurately estimating spare or replacement part demand to determine inventory stock levels. Estimating demand becomes challenging when parts experience intermittent demand/failures versus demand at more regular intervals or high quantities. In this paper, we develop a demand forecasting approach that utilizes Bayes’ rule to improve the forecast accuracy of parts from new equipment programs where established demand patterns have not had sufficient time to develop. In these instances, the best information available tends to be “engineering estimates” based on like /similar parts or engineering projections. A case study is performed to validate the forecasting methodology. The validation compared the performance of the proposed Bayesian method and traditional forecasting methods for both forecast accuracy and overall inventory fill rate performance. The analysis showed that for specific situations the Bayesian-based forecasting approach more accurately predicts part demand, impacting part availability (fill rate) and inventory cost. This improved forecasting ability will enable managers to make better inventory investment decisions for new equipment programs.     

Applications of artificial intelligence for optimization of compressor scheduling

This paper presents a feasibility study of evolutionary scheduling for gas pipeline operations. The problem is complex because of several constraints that must be taken into consideration during the optimization process. The objective of gas pipeline operations is to transfer sufficient gas from gas stations to consumers so as to satisfy customer demand with minimum costs. The scheduling involves selection of a set of compressors to operate during a shift. The scheduling decision has to be made so as to satisfy the dual objectives of minimizing the sum of fuel cost, start-up cost, the cost of gas wasted due to oversupply, and satisfying minimal operative and inoperative time of the compressors. The problem was decomposed into the two subproblems of gas load forecast and selection of compressors. Neural networks were used for forecasting the load; and genetic algorithms were used to search for a near optimal combination of compressors. The study was conducted on a subsystem of the pipeline network located in southeastern Saskatchewan, Canada. The results are compared with the solutions generated by an expert system and a fuzzy linear programming model.     

AKSNet: A novel convolutional neural network with adaptive kernel width and sparse regularization for machinery fault diagnosis

Convolutional kernels have significant affections on feature learning of convolutional neural network (CNN). However, it is still a challenging problem to determine appropriate kernel width. Moreover, some features learned by convolutional layers are still redundant and noisy. Thus, adaptive selection of kernel width and feature selection of feature maps are key techniques to improve feature learning performance of CNNs. In this paper, a new deep neural network (DNN) model, adaptive kernel sparse network (AKSNet) is proposed to extract multi-scale fault features from one-dimensional (1-D) vibration signals. Firstly, an adaptive kernel selection method is developed, where multiple branches with different kernels are used to extract multi-scale features from vibration signals. Channel-wise attention is developed to fuse features generated by these kernels to obtain different informative scales. Secondly, a spatial attention is used for dynamic receptive field to focus on salient region of feature maps. Thirdly, a sparse regularization layer is embedded in the deep network to further filter noise and highlight impaction of the feature maps. Finally, two cases are adopted to verify effectiveness of AKSNet-based feature learning for bearing fault diagnosis. Experimental results show that AKSNet can effectively extract features from multi-channel vibration signals and then improves fault diagnosis performance of the classifier significantly. AKSNet shows better recognition performance in comparison with that of shallow neural networks and other typical DNNs.     

A probabilistic methodology for quantifying,diagnosing and reducing model structural and predictive errors in short term water demand forecasting

Accurate forecasts of water demand are required for real-time control of water supply systems under normal and abnormal conditions. A methodology is presented for quantifying, diagnosing and reducing model structural and predictive errors for the development of short term water demand forecasting models. The methodology (re-)emphasises the importance of posterior predictive checks of modelling assumptions in model development, and to account for inherent demand uncertainty, quantifies model performance probabilistically through evaluation of the sharpness and reliability of model predictive distributions. The methodology, when applied to forecast demand for three District Meter Areas in the UK, revealed the inappropriateness of simplistic Gaussian residual assumptions in demand forecasting. An iteratively revised, parsimonious model using a formal Bayesian likelihood function that accounts for kurtosis and heteroscedasticity in the residuals led to sharper yet reliable predictive distributions that better quantifies the time varying nature of demand uncertainty across the day in water supply systems.     

泊松分布灰色理论在物流需求预测中的应用

研究物流需求预测准确度问题。物流需求预测中存在数据小以及非线性特点,使预测系统存在不确定性。为解决上述问题,提出了一种泊松分布的神经网络需求预测算法,采用泊松分布算法对物流的整体需求进行分类,然后采用灰色理论算法选择物流需求影响因子,对物流的需求进行实时预测,仿真结果表明,改进物流需求预测方法比传统的灰色理论预测模型以及BP神经网络具有更高的预测精确度,有效地提高了区域物流需求的预测准确度,具有一定的实际应用价值。     

嵌入互联网舆情强度的人民币汇率预测

针对目前人民币汇率预测研究存在的数据源单一导致难以提升预测效果的问题,提出一种嵌入互联网舆情强度的预测技术,通过融合多方面数据源进行对比分析,有效降低了人民币汇率的预测误差。首先,融合互联网外汇新闻数据和历史行情数据,并将多源文本数据转化为可计算的特征向量;其次,通过情感特征向量构建五种特征组合并对其进行对比,给出了嵌入互联网舆情强度的特征组合作为预测模型输入;最后,设计外汇舆情影响汇率预测的滑动时间窗口,建立基于机器学习的汇率预测模型。实验结果表明,嵌入互联网舆情的特征组合相对于不含舆情的特征组合在均方根误差（RMSE）和平均绝对误差（MAE）上分别提升了9.8%和16.2%;此外,长短期记忆网络（LSTM）预测模型比支持向量回归（SVR）、决策回归（DT）和深度神经网络（DNN）预测模型表现更好。     

基于改进遗传算法的SVR短期电力负载预测

为了有效且精确地预测电力负载,提出一种基于支持向量回归(Support Vector Regression, SVR)的预测方法对负载消耗进行建模,同时提出一种基于遗传算法(Genetic algorithm, GA)的两级改进遗传算法(Modified Genetic Algorithm, MGA)以调整SVR中的参数。在满足SVR约束条件的情况下选用平均绝对百分比误差(Mean Absolute Percentage Error, MAPE)作为MGA的适应度函数。最后使用一组实际数据对基于MGA的SVR预测方法的可行性和有效性进行了验证。      

Towards novel deep neuroevolution models: chaotic levy grasshopper optimization for short-term wind speed forecasting

High accurate wind speed forecasting plays an important role in ensuring the sustainability of wind power utilization. Although deep neural networks (DNNs) have been recently applied to wind time-series datasets, their maximum performance largely leans on their designed architecture. By the current state-of-the-art DNNs, their architectures are mainly configured in manual way, which is a time-consuming task. Thus, it is difficult and frustrating for regular users who do not have comprehensive experience in DNNs to design their optimal architectures to forecast problems of interest. This paper proposes a novel framework to optimize the hyperparameters and architecture of DNNs used for wind speed forecasting. Thus, we introduce a novel enhanced version of the grasshopper optimization algorithm called EGOA to optimize the deep long short-term memory (LSTM) neural network architecture, which optimally evolves four of its key hyperparameters. For designing the enhanced version of GOA, the chaotic theory and levy flight strategies are applied to make an efficient balance between the exploitation and exploration phases of the GOA. Moreover, the mutual information (MI) feature selection algorithm is utilized to select more correlated and effective historical wind speed time series features. The proposed model’s performance is comprehensively evaluated on two datasets gathered from the wind stations located in the United States (US) for two forecasting horizons of the next 30-min and 1-h ahead. The experimental results reveal that the proposed model achieves the best forecasting performance compared to seven prominent classical and state-of-the-art forecasting algorithms.

     

A comprehensive survey on optimizing deep learning models by metaheuristics

Deep neural networks (DNNs), which are extensions of artificial neural networks, can learn higher levels of feature hierarchy established by lower level features by transforming the raw feature space to another complex feature space. Although deep networks are successful in a wide range of problems in different fields, there are some issues affecting their overall performance such as selecting appropriate values for model parameters, deciding the optimal architecture and feature representation and determining optimal weight and bias values. Recently, metaheuristic algorithms have been proposed to automate these tasks. This survey gives brief information about common basic DNN architectures including convolutional neural networks, unsupervised pre-trained models, recurrent neural networks and recursive neural networks. We formulate the optimization problems in DNN design such as architecture optimization, hyper-parameter optimization, training and feature representation level optimization. The encoding schemes used in metaheuristics to represent the network architectures are categorized. The evolutionary and selection operators, and also speed-up methods are summarized, and the main approaches to validate the results of networks designed by metaheuristics are provided. Moreover, we group the studies on the metaheuristics for deep neural networks based on the problem type considered and present the datasets mostly used in the studies for the readers. We discuss about the pros and cons of utilizing metaheuristics in deep learning field and give some future directions for connecting the metaheuristics and deep learning. To the best of our knowledge, this is the most comprehensive survey about metaheuristics used in deep learning field.

     

基于非负矩阵分解的语音深层低维特征提取方法

作为一种基于深层神经网络提取的低维特征,瓶颈特征在连续语音识别中取得了很大的成功。然而训练瓶颈结构的深层神经网络时,瓶颈层的存在会降低网络输出层的帧准确率,进而反过来影响该特征的性能。针对这一问题,本文基于非负矩阵分解算法,提出一种利用不包含瓶颈层的深层神经网络提取低维特征的方法。该方法利用半非负矩阵分解和凸非负矩阵分解算法对隐含层权值矩阵分解得到基矩阵,将其作为新的特征层权值矩阵,然后在该层不设置偏移向量的情况下,通过数据前向传播提取新型特征。实验表明,该特征具有较为稳定的规律,且适用于不同的识别任务和网络结构。当使用训练数据充足的语料进行实验时,该特征表现出同瓶颈特征几乎相同的识别性能;而在低资源环境下,基于该特征识别系统的识别率明显优于深层神经网络混合识别系统和瓶颈特征识别系统。     

数字孪生甬江流域水资源“四预”应用研究

基于当前数字孪生流域发展背景及宁波市水资源利用特点，为满足宁波市水资源在不同时空下合理分配的需求，以甬江流域水资源管理与调配“四预”流程为切入点，深度融合数字孪生、BIM 建模等现代化信息技术，在建立来水预报分析、需水预测、水资源优化调配、水资源实时分析评价及水资源预警等各类模型的基础上， 构建具有“预报分析—监测预警—调配预演—调度预案—动态评价”功能的水资源管理与调配业务应用系统，以达到及时准确预报、全面精准预警、同步仿真预演、精细数字预案、多维动态评价的目标，最终实现水资源的智能优化调配与管理。     

Hybrid morphological methodology for software development cost estimation

In this paper we propose a hybrid methodology to design morphological-rank-linear (MRL) perceptrons in the problem of software development cost estimation (SDCE). In this methodology, we use a modified genetic algorithm (MGA) to optimize the parameters of the MRL perceptron, as well as to select an optimal input feature subset of the used databases, aiming at a higher accuracy level for SDCE problems. Besides, for each individual of MGA, a gradient steepest descent method is used to further improve the MRL perceptron parameters supplied by MGA. Finally, we conduct an experimental analysis with the proposed methodology using six well-known benchmark databases of software projects, where two relevant performance metrics and a fitness function are used to assess the performance of the proposed methodology, which is compared to classical machine learning models presented in the literature.     

鱼群涌现机制下集群机器人运动强化的迁移控制

采用鱼群模型驱动多智能体可以涌现出优良的运动特性,但是,由于机器人与真实鱼类相比具有较大的差异性,使得鱼群模型难以应用于真实机器人系统.为此,提出一种结合深度学习与强化学习的迁移控制方法,首先,使用鱼群运动数据训练深度网络(deep neural network, DNN)模型,以此作为机器人成对交互的基础;然后,连接强化学习的深度确定性策略梯度方法(deep deterministic policy gradient, DDPG)来修正DNN模型的输出,设计集群最大视觉尺寸方法挑选关键邻居,从而将DNN+DDPG模型拓展到多智能体的运动控制.集群机器人运动实验表明:所提出方法能使机器人仅利用单个邻居信息就能形成可靠、稳定的集群运动,与单纯DNN直接迁移控制相比,所提出DNN+DDPG控制框架既可以保存原有鱼群运动的灵活性,又能增强机器人系统的安全性与可控性,使得该方法在集群机器人运动控制领域具有较大的应用潜力.     

High quality voice conversion using prosodic and high-resolution spectral features

Voice conversion methods have advanced rapidly over the last decade. Studies have shown that speaker characteristics are captured by spectral feature as well as various prosodic features. Most existing conversion methods focus on the spectral feature as it directly represents the timbre characteristics, while some conversion methods have focused only on the prosodic feature represented by the fundamental frequency. In this paper, a comprehensive framework using deep neural networks to convert both timbre and prosodic features is proposed. The timbre feature is represented by a high-resolution spectral feature. The prosodic features include F0, intensity and duration. It is well known that DNN is useful as a tool to model high-dimensional features. In this work, we show that DNN initialized by our proposed autoencoder pretraining yields good quality DNN conversion models. This pretraining is tailor-made for voice conversion and leverages on autoencoder to capture the generic spectral shape of source speech. Additionally, our framework uses segmental DNN models to capture the evolution of the prosodic features over time. To reconstruct the converted speech, the spectral feature produced by the DNN model is combined with the three prosodic features produced by the DNN segmental models. Our experimental results show that the application of both prosodic and high-resolution spectral features leads to quality converted speech as measured by objective evaluation and subjective listening tests.     

Impact of forecasting error on the performance of capacitated multi-item production systems

A computer model is built to simulate master production scheduling activities in a capacitated multi-item production system under demand uncertainty and a rolling time horizon. The output from the simulation is analyzed through statistical software. The results of the study show that forecasting errors have significant impacts on total cost, schedule instability and system service level, and the performance of forecasting errors is significantly influenced by some operational factors, such as capacity tightness and cost structure. Furthermore, the selection of the master production schedule freezing parameters is also significantly influenced by forecasting errors. The findings from this study can help managers optimize their production plans by selecting more reasonable forecasting methods and scheduling parameters, thus improving the performance of production systems.     

Forecasting model selection through out-of-sample rolling horizon weighted errors

Demand forecasting is an essential process for any firm whether it is a supplier, manufacturer or retailer. A large number of research works about time series forecast techniques exists in the literature, and there are many time series forecasting tools. In many cases, however, selecting the best time series forecasting model for each time series to be dealt with is still a complex problem. In this paper, a new automatic selection procedure of time series forecasting models is proposed. The selection criterion has been tested using the set of monthly time series of the M3 Competition and two basic forecasting models obtaining interesting results. This selection criterion has been implemented in a forecasting expert system and applied to a real case, a firm that produces steel products for construction, which automatically performs monthly forecasts on tens of thousands of time series. As result, the firm has increased the level of success in its demand forecasts.