Search

With the growing demand for various IoT applications, fog nodes frequently become overloaded. Fog computing requires effective load balancing to maximize resource utilization. It is essential to determine the load on host to obtain workload consolidation. Various random parameters, including CPU utilization, the number of CPU cores, RAM, memory allocated, memory available, disk I/O, and network I/O are employed to better comprehend host workloads. In the proposed work, the host's load status is detected using an ensemble approach into three categories: under-loaded, balanced and overloaded. Further, the proposed work considers three case studies and varying numbers of virtual machines (VMs) are executed with various parameter combinations. In each case study, a different number of VMs are executed in parallel on two different platforms. In the proposed study, we predicted the load on multiple hosts by employing a variety of advanced machine-learning models. To construct an ensemble model, we selected models with higher accuracy based on retrieved performance evaluation criteria. The ensemble method is applied to deal with the worst-case scenario of the model prediction. For a selected number of case studies, the Random Forest model, Ada Boost Classifier, Gradient Boost and Decision Tree models perform better than other models. These state-of-the-art predictive models are outperformed by our proposed ensemble model and achieves an improved accuracy of nearly 82% by correctly classifying hosts as overloaded, underloaded and balanced. [ABSTRACT FROM AUTHOR]

To address the environmental impact, low efficiency, and poor accuracy of existing power load prediction methods, this study innovatively proposes a power load prediction system that combines wavelet transform with digital twin technology. Compared with similar power load prediction methods, the proposed method achieved the highest power load prediction accuracy rate of 97.26%, with the lowest MAPE and RMSE being only 3.96% each. Our proposed method has good noise resistance and overcomes the disadvantage of traditional power load prediction methods that are easily affected by the environment. Moreover, the false detection rate of the load information data obtained from the power system in the Fuxin area from 2022 to 2023 was less than 5%, further verifying the reliability of the proposed method. This achievement is attributed to the powerful signal processing capabilities of the discrete wavelet transform, advanced pattern recognition and prediction capabilities of these three deep learning network algorithms, and the intelligence of digital twin technology. The combination of these three elements has brought new technological breakthroughs to the field of power load prediction. [ABSTRACT FROM AUTHOR]

Cloud Computing has immersed researchers in accessing the resources on-demand for deploying various applications by offering infinite services. But, as the demand for cloud resources is dynamic, it significantly affects the load on the system. Thus, this research emphasizes deploying a dynamic and autonomic load prediction framework. This paper proposes an Ensemble Time-Series Approach for Load Prediction (ETSA-LP), which integrates various time-series analysis techniques for predicting CPU and memory utilization. To evaluate the efficiency of the proposed approach, a series of experiments on Google and PlanetLab traces have been conducted in a real Cloud environment. The results were compared according to different performance metrics and models, the accuracy determined and the minimal error rate selected as the best among others. The proposed ensemble approach gives the best performance over the existing models showing the remarkable accuracy improvement and reducing the error rate and execution time. [ABSTRACT FROM AUTHOR]

Virtual machine (VM) load prediction is a critical task in cloud computing. Accurate VM load prediction can help to improve resource utilization, reduce costs, and improve the quality of service. For the Hybrid LSTM and AdaBoost model, a novel approach is proposed for accurate VM load prediction in cloud environments. The proposed model combines the power of LSTM and AdaBoost model, aiming to capture temporal dependencies in the VM load data and enhance prediction accuracy. The proposed model leverages LSTM to learn patterns and dynamics from historical load data, while AdaBoost is used to create an ensemble of weak regressors that collectively make load predictions. The model follows a two-step process: first, LSTM is trained on historical load data to extract informative features, and then AdaBoost is trained to combine the predictions from multiple weak regressors. The hybrid model demonstrates improved performance in VM load prediction by effectively handling non-linear relationships, temporal dependencies, and complex load patterns. The outcome of the proposed model is calculated using metrics, i.e. MAE, MAPE, MSE, RMSE, R2, and compared with existing machine learning algorithms i.e., Adaboost, KNN, SVM and deep learning algorithms i.e. LSTM, RNN. The results clearly show the superiority of the proposed hybrid approach in accurately predicting virtual machine load, enabling efficient resource allocation and management in cloud computing environments. [ABSTRACT FROM AUTHOR]

Most of today applications are built on a micro-service architecture, where a large application is divided into different functional parts that can be deployed on many containers that enable good load balancing. Container management tools need system load forecasting means to timely balance system load. It is an important problem for systems with direct streams of popular events which periodically have large splashes of a load. In this paper, we propose the load prediction method for such systems in two cases: usual and broadcasting workload. Also, we propose an architecture of adaptive infrastructure using our load forecasting method. [ABSTRACT FROM AUTHOR]

Abstract This manuscript introduced a domestic energy consumption modeling method which incorporated physical characteristics and behavioral factors. The model was established by considering regular load modeling and prediction characteristics as well as user behavior. A typical home energy hub serves as the research object; physical and operating features were input to model appliance energy use. We then simulated user energy consumption behaviors by combining non-intrusive load decomposition and Markov Chain methods based on the proposed model. Practical examples were used to determine the model's prediction and the effects of demand response. The proposed model showed an independent ability to depict the load portrait curve with relatively little input data. [ABSTRACT FROM AUTHOR]

• Operation optimization of phase change thermal energy storage heating system. • Novel control method combining load prediction and operation optimization. • The load prediction model integrating multiple deep learning algorithms. • The operation cost of the heating season can be saved by 10%. • The matching degree of heat supply and demand can be close to 100%. Due to the high latent heat and load-shifting capacity, phase-change thermal energy storage technology is an effective way to reduce energy costs under time-of-use electricity pricing. A favorable operation strategy is essential to exploit the advantage of the phase change thermal energy storage system. Previous studies on the operation strategy lack consideration of load prediction, which could reduce the matching degree of heat supply and demand. This study proposed a control method combing load prediction and operation optimization based on an electric boiler-phase change thermal energy storage heating system. A deep learning-based heating load prediction model was built; on this basis, an operation optimization method using dynamic programming was formulated subsequently. The proposed method was applied to a case building located in Beijing, China. By applying the proposed control method, the energy costs of the system were saved by 10% compared to the original operation strategy in the whole heating season, and the matching degree of heat supply and demand could be close to 100%, demonstrating that the method can provide users with cheaper and the best thermally comfortable heating. The proposed method is capable of achieving optimal operation of a phase change thermal energy storage system in practical applications. [Display omitted] [ABSTRACT FROM AUTHOR]

The recent advancements in sensing technologies allow us to record measurements from target structures at multiple locations and with relatively high spatial resolution. Such measurements can be used to develop data-driven methodologies for condition assessment, control, and health monitoring of target structures. One of the state-of-the-art technologies, Fiber Optic Strain Sensors (FOSS), is developed at NASA Armstrong Flight Research Center, and is based on Fiber Bragg Grating (FBG) sensors. These strain sensors are accurate, lightweight, and can provide almost continuous strainfield measurements along the length of the fiber. The strain measurements can then be used for real-time shape-sensing and operational load-estimation of complex structural systems. While several works have demonstrated the successful implementation of FOSS on large-scale real-life aerospace structures (i.e., airplane wings), there is paucity of studies in the literature that have investigated the potential of extending the application of FOSS into civil structures (e.g., tall buildings, bridges, etc.). This work assesses the feasibility of using FOSS to predict operational loads (e.g., wind loads) on chain-like structures. A thorough investigation is performed using analytical, computational, and experimental models of a 4-story steel building test specimen, developed at the University of Southern California. This study provides guidelines on the implementation of the FOSS technology on building-like structures, addresses the associated technical challenges, and suggests potential modifications to a load-estimation algorithm, to achieve a robust methodology for predicting operational loads using strain-field measurements. [ABSTRACT FROM AUTHOR]

Rural houses are recently built using confined masonry walls. The concrete beam and column confining the wall are usually considered to resist lateral loads, such as earthquake load. The Indonesian code requires reinforcement of 12 mm diameter in each corner of the beam and column with 8 mm diameter of stirrups. Such houses are usually termed of non-engineered structures, i.e buildings which the strength is not technically calculated but is established by the communities based on their experiences. This research was aimed to relate the in plane lateral ground acceleration and the resulting lateral displacement on the top of the wall at the first fundamental frequency. As the wall occasionally breaks at certain lateral drift ratio (H/V ratio), therefore, a prediction of the maximum lateral ground acceleration of the wall due to earthquake shake may also be established. Two wall specimens were made of local masonries with dimensions of 3m x 3m and 0.15 m thick. Lateral loads were applied in several stages and ambient vibrations of the two accelerometers, mounted on the base and on the top of the wall, were also recorded in stages. The maximum lateral force prediction, calculated using allowable lateral ground acceleration, was closely approaching the experiment result and numerical calculation. [ABSTRACT FROM AUTHOR]

Community building load prediction is the foundation and key to community building energy planning and community energy supply system design. The present load prediction methods for a single building are relatively mature. But load prediction for community buildings are more complicated than that for single buildings. In this paper, load prediction methods currently used for community buildings are first discussed categorically. Based on the previous work, prototypical building scenario load prediction method is proposed. Then taking Dawangdian community as an example, prototypical building scenario load prediction method is used to predict the cooling, heating and power loads. [ABSTRACT FROM AUTHOR]

Abstract: The density distribution and, thus, mechanical properties of long bones such as the femur are dependent on their loading. Many bone tissue adaptation theories are proposed to describe the density distribution that results from a given set of loading parameters. It is relatively easy to measure the density distribution of long bones, for example, using Computed Tomography (CT). However, there is no easy non-invasive method for in-vivo measurement of musculoskeletal loads. It is therefore interesting to investigate whether or not it is possible to predict the musculoskeletal loads that have resulted in a certain measured density distribution using bone tissue adaptation models. An inverse problem has to be solved for that purpose. In this paper, we use Artificial Neural Networks (ANNs) to solve the associated inverse problem and estimate the loading parameters that have resulted in the CT-measured three-dimensional density distribution of a proximal femur. An ANN is trained using a dataset generated by solving the forward tissue adaptation model for a large number of loading parameters. Before training the ANN with the generated training dataset, a Gaussian noise component is added to the density distribution. This improves the robustness of the trained ANN against deviations of the measured density distribution from the predictions of the forward bone tissue adaptation model. It is shown that the proposed technique is capable of predicting loading parameters that result in a density distribution close to the measured density distribution. [Copyright &y& Elsevier]

Acoustic emission (AE) nondestructive testing was used to monitor fiberglass/epoxy I-beams. The experiment consisted of loading the I-beams in cantilever fashion with a hydraulic ram. While testing, AE waveforms were collected from the onset of loading to failure. After acquisition, the AE data from each test were filtered to include only data collected up to 50 % of the theoretical ultimate load for further analysis. A Kohonen self-organizing map (SOM) was utilized to separate individual data points into failure mechanism clusters. Then a multiple linear regression analysis was performed using the percentage of hits associated with each failure mechanism along with the epoxy type to develop a prediction equation. The results of this analysis provided a prediction to within a 36.0 % error. A second analysis was performed utilizing a back-propagation neural network. The inputs to the network included a categorical variable for the epoxy type together with the amplitude frequencies from 30-100 dB. The optimized network contained two hidden layers having nine neurons apiece. Here the ultimate load prediction was within 48 lbf for a 9.5 % error. Thus, the back-propagation neural network produced far better results than the SOM/multiple linear regression, probably because of unwanted noise and perhaps nonlinearities in the data. [ABSTRACT FROM AUTHOR]

Aimed at the difficulty in predicting the city gas load due to many factors involved, the accumulation model in econometrics was adopted to predict the gas load. The city population and industrial production, which are two main factors that influence the gas load, were introduced into the model. The prediction result of the accumulation model was compared with that of the GM( 1 ,N) model in aspects of relative error, variance and Tale difference coefficient. It is indicated that the former is better than the latter, which proves that the proposed model is effective in predicting the gas load in a city with advantages of easy to be understood and easy for computer programming. [ABSTRACT FROM AUTHOR]

• Five prediction models (MLR, AR, ARX, MNR, and BP models) were developed for short term cooling load prediction. • The prediction accuracy on different training time scales (T-1 to T-6) were analyzed, and was evaluated by five evaluation indices. • Performance results showed that prediction accuracy of MNR was highest than MLR, AR, ARX and BP models when using short time training scale T-1. • For short time scale prediction, MNR model requires less training data, low hardware requirements and low computational complexity, which can be easily used to quickly on-site load prediction. Data-mining approaches for improving building cooling load predictions are presented and analyzed on different training time scales (T-1 to T-6) in this paper. Multiple linear regression (MLR), auto regression (AR), autoregressive exogenous (ARX) model, multiple nonlinear regression (MNR), and a back propagation neural network (BP). In this study, a real case in Guangzhou was examined to verify the performance of the prediction models on different time scales using measured cooling load and weather data. Based on the sensitivity analysis, the variable with largest sensitivity is selected as the input to the proposed MNR model, which can ensure prediction accuracy while increasing prediction speed. The case study shows that the prediction from the MNR model is comparable to or better than that from the BP model on different training time scales. The results also show that the MNR model could provide satisfactory prediction accuracy with less data (e.g., 3 d of weather and historical load data), while the BP model requires at least one month of training data to provide the same prediction accuracy. An analysis of the prediction performance results shows that the prediction accuracy from the regression models will decrease and the prediction accuracy from the BP model gradually increases as the training time scale increases. Surprisingly, the prediction accuracy with the MNR model is superior to that from the BP model with T-1 to T-4 training time scales. Finally, it is shown that the MNR model with the limited variable as inputs provides higher prediction accuracy than four other models with training time scales T-1 to T-4. The MNR model requires less training data, simpler hardware, and low computational complexity, making it preferable for quick on-site load prediction. [ABSTRACT FROM AUTHOR]