Forecasting the energy demand of Turkey with a NN based on an improved Particle Swarm Optimization

Forecasting the future energy demand accurately is a critical issue, especially for countries like Turkey where the energy dependency ratio is high. This paper presents a neural network based on the particle swarm optimization algorithm with mutation (PSOM-NN) to enhance the prediction accuracy of the energy demand of Turkey. Unlike some studies in the field which are using all the observed data for training purpose, the proposed network used only a part of these data for training. Approximately 63 % and 37 % of the mentioned data are used for the training and test, respectively. Detrending is applied to the data to avoid nonlinear transfer functions that constrain the model to the input range values. The analysis of the results revealed that PSOM-NN produced better forecasts of energy demand compared to the earlier studies in terms of root-mean-square error, mean absolute percentage error and mean absolute deviation. Finally, future projections under different scenarios are also employed and discussed. It is believed that the proposed model could be applied to any country that needs accurate forecasts of the energy demand for sustainable energy policies.

     

Semiparametric mixed-effect least squares support vector machine for analyzing pharmacokinetic and pharmacodynamic data

In this paper we propose a semiparametric mixed-effect least squares support vector machine (LS-SVM) regression model for the analysis of pharmacokinetic (PK) and pharmacodynamic (PD) data. We also develop the generalized cross-validation (GCV) method for choosing the hyperparameters which affect the performance of the proposed LS-SVM. The performance of the proposed LS-SVM is compared with those of NONMEM and the regular semiparametric LS-SVM via four measures, which are mean squared error (MSE), mean absolute error (MAE), mean relative absolute error (MRAE) and mean relative prediction error (MRPE). Through paired-t test statistic we find that the absolute values of four measures of the proposed LS-SVM are significantly smaller than those of NONMEM for PK and PD data. We also investigate the coefficient of determinations R²'s of predicted and observed values. The R²'s of NONMEM are 0.66 and 0.59 for PK and PD data, respectively, while the R²'s of the proposed LS-SVM are 0.94 and 0.96. Through cross validation technique we also find that the proposed LS-SVM shows better generalization performance than the regular semiparametric LS-SVM for PK and PD data. These facts indicate that the proposed LS-SVM is an appealing tool for analyzing PK and PD data.     

Improving the performance of LSSVM model in predicting the safety factor for circular failure slope through optimization algorithms

Circular failure can be seen in weak rocks, the slope of soil, mine dump, and highly jointed rock mass. The challenging issue is to accurately predict the safety factor (SF) and the behavior of slopes. The aim of this study is to offer advanced and accurate models to predict the SF of slopes through machine learning methods improved by optimization algorithms. To this view, three different methods, i.e., trial and error (TE) method, gravitational search algorithm (GSA), and whale optimization algorithm (WOA) were used to investigate the proper control parameters of least squares support vector machine (LSSVM) method. In the constructed LSSVM-TE, LSSVM-GSA and LSSVM-WOA methods, six effective parameters on the SF, such as pore pressure ratio and angle of internal friction, were used as the input parameters. The results of the error criteria indicated that both GSA and WOA can improve the performance prediction of the LSSVM method in predicting the SF. However, the LSSVM-WOA method, with root mean square error of 0.141, performed better than the LSSVM-GSA with root mean square error of 0.170.

     

基于需求预测的PaaS平台资源分配方法

针对缺乏PaaS平台下资源需求的有效预测与优化分配的问题，提出一种资源需求预测模型和分配方法。首先，根据PaaS平台中应用对资源需求的周期性来对资源序列进行切分，并在短期预测的基础上结合应用的多周期性特征，利用多元回归算法建立综合的预测模型。然后，基于MapReduce架构设计实现了一个Master-Slave模式的PaaS平台资源分配系统。最后，结合当前任务请求和资源需求预测结果进行资源分配。实验结果表明，采用该资源需求预测模型和分配方法后，相比于自回归模型和指数平滑算法，平均绝对百分比误差分别下降8.71个百分点和2.07个百分点，均方根误差分别下降2.01个百分点和0.46个百分点。所提预测模型的预测结果不仅误差小，与真实值的拟合程度也较高，而且利用较小的时间开销就可以获得较高的准确度。此外，使用该预测模型的PaaS平台的资源请求的平均等待时间有了明显的下降。     

Prediction of rice crop yield using MODIS EVI−LAI data in the Mekong Delta,Vietnam

Predicting rice crop yield at the regional scale is important for production estimates that ensure food security for a country. This study aimed to develop an approach for rice crop yield prediction in the Vietnamese Mekong Delta using the Moderate Resolution Imaging Spectroradiometer (MODIS) enhanced vegetation index (EVI) and leaf area index (LAI). Data processing consisted of four main steps: (1) constructing time-series vegetation indices, (2) noise filtering of time-series data using the empirical mode decomposition (EMD), (3) establishment of crop yield models, and (4) model validation. The results indicated that the quadratic model using two variables (EVI and LAI) produced more accurate results than other models (i.e. linear, interaction, pure quadratic, and quadratic with a single variable). The highest correlation coefficients obtained at the ripening period for the spring–winter and autumn–summer crops were 0.70 and 0.74, respectively. The robustness of the established models was evaluated by comparisons between the predicted yields and crop yield statistics for 10 sampling districts in 2006 and 2007. The comparisons revealed satisfactory results for both years, especially for the spring–winter crop. In 2006, the root mean squared error (RMSE), mean absolute error (MAE), and mean bias error (MBE) for the spring–winter crop were 10.18%, 8.44% and 0.9%, respectively, while the values for the autumn–summer crop were 17.65%, 14.06%, and 3.52%, respectively. In 2007, the spring–winter crop also yielded better results (RMSE = 10.56%, MAE = 9.14%, MBE = 3.68%) compared with the autumn–summer crop (RMSE = 17%, MAE = 12.69%, MBE = 2.31%). This study demonstrates the merit of using MODIS data for regional rice crop yield prediction in the Mekong Delta before the harvest period. The methods used in this study could be transferable to other regions around the world.     

Neural network committee to predict the AMEn of poultry feedstuffs

A committee of neural networks is the aggregation of two or more neural networks for making overall predictions that are supposedly more accurate than those obtained by the individual networks. The objective of this paper was to assign some uncertainty over the predictions of neural networks, using a network committee to estimate the nitrogen-corrected metabolizable energy (AMEn) values of the energetic and protein concentrate feedstuffs for broilers. The dataset used to implement each expert network contains 568 experimental results. Another dataset with 48 bioassay results was used as test data. From several implemented multilayer perceptrons, the networks that presented the best generalization performance were selected to constitute the committee. The percentage of correct predictions was used as the criterion to compare committees that contained different numbers of networks. The highest probability density intervals were obtained for each feedstuff in the test data in this comparison. The estimator that ensured more accurate predictions was selected. The highest accuracy for predicting the AMEn values of concentrate feedstuffs for broilers was achieved by a committee with 1,000 networks with the use of the mode of the empirical distribution obtained from 1,000 estimated values of the AMEn. The accuracy of the models was evaluated based on their values of error measures between the observed and predicted values, in which the mode of the empirical distribution presented lower values of mean squared error (MSE = 45,285.43), mean absolute deviation (MAD = 177.66) and mean absolute percentage error (MAPE = 5.97 %) compared to the mean and the median.     

Canopy level estimations of chlorophyll and LAI for two tropical species (teak and bamboo) from Hyperion (EO1) data

Current economic development in tropical regions (especially in India, China, and Brazil) is putting tremendous pressure on tropical forest cover. Some of the dominant and economically important species are planted at large scale in these countries. Teak and bamboo are two important species of tropical regions because of their commercial and conservation values. Accurate estimates of foliar chemistry can help in evaluating the health status of vegetation in these regions. An attempt has been made to derive canopy level estimation of chlorophyll and leaf area index (LAI) for these species utilizing Hyperion data. Partial least square (PLS) regression analysis was carried out to identify the correlation between measured parameters (chlorophyll and LAI) and Hyperion reflectance spectra. PLS regression identified 600–750 nm as a sensitive spectral region for chlorophyll and 1000–1507 nm for LAI. The PLS regression model tested in this study worked well for the estimation of chlorophyll (R ² = 0.90, root mean square error (RMSE) = 0.182 for teak and R ² = 0.84, RMSE = 0.113 for bamboo) and for the estimation of LAI (R ² = 0.87, RMSE = 0.425). The lower predictive error obtained indicates the robustness of the data set and also of the applicability of the PLS regression analysis. Wavelengths recognized by the PLS regression model were utilized for the development of vegetation indices for estimating chlorophyll and LAI. Predictive performances of the developed simple ratios (SRs) were evaluated using the cross-validation method. SR 743/692 gave the best results for the prediction of chlorophyll with the leave-one-out cross-validation (LOO-CV) method (R ² = 0.73, RMSE = 0.28 for teak and R ² = 0.71, RMSE = 0.15 for bamboo). The normalized difference ratio (ND 1457/1084) gave the best results for the prediction of LAI with LOO-CV (R ² = 0.66, RMSE = 0.57). Ratios developed here can be tested for teak and bamboo cover spread in tropical regions with similar environmental conditions.     

Prediction of suspended sediment concentration from water quality variables

This study investigates use of water quality (WQ) variables, namely total chromium concentration, total iron concentration, and turbidity for predicting suspended sediment concentration (SSC). For this purpose, the artificial neural networks (ANNs) and regression analysis (RA) models are employed. Seven different RA models are constructed, considering the functional relation between measured WQ variables and SSC. The WQ and SSC data are fortnightly obtained from six monitoring stations, located on the stream Harsit, Eastern Black Sea Basin, Turkey. A total of 132 water samples are collected from April 2009 to February 2010. Model prediction results reveal that ANN is able to predict SSC from WQ data, with mean absolute error (MAE) of 10.30 mg/L and root mean square error (RMSE) of 13.06 mg/L. Among seven RA models, the best one, which has the form including all independent parameters, produces results comparable to those of ANN, with MAE = 14.28 mg/L and RMSE = 15.35 mg/L. The sensitivity analysis results reveal that the most effective parameter on the SSC is total chromium concentration. These results have time- and cost-saving implications.     

Development of an IoT-based big data platform for day-ahead prediction of building heating and cooling demands

The emerging technologies of the Internet of Things (IoT) and big data can be utilised to derive knowledge and support applications for energy-efficient buildings. Effective prediction of heating and cooling demands is fundamental in building energy management. In this study, a 4-layer IoT-based big data platform is developed for day-ahead prediction of building energy demands, while the core part is the hybrid machine learning-based predictive model. The proposed energy demand predictive model is based on the hybrids of k-means clustering and artificial neural network (ANN). Due to different temperatures of walls, windows, grounds, roofs and indoor air, various IoT sensors are installed at different locations of the building. To determine the input variables to the hybrid machine learning-based predictive model, correlation analysis is adopted. Through clustering analysis, the characteristic patterns of daily weather profile are identified. Thus, the annual profile is classified into several featuring groups. Each group of weather profile, along with IoT sensor readings, building operating schedules as well as heating and cooling demands, is used to train the sub-ANN predictive models. Due to the involvement of IoT sensors, the overall prediction accuracy can be improved. It is found that the mean absolute percentage error of energy demands prediction is 3% and 8% in training and testing cases, respectively.     

WSN中利用XGBoost和加权自适应HFLMS的数据约减组合预测方法

针对无线传感器网络(WSN)中能量、带宽和内存等各种资源的限制问题,提出了一种XGBoost结合加权自适应分层分数最小均方误差(hierarchical fractional least-mean-square,HFLMS)的数据约减组合预测方法。首先,利用XGBoost方法对损失函数进行了二阶的泰勒展开,权衡模型的复杂度和损失函数的下降速度,实现了资源限制的稳定预测;然后提出自适应HFLMS滤波器实现WSN数据约简的传输,并基于误差估计来预测所感测的数据,有效降低了WSN中的能量约束;最后,利用两个评估参数(能量和预测误差)来验证所提组合预测方法的性能。实验结果表明,相比没有预测、近似最速下降算法和分层最小均方滤波技术,提出的预测方法获得的预测结果更好。     

High performance concrete compressive strength forecasting using ensemble models based on discrete wavelet transform

This paper investigates the use of wavelet ensemble models for high performance concrete (HPC) compressive strength forecasting. More specifically, we incorporate bagging and gradient boosting methods in building artificial neural networks (ANN) ensembles (bagged artificial neural networks (BANN) and gradient boosted artificial neural networks (GBANN)), first. Coefficient of determination (R²), mean absolute error (MAE) and the root mean squared error (RMSE) statics are used for performance evaluation of proposed predictive models. Empirical results show that ensemble models (R²_BANN=0.9278, R²_GBANN=0.9270) are superior to a conventional ANN model (R²_ANN=0.9088). Then, we use the coupling of discrete wavelet transform (DWT) and ANN ensembles for enhancing the prediction accuracy. The study concludes that DWT is an effective tool for increasing the accuracy of the ANN ensembles (R²_WBANN=0.9397, R²_WGBANN=0.9528).     

Linear genetic programming for shear strength prediction of reinforced concrete beams without stirrups

A new design equation is proposed for the prediction of shear strength of reinforced concrete (RC) beams without stirrups using an innovative linear genetic programming methodology. The shear strength was formulated in terms of several effective parameters such as shear span to depth ratio, concrete cylinder strength at date of testing, amount of longitudinal reinforcement, lever arm, and maximum specified size of coarse aggregate. A comprehensive database containing 1938 experimental test results for the RC beams was gathered from the literature to develop the model. The performance and validity of the model were further tested using several criteria. An efficient strategy was considered to guarantee the generalization of the proposed design equation. For more verification, sensitivity and parametric analysis were conducted. The results indicate that the derived model is an effective tool for the estimation of the shear capacity of members without stirrups (R = 0.921). The prediction performance of the proposed model was found to be better than that of several existing buildings codes.     

ORESP：基于有序回归的软件缺陷严重程度预测方法

为提高软件缺陷严重程度的预测性能,通过充分考虑软件缺陷严重程度标签间的次序性,提出一种基于有序回归的软件缺陷严重程度预测方法ORESP.该方法首先使用基于Spearman的特征选择方法来识别并移除数据集内的冗余特征,随后使用基于比例优势模型的神经网络来构建预测模型.通过与五种经典分类方法的比较,所提的ORESP方法在四种不同类型的度量下均可取得更高的预测性能,其中基于平均0-1误差(MZE)评测指标,预测模型性能最大可提升10.3％;基于平均绝对误差(MAE)评测指标,预测模型性能最大可提升12.3％.除此之外,发现使用基于Spearman的特征选择方法可以有效提升ORESP方法的预测性能.     

基于时间点过程的时间序列预测模型

对城市中发生的事件进行有效预测,可以为政府避免、控制或减轻相关的社会风险提供决策支撑。首先,提出基于积分求导法的条件强度函数式,提高序列预测精度;其次,构建基于递归神经网络和累积危险函数的时间点过程模型,通过递归神经网络捕获历史事件的非线性依赖关系,利用全连接网络获得累积危险函数;最后,选择具有代表性的合成数据集和真实数据集对几种模型的性能进行对比分析。实验结果表明,所提模型可以更好地进行城市事件的时间序列预测,在平均绝对误差、平均负对数似然值等方面均优于传统的时间点过程模型,说明了模型的优越性。     

A low energy adaptive motion estimation hardware for H.264 multiview video coding

Multiview video coding (MVC) is the process of efficiently compressing stereo (two views) or multiview video signals. The improved compression efficiency achieved by H.264 MVC comes with a significant increase in computational complexity. Temporal prediction and inter-view prediction are the most computationally intensive parts of H.264 MVC. Therefore, in this paper, we propose novel techniques for reducing the amount of computations performed by temporal and inter-view predictions in H.264 MVC. The proposed techniques reduce the amount of computations performed by temporal and inter-view predictions significantly with very small PSNR loss and bit rate increase. We also propose a low energy adaptive H.264 MVC motion estimation hardware for implementing the temporal and inter-view predictions including the proposed computation reduction techniques. The proposed hardware is implemented in Verilog HDL and mapped to a Xilinx Virtex-6 FPGA. The FPGA implementation is capable of processing 30 × 8 = 240 frames per second (fps) of CIF (352 × 288) size eight view video sequence or 30 × 2 = 60 fps of VGA (640 × 480) size stereo (two views) video sequence. The proposed techniques reduce the energy consumption of this hardware significantly.     

Hybrid model combining empirical mode decomposition,singular spectrum analysis,and least squares for satellite-derived sea-level anomaly prediction

In this study, to meet the need for the accurate prediction of sea level anomaly (SLA), a hybrid model is proposed. In this model, empirical mode decomposition is combined with singular spectrum analysis and least-squares extrapolation to predict satellite-derived SLA. Each intrinsic mode function series of an empirical mode decomposition is decomposed and reconstructed using singular spectrum analysis. The reconstructed components and the residual series are predicted using least-squares extrapolation. This hybrid model was used for satellite-derived SLAs that were obtained using multi-mission along-track satellite altimetry data from September 1992 to January 2018, and the prediction errors for 3 years lead times were analysed. The observations and predictions of the principal components for annual or interannual periods correlated well, and the proposed hybrid model effectively predicted the SLAs. For the 3 years lead time predictions, the mean absolute error and root-mean-square error were 1.03 and 1.32 cm, respectively, which were less than those reported for existing methods.     

A novel bankruptcy prediction model based on an adaptive fuzzy k-nearest neighbor method

Bankruptcy prediction is one of the most important issues in financial decision-making. Constructing effective corporate bankruptcy prediction models in time is essential to make companies or banks prevent bankruptcy. This study proposes a novel bankruptcy prediction model based on an adaptive fuzzy k-nearest neighbor (FKNN) method, where the neighborhood size k and the fuzzy strength parameter m are adaptively specified by the continuous particle swarm optimization (PSO) approach. In addition to performing the parameter optimization for FKNN, PSO is also utilized to choose the most discriminative subset of features for prediction. Adaptive control parameters including time-varying acceleration coefficients (TVAC) and time-varying inertia weight (TVIW) are employed to efficiently control the local and global search ability of PSO algorithm. Moreover, both the continuous and binary PSO are implemented in parallel on a multi-core platform. The proposed bankruptcy prediction model, named PTVPSO-FKNN, is compared with five other state-of-the-art classifiers on two real-life cases. The obtained results clearly confirm the superiority of the proposed model in terms of classification accuracy, Type I error, Type II error and area under the receiver operating characteristic curve (AUC) criterion. The proposed model also demonstrates its ability to identify the most discriminative financial ratios. Additionally, the proposed model has reduced a large amount of computational time owing to its parallel implementation. Promisingly, PTVPSO-FKNN might serve as a new candidate of powerful early warning systems for bankruptcy prediction with excellent performance.     

桥梁钢-混凝土组合结构抗震性能预测仿真

为了及时预测桥梁主体结构的抗震性能,为抗震规划以及桥梁结构维修加固提供参考依据,针对当前桥梁钢-混凝土组合结构抗震性能预测方法中存在的过程复杂、耗时长,以及数据测量误差引起的准确率较低等问题,提出基于支持向量机回归法的桥梁钢-混凝土组合结构抗震性能预测方法。根据主体结构-荷载位移滞回实验的结果曲线,获取桥梁钢-混凝土组合结构恢复力矩,结合受力分析,计算组合结构的累积滞回耗能。根据累计滞回耗能值变化,结合支持向量机回归法构建钢-混凝土组合结构抗震性能预测模型;利用粒子群优化算法对模型进行优化,利用优化后模型完成桥梁钢-混凝土组合结构抗震性能预测。实验结果表明,与当前结构抗震性能预测方法相比,所提方法预测结果的准确率有一定提高,且预测耗时更短,满足抗震性能预测及时性需求。     

Estimating aboveground biomass of a mangrove plantation on the Northern coast of Vietnam using machine learning techniques with an integration of ALOS-2 PALSAR-2 and Sentinel-2A data

ABSTRACT

Aboveground biomass (AGB) of mangrove forest plays a crucial role in global carbon cycle by reducing greenhouse gas emissions and mitigating climate change impacts. Monitoring mangrove forests biomass accurately still remains challenging compared to other forest ecosystems. We investigated the usability of machine learning techniques for the estimation of AGB of mangrove plantation at a coastal area of Hai Phong city (Vietnam). The study employed a GIS database and support vector regression (SVR) to build and verify a model of AGB, drawing upon data from a survey in 25 sampling plots and an integration of Advanced Land Observing Satellite-2 Phased Array Type L-band Synthetic Aperture Radar-2 (ALOS-2 PALSAR-2) dual-polarization horizontal transmitting and horizontal receiving (HH) and horizontal transmitting and vertical receiving (HV) and Sentinel-2A multispectral data. The performance of the model was assessed using root mean square error (RMSE), mean absolute error (MAE), coefficient of determination (R²), and leave-one-out cross-validation. Usability of the SVR model was assessed by comparing with four state-of-the-art machine learning techniques, i.e. radial basis function neural networks, multi-layer perceptron neural networks, Gaussian process, and random forest. The SVR model shows a satisfactory result (R² = 0.596, RMSE = 0.187, MAE = 0.123) and outperforms the four machine learning models. The SVR model-estimated AGB ranged between 36.22 and 230.14 Mg ha^?1 (average = 87.67 Mg ha^?1). We conclude that an integration of ALOS-2 PALSAR-2 and Sentinel-2A data used with SVR model can improve the AGB accuracy estimation of mangrove plantations in tropical areas.     

Comparison of individual and combined ANN models for prediction of air and dew point temperature

Predicted air and dew point temperatures can be valuable in decision making in many areas including protecting crops from damage, avoiding heat stress on animals and humans, and in planning related to energy management. Current web-based artificial neural network (ANN) models on the Automated Environment Monitoring Network (AEMN) in Georgia predict hourly air and dew point temperature for twelve prediction horizons, using 24 models. The observed air temperature may approach the observed dew point temperature, but never goes below it. Current web based ANN models have prediction errors which, when the air and dew point temperatures are close, may cause air temperature to be predicted below the dew point temperature. Herein this error is referred to as a prediction anomaly. The goal of this research was to improve the prediction accuracy of existing air and dew point temperature ANN models by combining the two weather variables into a single ANN model for each prediction horizon. The objectives of this study were to reduce the mean absolute error (MAE) of prediction and to reduce the number of prediction anomalies. The combined models produced a reduction in the air temperature MAE for ten of twelve prediction horizons with an average reduction in MAE of 1.93 %. The combined models produced a reduction in the dew point temperature MAE for only six of twelve prediction horizons with essentially no average decrease in MAE. However, the combined models showed a marked reduction in prediction anomalies for all twelve prediction horizons with an average reduction of 34.1 %. The reduction in prediction anomalies ranged from 4.6 % at the one-hour horizon to 60.5 % at the eleven-hour horizon.