Transformer-based correction scheme for short-term bus load prediction in holidays

To tackle the problem of inaccurate short-term bus load prediction,especially during holidays,a Transformer-based scheme with tailored architectural enhancements is proposed.First,the input data are clustered to reduce complexity and capture inherent characteristics more effectively.Gated residual connections are then employed to selectively propagate salient features across layers,while an attention mechanism focuses on identifying prominent patterns in multivariate time-series data.Ultimately,a pre-trained structure is incorporated to reduce computational complexity.Experimental results based on extensive data show that the proposed scheme achieves improved prediction accuracy over comparative algorithms by at least 32.00%consistently across all buses evaluated,and the fitting effect of holiday load curves is outstanding.Meanwhile,the pre-trained structure drastically reduces the training time of the proposed algorithm by more than 65.75%.The proposed scheme can efficiently predict bus load results while enhancing robustness for holiday predictions,making it better adapted to real-world prediction scenarios.

Load prediction of grid computing resources based on ARSVR method

Based on the monitoring and discovery service 4 （MDS4） model, a monitoring model for a data grid which supports reliable storage and intrusion tolerance is designed. The load characteristics and indicators of computing resources in the monitoring model are analyzed. Then, a time-series autoregressive prediction model is devised. And an autoregressive support vector regression（ ARSVR） monitoring method is put forward to predict the node load of the data grid. Finally, a model for historical observations sequences is set up using the autoregressive （AR） model and the model order is determined. The support vector regression（SVR） model is trained using historical data and the regression function is obtained. Simulation results show that the ARSVR method can effectively predict the node load.

Server load prediction algorithm based on CM-MC for cloud systems

Accurate prediction of server load is important to cloud systems for improving the resource utilization, reducing the energy consumption and guaranteeing the quality of service(QoS).This paper analyzes the features of cloud server load and the advantages and disadvantages of typical server load prediction algorithms, integrates the cloud model(CM) and the Markov chain(MC) together to realize a new CM-MC algorithm, and then proposes a new server load prediction algorithm based on CM-MC for cloud systems. The algorithm utilizes the historical data sample training method of the cloud model, and utilizes the Markov prediction theory to obtain the membership degree vector, based on which the weighted sum of the predicted values is used for the cloud model. The experiments show that the proposed prediction algorithm has higher prediction accuracy than other typical server load prediction algorithms, especially if the data has significant volatility. The proposed server load prediction algorithm based on CM-MC is suitable for cloud systems, and can help to reduce the energy consumption of cloud data centers.

Novel Hybrid Physics‑Informed Deep Neural Network for Dynamic Load Prediction of Electric Cable Shovel

Electric cable shovel(ECS)is a complex production equipment,which is widely utilized in open-pit mines.Rational valuations of load is the foundation for the development of intelligent or unmanned ECS,since it directly influences the planning of digging trajectories and energy consumption.Load prediction of ECS mainly consists of two types of methods:physics-based modeling and data-driven methods.The former approach is based on known physical laws,usually,it is necessarily approximations of reality due to incomplete knowledge of certain processes,which introduces bias.The latter captures features/patterns from data in an end-to-end manner without dwelling on domain expertise but requires a large amount of accurately labeled data to achieve generalization,which introduces variance.In addition,some parts of load are non-observable and latent,which cannot be measured from actual system sensing,so they can’t be predicted by data-driven methods.Herein,an innovative hybrid physics-informed deep neural network(HPINN)architecture,which combines physics-based models and data-driven methods to predict dynamic load of ECS,is presented.In the proposed framework,some parts of the theoretical model are incorporated,while capturing the difficult-to-model part by training a highly expressive approximator with data.Prior physics knowledge,such as Lagrangian mechanics and the conservation of energy,is considered extra constraints,and embedded in the overall loss function to enforce model training in a feasible solution space.The satisfactory performance of the proposed framework is verified through both synthetic and actual measurement dataset.

Resource Load Prediction of Internet of Vehicles Mobile Cloud Computing

Load-time series data in mobile cloud computing of Internet of Vehicles(IoV)usually have linear and nonlinear composite characteristics.In order to accurately describe the dynamic change trend of such loads,this study designs a load prediction method by using the resource scheduling model for mobile cloud computing of IoV.Firstly,a chaotic analysis algorithm is implemented to process the load-time series,while some learning samples of load prediction are constructed.Secondly,a support vector machine(SVM)is used to establish a load prediction model,and an improved artificial bee colony(IABC)function is designed to enhance the learning ability of the SVM.Finally,a CloudSim simulation platform is created to select the perminute CPU load history data in the mobile cloud computing system,which is composed of 50 vehicles as the data set;and a comparison experiment is conducted by using a grey model,a back propagation neural network,a radial basis function(RBF)neural network and a RBF kernel function of SVM.As shown in the experimental results,the prediction accuracy of the method proposed in this study is significantly higher than other models,with a significantly reduced real-time prediction error for resource loading in mobile cloud environments.Compared with single-prediction models,the prediction method proposed can build up multidimensional time series in capturing complex load time series,fit and describe the load change trends,approximate the load time variability more precisely,and deliver strong generalization ability to load prediction models for mobile cloud computing resources.

A Power Load Prediction by LSTM Model Based on the Double Attention Mechanism for Hospital Building

This work proposed a LSTM(long short-term memory)model based on the double attention mechanism for power load prediction,to further improve the energy-saving potential and accurately control the distribution of power load into each department of the hospital.Firstly,the key influencing factors of the power loads were screened based on the grey relational degree analysis.Secondly,in view of the characteristics of the power loads affected by various factors and time series changes,the feature attention mechanism and sequential attention mechanism were introduced on the basis of LSTM network.The former was used to analyze the relationship between the historical information and input variables autonomously to extract important features,and the latter was used to select the historical information at critical moments of LSTM network to improve the stability of long-term prediction effects.In the end,the experimental results from the power loads of Shanxi Eye Hospital show that the LSTM model based on the double attention mechanism has the higher forecasting accuracy and stability than the conventional LSTM,CNN-LSTM and attention-LSTM models.

Two-Stage IoT Computational Task Offloading Decision-Making in MEC with Request Holding and Dynamic Eviction

The rapid development of Internet of Things(IoT)technology has led to a significant increase in the computational task load of Terminal Devices(TDs).TDs reduce response latency and energy consumption with the support of task-offloading in Multi-access Edge Computing(MEC).However,existing task-offloading optimization methods typically assume that MEC’s computing resources are unlimited,and there is a lack of research on the optimization of task-offloading when MEC resources are exhausted.In addition,existing solutions only decide whether to accept the offloaded task request based on the single decision result of the current time slot,but lack support for multiple retry in subsequent time slots.It is resulting in TD missing potential offloading opportunities in the future.To fill this gap,we propose a Two-Stage Offloading Decision-making Framework(TSODF)with request holding and dynamic eviction.Long Short-Term Memory(LSTM)-based task-offloading request prediction and MEC resource release estimation are integrated to infer the probability of a request being accepted in the subsequent time slot.The framework learns optimized decision-making experiences continuously to increase the success rate of task offloading based on deep learning technology.Simulation results show that TSODF reduces total TD’s energy consumption and delay for task execution and improves task offloading rate and system resource utilization compared to the benchmark method.

Short-term Load Prediction of Integrated Energy System with Wavelet Neural Network Model Based on Improved Particle Swarm Optimization and Chaos Optimization Algorithm

To improve energy efficiency and protect the environment,the integrated energy system(IES)becomes a significant direction of energy structure adjustment.This paper innovatively proposes a wavelet neural network(WNN)model optimized by the improved particle swarm optimization(IPSO)and chaos optimization algorithm(COA)for short-term load prediction of IES.The proposed model overcomes the disadvantages of the slow convergence and the tendency to fall into the local optimum in traditional WNN models.First,the Pearson correlation coefficient is employed to select the key influencing factors of load prediction.Then,the traditional particle swarm optimization(PSO)is improved by the dynamic particle inertia weight.To jump out of the local optimum,the COA is employed to search for individual optimal particles in IPSO.In the iteration,the parameters of WNN are continually optimized by IPSO-COA.Meanwhile,the feedback link is added to the proposed model,where the output error is adopted to modify the prediction results.Finally,the proposed model is employed for load prediction.The experimental simulation verifies that the proposed model significantly improves the prediction accuracy and operation efficiency compared with the artificial neural network(ANN),WNN,and PSO-WNN.

A mixture of HMM,GA,and Elman network for load prediction in cloud-oriented data centers

The rapid growth of computational power demand from scientific,business,and Web applications has led to the emergence of cloud-oriented data centers.These centers use pay-as-you-go execution environments that scale transparently to the user.Load prediction is a significant cost-optimal resource allocation and energy saving approach for a cloud computing environment.Traditional linear or nonlinear prediction models that forecast future load directly from historical information appear less effective.Load classification before prediction is necessary to improve prediction accuracy.In this paper,a novel approach is proposed to forecast the future load for cloud-oriented data centers.First,a hidden Markov model(HMM) based data clustering method is adopted to classify the cloud load.The Bayesian information criterion and Akaike information criterion are employed to automatically determine the optimal HMM model size and cluster numbers.Trained HMMs are then used to identify the most appropriate cluster that possesses the maximum likelihood for current load.With the data from this cluster,a genetic algorithm optimized Elman network is used to forecast future load.Experimental results show that our algorithm outperforms other approaches reported in previous works.

Prediction of Rolling Load in Hot Strip Mill byInnovations Feedback Neural Networks

Because the structure of the classical mathematical model of rolling load is simple, even with the self-adapting technology, it is difficult to accommodate the increasing dimensional accuracy. Motivated by this fact, an Innovations Feedback Neural Networks （IFNN） was presented based on the idea of Kalman prediction. The neural networks used the Back Propagation （BP） algorithm and applied it to the prediction of rolling load in hot strip mill. The theoretical results and the off-line simulation show that the prediction capability of IFNN is better than that of normal BP networks, namely, for the prediction of the rolling load in hot strip mill, the prediction precision of IFNN is higher than that of normal BP networks. Finally, a relative complete rolling load prediction system was developed on Windows 2003/XP platform using the OOP programming method and the SQL server2000 database. With this sys- tem, the rolling load of a 1700 strip mill was calculated, and the prediction results obtained correspond well with the field data. It shows that IFNN is valid for rolling load prediction.

Room thermal load prediction based on analytic hierarchy process and back-propagation neural networks

Accurate prediction of the heat load is the basic premise of intelligent regulation of the heating system,which helps to realize effective management of heating,ventilation,air conditioning system.For the problem that the weight of each influencing factor is not taken into account in the current heat load prediction and is not highly targeted,this article deeply explores the influence of different factors on the room heat load,and we propose a method to calculate room heat load prediction based on the combination of analytic hierarchy process(AHP)and back-propagation(BP)neural network.Firstly,eight environmental factors affecting the heat load are selected as prediction inputs through parametric analysis,and then the weights of each input are determined by AHP and normalize the prediction data by combining expert opinions,and finally do one-to-one training the quantified score and the room heat load to predict the future heat load by BP neural network.The simulation tests show that the mean absolute relative error(MARE)of the proposed prediction method is 5.40%.This article also verifies the influence of different expert opinions on the stability of the model.The results show that the proposed method can guarantee higher prediction accuracy and stability.

Host load prediction in cloud based on classification methods

The host load prediction problem in cloud computing has also been received much attention. To solve this problem, we have to use the historical load data to predict the future load level. Accurate prediction methods are useful for host load balance and virtual machine migration. Although cloud is likely to grids at some extent, the length of tasks are much shorter and host loads change more frequently with higher noise. The above characteristics introduce challenges for host load prediction. In this paper, based on the proposed exponentially segmented pattern and the corresponding transformation, prediction problem is transformed into the traditional classification problem, This classification problem can be solved based on the traditional methods, and features are given for training the classification model. For achieving accurate prediction, a new feature periodical coefficient is introduced and some existed classification methods are implemented. Experiments on the real world dataset invalidate the efficiency of the new proposed feature, which is in the most effective combinations of features, it increases successful rate （SR） 1.33%-2.82% and decreases the mean square error （MSE） 1.37%-2.91%. And the results also show that support vector machine （SVM） method can achieve nearly the same performance as the Bayes methods and their performance is about 50% higher in successful rate and 17% better in the mean square error compared to the existed methods.

Strength Prediction of Cruciform Specimen Under Biaxial Loading

In order to achieve a better understanding of failure behavior of cruciform specimen under different biaxial loading conditions,a three-dimensional finite element model is established with solid and interface elements.Maximum stress criterion,two Hashin-type criteria and the new proposed criteria are used to predict the strength of plain woven textile composites when biaxial loading ratio equals 1.Compared with experimental data,only the new proposed criteria can reach reasonable results.The applicability of the new proposed criteria is also verified by predicting the tensile and compressive strength of cruciform specimen under different biaxial loading ratios.Moreover,the introduction of interface element makes it more intuitive to recognize delamination failure.The shape of the predicted delamination failure region in the interface layer is similar to that of the failure region in neighboring entity layers,but the area of delamination failure region is a little larger.

ACO-Inspired Load Balancing Strategy for Cloud-Based Data Centre with Predictive Machine Learning Approach

Virtual Machines are the core of cloud computing and are utilized toget the benefits of cloud computing. Other essential features include portability,recovery after failure, and, most importantly, creating the core mechanismfor load balancing. Several study results have been reported in enhancing loadbalancingsystems employing stochastic or biogenetic optimization methods.It examines the underlying issues with load balancing and the limitationsof present load balance genetic optimization approaches. They are criticizedfor using higher-order probability distributions, more complicated solutionsearch spaces, and adding factors to improve decision-making skills. Thus, thispaper explores the possibility of summarizing load characteristics. Second,this study offers an improved prediction technique for pheromone level predictionover other typical genetic optimization methods during load balancing.It also uses web-based third-party cloud service providers to test and validatethe principles provided in this study. It also reduces VM migrations, timecomplexity, and service level agreements compared to other parallel standardapproaches.

A method for predicting critical load evaluating adhesion of coatings in scratch testing

In this paper based on the experiment principle of evaluating adhesion property by scratch testing, the peeling mechanism of thin films is discussed by applying contact theory and surface physics theory. A mathematical model predicting the critical load is proposed for calculating critical load as determined byscratch testing. The factors for correctly evaluating adhesion of coatings according to the experimental data arediscussed.

Optimization of Load Balancing Algorithm for Virtual Machine Dynamic Migration under Mobile Cloud Computing

As one of the key technologies of cloud computing,the virtualization technology can virtualize all kinds of resources and integrate them into the unified planning of the cloud computing management platform.The migration of virtual machines is one of the important technologies of virtual machine applications.However,there are still many deficiencies in the implementation of load balancing by virtual machine dynamic migration in cloud computing.Traditional triggering strategy thresholds are mostly fixed.If there is an instantaneous peak,it will cause migration,which will cause a waste of resources.In order to solve this problem,based on improving the dynamic migration framework,this paper proposes node selection optimization algorithm and node load balancing strategy and designs a prediction module,which uses a one-time smooth prediction to avoid the shortcoming of peak load moment.The simulation experiments and conclusions analysis results show that the fusion algorithm has performance advantages obvious.

Surrogate modeling for long-term and high-resolution prediction of building thermal load with a metric-optimized KNN algorithm

During the pre-design stage of buildings,reliable long-term prediction of thermal loads is significant for cool-ing/heating system configuration and efficient operation.This paper proposes a surrogate modeling method to predict all-year hourly cooling/heating loads in high resolution for retail,hotel,and office buildings.16384 surrogate models are simulated in EnergyPlus to generate the load database,which contains 7 crucial building features as inputs and hourly loads as outputs.K-nearest-neighbors(KNN)is chosen as the data-driven algorithm to approximate the surrogates for load prediction.With test samples from the database,performances of five different spatial metrics for KNN are evaluated and optimized.Results show that the Manhattan distance is the optimal metric with the highest efficient hour rates of 93.57%and 97.14%for cooling and heating loads in office buildings.The method is verified by predicting the thermal loads of a given district in Shanghai,China.The mean absolute percentage errors(MAPE)are 5.26%and 6.88%for cooling/heating loads,respectively,and 5.63%for the annual thermal loads.The proposed surrogate modeling method meets the precision requirement of engineering in the building pre-design stage and achieves the fast prediction of all-year hourly thermal loads at the district level.As a data-driven approximation,it does not require as much detailed building information as the commonly used physics-based methods.And by pre-simulation of sufficient prototypical models,the method overcomes the gaps of data missing in current data-driven methods.

Future Vision on Artificial Intelligence Assisted Green Energy Efficiency Network

To meet the key performance requirement of the 5G network and the demand of the growing number of mobile subscribers,millions of base stations are being constructed.5G New Radio is designed to enable denser network deployments,which raises significant concerns about network energy consumption.Machine learning(ML),as a kind of artificial intelligence(AI)technologies,can enhance network optimization performance and energy efficiency.In this paper,we propose AI/ML-assisted energy-saving strategies to achieve optimal performance in terms of cell shutdown duration and energy efficiency.To realize network intelligence,we put forward the concept of intrinsic AI,which integrates AI into every aspect of wireless communication networks.

Prediction of bed load sediments using different artificial neural network models

Modeling and prediction of bed loads is an important but difficult issue in river engineering.The introduced empirical equations due to restricted applicability even in similar conditions provide different accuracies with each other and measured data.In this paper,three different artificial neural networks(ANNs)including multilayer percepterons,radial based function(RBF),and generalized feed forward neural network using five dominant parameters of bed load transport formulas for the Main Fork Red River in Idaho-USA were developed.The optimum models were found through 102 data sets of flow discharge,flow velocity,water surface slopes,flow depth,and mean grain size.The deficiency of empirical equations for this river by conducted comparison between measured and predicted values was approved where the ANN models presented more consistence and closer estimation to observed data.The coefficient of determination between measured and predicted values for empirical equations varied from 0.10 to 0.21 against the 0.93 to 0.98 in ANN models.The accuracy performance of all models was evaluated and interpreted using different statistical error criteria,analytical graphs and confusion matrixes.Although the ANN models predicted compatible outputs but the RBF with 79%correct classification rate corresponding to 0.191 nctwork error was outperform than others.

Theoretical model for loads prediction on shield tunneling machine with consideration of soil-rock interbedded ground

The loads acting on shield tunneling machines are basic parameters for the equipment design as well as key control parameters throughout the entire operation of the equipment. In the study, a mechanical analysis for the coupled interactive system between the cutterhead and the ground at the excavation face is conducted. The normal and tangential loads acting on the cutterhead are decoupled and solved, with consideration of the influence of three key factors on loads: geological condition, operating status and equipment structure. Then analytical expressions for the thrust and the torque acting on the equipment under uniform geological condition are established. On this basis, the impact of soil-rock interbedded ground on acting loads is further considered. A theoretical model for loads prediction of earth pressure balance (EPB) shield machines working under soil-rock interbedded ground is proposed. This model is subsequently applied to loads prediction for a shield tunneling project under soil-rock interbedded ground. The computational value of the thrust and the torque, the measured loads and the load ranges from Krause empirical formula are compared. Thus, this model for loads prediction acting on shield tunneling machines under soil-rock interbedded ground has been proved to be effective.