On-line forecasting model for zinc output based on self-tuning support vector regression and its application

An on-line forecasting model based on self-tuning support vectors regression for zinc output was put forward to maximize zinc output by adjusting operational parameters in the process of imperial smelting furnace. In this model, the mathematical model of support vector regression was converted into the same format as support vector machine for classification. Then a simplified sequential minimal optimization for classification was applied to train the regression coefficient vector α- α* and threshold b. Sequentially penalty parameter C was tuned dynamically through forecasting result during the training process. Finally, an on-line forecasting algorithm for zinc output was proposed. The simulation result shows that in spite of a relatively small industrial data set, the effective error is less than 10% with a remarkable performance of real time. The model was applied to the optimization operation and fault diagnosis system for imperial smelting furnace.

STRONGLY CONVERGENT INERTIAL FORWARD-BACKWARD-FORWARD ALGORITHM WITHOUT ON-LINE RULE FOR VARIATIONAL INEQUALITIES

This paper studies a strongly convergent inertial forward-backward-forward algorithm for the variational inequality problem in Hilbert spaces.In our convergence analysis,we do not assume the on-line rule of the inertial parameters and the iterates,which have been assumed by several authors whenever a strongly convergent algorithm with an inertial extrapolation step is proposed for a variational inequality problem.Consequently,our proof arguments are different from what is obtainable in the relevant literature.Finally,we give numerical tests to confirm the theoretical analysis and show that our proposed algorithm is superior to related ones in the literature.

Medium-Term Electric Load Forecasting Using Multivariable Linear and Non-Linear Regression

Medium-term forecasting is an important category of electric load forecasting that covers a time span of up to one year ahead. It suits outage and maintenance planning, as well as load switching operation. We propose a new methodol-ogy that uses hourly daily loads to predict the next year hourly loads, and hence predict the peak loads expected to be reached in the next coming year. The technique is based on implementing multivariable regression on previous year's hourly loads. Three regression models are investigated in this research: the linear, the polynomial, and the exponential power. The proposed models are applied to real loads of the Jordanian power system. Results obtained using the pro-posed methods showed that their performance is close and they outperform results obtained using the widely used ex-ponential regression technique. Moreover, peak load prediction has about 90% accuracy using the proposed method-ology. The methods are generic and simple and can be implemented to hourly loads of any power system. No extra in-formation other than the hourly loads is required.

Study on the non-linear forecast method for water inrush from coal seam floor based on wavelet neural network

Directing at the non-linear dynamic characteristics of water inrush from coal seam floor and by the analysis of the shortages of current forecast methods for water inrush from coal seam floor, a new forecast method was raised based on wavelet neural network （WNN） that was a model combining wavelet function with artificial neural network. Firstly basic principle of WNN was described, then a forecast model for water inrush from coal seam floor based on WNN was established and analyzed, finally an example of forecasting the quantity of water inrush from coal floor was illustrated to verify the feasibility and superiority of this method. Conclusions show that the forecast result based on WNN is more precise and that using WNN model to forecast the quantity of water inrush from coal seam floor is feasible and practical.

Impact of ocean data assimilation on the seasonal forecast of the 2014/15 marine heatwave in the Northeast Pacific Ocean

A remarkable marine heatwave,known as the“Blob”,occurred in the Northeast Pacific Ocean from late 2013 to early 2016,which displayed strong warm anomalies extending from the surface to a depth of 300 m.This study employed two assimilation schemes based on the global Climate Forecast System of Nanjing University of Information Science(NUIST-CFS 1.0)to investigate the impact of ocean data assimilation on the seasonal prediction of this extreme marine heatwave.The sea surface temperature(SST)nudging scheme assimilates SST only,while the deterministic ensemble Kalman filter(EnKF)scheme assimilates observations from the surface to the deep ocean.The latter notably improves the forecasting skill for subsurface temperature anomalies,especially at the depth of 100-300 m(the lower layer),outperforming the SST nudging scheme.It excels in predicting both horizontal and vertical heat transport in the lower layer,contributing to improved forecasts of the lower-layer warming during the Blob.These improvements stem from the assimilation of subsurface observational data,which are important in predicting the upper-ocean conditions.The results suggest that assimilating ocean data with the EnKF scheme significantly enhances the accuracy in predicting subsurface temperature anomalies during the Blob and offers better understanding of its underlying mechanisms.

How Do Deep Learning Forecasting Models Perform for Surface Variables in the South China Sea Compared to Operational Oceanography Forecasting Systems?

It is fundamental and useful to investigate how deep learning forecasting models(DLMs)perform compared to operational oceanography forecast systems(OFSs).However,few studies have intercompared their performances using an identical reference.In this study,three physically reasonable DLMs are implemented for the forecasting of the sea surface temperature(SST),sea level anomaly(SLA),and sea surface velocity in the South China Sea.The DLMs are validated against both the testing dataset and the“OceanPredict”Class 4 dataset.Results show that the DLMs'RMSEs against the latter increase by 44%,245%,302%,and 109%for SST,SLA,current speed,and direction,respectively,compared to those against the former.Therefore,different references have significant influences on the validation,and it is necessary to use an identical and independent reference to intercompare the DLMs and OFSs.Against the Class 4 dataset,the DLMs present significantly better performance for SLA than the OFSs,and slightly better performances for other variables.The error patterns of the DLMs and OFSs show a high degree of similarity,which is reasonable from the viewpoint of predictability,facilitating further applications of the DLMs.For extreme events,the DLMs and OFSs both present large but similar forecast errors for SLA and current speed,while the DLMs are likely to give larger errors for SST and current direction.This study provides an evaluation of the forecast skills of commonly used DLMs and provides an example to objectively intercompare different DLMs.

AI-based Correction of Wave Forecasts Using the Transformer-enhanced UNet Model

Grid forecasting can be used to effectively enhance the spatial and temporal density of forecast products,thereby improving the capability of short-term marine disaster forecasting and warnings in terms of proximity.The traditional method that relies on forecasters'subjective correction of station observation data for forecasting has been unable to meet the practical needs of refined forecasting.To address this problem,this paper proposes a Transformer-enhanced UNet(TransUNet)model for wave forecast AI correction,which fuses wind and wave information.The Transformer structure is integrated into the encoder of the UNet model,and instead of using the traditional upsampling method,the dual-sampling module is employed in the decoder to enhance the feature extraction capability.This paper compares the TransUNet model with the traditional UNet model using wind speed forecast data,wave height forecast data,and significant wave height reanalysis data provided by ECMWF.The experimental results indicate that the TransUNet model yields smaller root-meansquare errors,mean errors,and standard deviations of the corrected results for the next 24-h forecasts than does the UNet model.Specifically,the root-mean-square error decreased by more than 21.55%compared to its precorrection value.According to the statistical analysis,87.81%of the corrected wave height errors for the next 24-h forecast were within±0.2m,with only 4.56%falling beyond±0.3 m.This model effectively limits the error range and enhances the ability to forecast wave heights.

TianXing:A Linear Complexity Transformer Model with Explicit Attention Decay for Global Weather Forecasting

In this paper,we introduce TianXing,a transformer-based data-driven model designed with physical augmentation for skillful and efficient global weather forecasting.Previous data-driven transformer models such as Pangu-Weather,FengWu,and FuXi have emerged as promising alternatives for numerical weather prediction in weather forecasting.However,these models have been characterized by their substantial computational resource consumption during training and limited incorporation of explicit physical guidance in their modeling frameworks.In contrast,TianXing applies a linear complexity mechanism that ensures proportional scalability with input data size while significantly diminishing GPU resource demands,with only a marginal compromise in accuracy.Furthermore,TianXing proposes an explicit attention decay mechanism in the linear attention derived from physical insights to enhance its forecasting skill.The mechanism can reweight attention based on Earth's spherical distances and learned sparse multivariate coupling relationships,promptingTianXing to prioritize dynamically relevant neighboring features.Finally,to enhance its performance in mediumrange forecasting,TianXing employs a stacked autoregressive forecast algorithm.Validation of the model's architecture is conducted using ERA5 reanalysis data at a 5.625°latitude-longitude resolution,while a high-resolution dataset at 0.25°is utilized for training the actual forecasting model.Notably,the TianXing exhibits excellent performance,particularly in the Z500(geopotential height)and T850(temperature)fields,surpassing previous data-driven models and operational fullresolution models such as NCEP GFS and ECMWF IFS,as evidenced by latitude-weighted RMSE and ACC metrics.Moreover,the TianXing has demonstrated remarkable capabilities in predicting extreme weather events,such as typhoons.

Fortified Financial Forecasting Models Based on Non-Linear Searching Approaches

The paper＇s aim is how to forecast data with variations involving at times series data to get the best forecasting model. When researchers are going to forecast data with variations involving at times series data （i.e., secular trends, cyclical variations, seasonal effects, and stochastic variations）, they believe the best forecasting model is the one which realistically considers the underlying causal factors in a situational relationship and therefore has the best ＂track records＂ in generating data. Paper＇s models can be adjusted for variations in related a time series which processes a great deal of randomness, to improve the accuracy of the financial forecasts. Because of Na＇fve forecasting models are based on an extrapolation of past values for future. These models may be adjusted for seasonal, secular, and cyclical trends in related data. When a data series processes a great deal of randomness, smoothing techniques, such as moving averages and exponential smoothing, may improve the accuracy of the financial forecasts. But neither Na＇fve models nor smoothing techniques are capable of identifying major future changes in the direction of a situational data series. Hereby, nonlinear techniques, like direct and sequential search approaches, overcome those shortcomings can be used. The methodology which we have used is based on inferential analysis. To build the models to identify the major future changes in the direction of a situational data series, a comparative model building is applied. Hereby, the paper suggests using some of the nonlinear techniques, like direct and sequential search approaches, to reduce the technical shortcomings. The final result of the paper is to manipulate, to prepare, and to integrate heuristic non-linear searching methods to serve calculating adjusted factors to produce the best forecast data.

Regional Storm Surge Forecast Method Based on a Neural Network and the Coupled ADCIRC-SWAN Model

Timely and accurate forecasting of storm surges can effectively prevent typhoon storm surges from causing large economic losses and casualties in coastal areas.At present,numerical model forecasting consumes too many resources and takes too long to compute,while neural network forecasting lacks regional data to train regional forecasting models.In this study,we used the DUAL wind model to build typhoon wind fields,and constructed a typhoon database of 75 processes in the northern South China Sea using the coupled Advanced Circulation-Simulating Waves Nearshore(ADCIRC-SWAN)model.Then,a neural network with a Res-U-Net structure was trained using the typhoon database to forecast the typhoon processes in the validation dataset,and an excellent storm surge forecasting effect was achieved in the Pearl River Estuary region.The storm surge forecasting effect of stronger typhoons was improved by adding a branch structure and transfer learning.

Multi-Sensor Intelligent System for On-Line and Real-Time Moneitoring Tool Cutting State in FMS

The principle and the constitution of an intelligent system for on-line and real-time montitoring tool cutting state were discussed and a synthetic sensors schedule combined a new type fluid acoustic emission sensor (AE) with motor current sensor was presented. The parallel communication between control system of machine tools, the monitoring intelligent system,and several decision-making systems for identifying tool cutting state was established It can auto - matically select the sensor way ,monitoring mode and identifying method in machining process- ing so as to build a successful and effective intelligent system for on -line and real-time moni- toring cutting tool states in FMS.

On-line血液透析滤过对移植肾功能延迟恢复的影响

目的 研究On -line血液透析滤过 (HDF)在移植肾功能延迟恢复患者中的应用。方法 我院2 0 0 2年 1月至 2 0 0 3年 7月肾移植术后发生移植肾功能延迟恢复患者 38例 ,2 9例采用Freseniuson -lineHDF透析机 ,F6 0透析器 ,透析滤过时间 3～ 4h ,血流量在 180～ 2 5 0ml/min ,置换液采取前稀释法 ,9例行HD治疗 ,采用GambroAK2 0 0透析机 ,F7透析器。结果 38例移植肾功能延迟恢复患者 ,2 9例用联机血液透析滤过 ,平均肾功能恢复时间为 ( 11.2± 5 .6 )天 ,治疗期间收缩压变化在 110～ 170mmHg,透析期间低血压发生率为 2 .6 % ,9例HD治疗者平均肾功能恢复时间为 ( 2 2 .3± 15 .7)天 ,透析期间低血压发生率为 13.2 %。结论 与HD比较 ,On -lineHDF对血压影响较小 ,可缩短移植肾功能恢复的时间。

Development and Application of On-Line Corrosion Monitoring Device for Condenser Tube

This paper introduces the development and industrial application of an on-line corrosion monitoring device for condenser tubes. Corrosion sensors are made up of representative condenser tubes chosen by eddy current test, which enable the monitoring result to be consistent with the corrosion of actual condenser tubes. Localized corrosion rate of condenser tubes can be measured indirectly by a galvanic couple made up of tube segments with and without pits. Using this technology, corrosion problems can be found in time and accurately, and anticorrosive measures be made more economic and effective. Applications in two power plants showed the corrosion measurements are fast and accurate.

平行机的实时到达on-line算法下界的改进

本文考虑了平行机实时到达的在线问题．模型中，工件是陆续到达的，工件的个数、到达时间是事先未知的，而且只有当工件到达，才知其加工时间，目标是使所有工件都加工完的时间达到最小．Ｃｈｅｎ与Ｖｅｓｔｊｅｎｓ（１９９６年）证明了该在线问题不存在性能比小于１．３４７３的ｏｎ－ｌｉｎｅ算法．本文将此界改进为（５－５）／２．

Effects of on-line solution and off-line heat treatment on microstructure and hardness of die-cast AZ91D alloy

The effects of on-line solution, off-line solution and aging heat treatment on the microstructure and hardness of the die-cast AZ91D alloys were investigated. Brinell hardness of die-cast AZ91D alloy increases through on-line solution and off-line aging treatment but decreases after off-line solution treatment. By X-ray diffractometry, optical microscopy, differential thermal analysis, scanning electron microscopy and X-ray energy dispersive spectroscopy, it is found that the microstructures of the die-cast AZ91D magnesium alloy before and after on-line solution and off-line aging are similar, consisting of α-Mg and β-Al12Mg17. The precipitation of Al element is prevented by on-line solution so that the effect of solid solution strengthening is enhanced. The β-Al12Mg17 phases precipitate from supersaturated Mg solid solution after off-line aging treatment, and lead to microstructure refinement of AZ91D alloy, so the effect of precipitation hardening is enhanced. The β-Al12Mg17 phases dissolve in the substructure after off-line solution treatment, which leads to that the grain boundary strengthening phase is reduced significantly and the hardness of die cast AZ91D is reduced.

On-line血液透析滤过对维持性血液透析患者心率变异性的影响

目的探讨On-line血液透析滤过技术对维持性血液透析患者心率变异性(HRV)的影响。方法 50例维持性血液透析(MHD)患者随机分为常规血液透析治疗组(HD组)和On-line血液透析滤过治疗组(OL-HDF组),每组各25例。患者每周行3次、每次4 h的血液透析治疗或OL-HDF,治疗24个月。比较两组常规的实验室检查和24h动态心电图监测结果。结果治疗24个月后,OL-HDF组的β2-微球蛋白(β2-MG)由基线值30.1±7.2 mg/dl降至21.7±6.1 mg/dl(P<0.05),HD组的β2-MG则无明显下降。OL-HDF组HRV的时域性和频域性指标均值高于基线值(P<0.05),HD组的HRV指标则无显著性变化(P>0.05)。结论与普通血液透析比较,OL-HDF能更好地改善MHD患者的HRV,改善自主神经功能。

On-line置换液与生理盐水在血液透析滤过回血中的应用效果比较

[目的]探讨On-line置换液与生理盐水在血液透析滤过(HDF)病人回血中的应用效果,以期为临床护理工作提供理论依据。[方法]选取30例规律HDF病人为研究对象,采用自身对照方法,每例病人每周依次采用On-line置换液和生理盐水回血,比较回血过程中残余血量、静脉壶液体温度、护士的工作量及经济效益。[结果]采用On-line置换液回血法残余血量及回血后静脉壶温度分别为(1.89±0.91)mL,(33.36±0.15)℃,而采用生理盐水法为(2.88±0.23)mL、(27.13±0.14)℃,差异均有统计学意义(P<0.01);而透析器凝血状况比较差异无统计学意义(P=0.08);采用On-line置换液回血时护士工作量减少,回血液体成本降低。[结论]应用接近生理温度的On-line置换液回血能稀释血液、降低血液黏稠度,减少血路管残余血量,增加病人舒适感,提高工作效率。

On-lineHDF的临床观察及护理

目的 探讨应用On -lineHDF治疗慢性肾衰尿毒症患者的临床效果及护理对策。方法 回顾性分析 15例维持性HD患者应用On -lineHDF的治疗及护理过程。结果 15例患者经每 1～ 2周 1次的On -lineHDF治疗后 ,病情稳定 ,部分患者心功能、血压、食欲、体力及睡眠不同成度得到改善 ,在促红细胞生成素使用量不变的情况下 ,4例病人的血红蛋白有所增加。结论 On -lineHDF是当今最佳的肾脏替代疗法技术之一。