Power Consumption Estimation for Building Air Conditioning Systems Using Recurrent Neural Network

Prediction of power consumption and product size in cone crushing

Currently many commercial buildings are not continuously monitored for energy consumption, especially small buildings which constitute 90% of all such buildings. However, readily available data from the electric meters can be used for monitoring and analyzing energy consumption. Efficient utilization of available historical data (from these meters) can potentially improve energy efficiency, help to identify common energy wasting problems, and, in the future, enable various Smart Grid programs, such as demand response, real-time pricing etc. This paper describes application of computational intelligence techniques for prediction of electric power consumption. The proposed approach combines regression and clustering methods, in order to improve the prediction accuracy of power consumption, as a function of time (of the day) and temperature, using real-life data from several commercial and government buildings. Empirical comparisons show that the proposed approach provides an improvement over the currently used bin-based method for modeling power consumption.

Effective prediction of electric power consumption is vital for accurate energy management systems, particularly in growing urban regions. Conventional statistical techniques usually fail to capture the complicated patterns and non-linear tendencies in time-series energy data. The development of deep learning models has demonstrated encouraging findings in a variety of prediction tasks. Therefore, this paper concentrates on the application of diverse deep learning models, comprising One-dimensional Convolutional Neural Network (CNN), Long Short-Term Memory (LSTM), Bidirectional LSTM, and a hybrid model for power consumption prediction based on a dataset gathered from January 1, 2017, until January 1, 2018, in Tetouan, Morocco. These applied models were assessed utilizing diverse robust assessment metrics, and the findings demonstrated that the hybrid model (One-dimensional CNN with LSTM) reached superior performance with a Mean Squared Error (MSE) of 0.42548, Root-MSE of 0.65228, Mean Absolute Error (MAE) of 0.54395, and Median-AE of 0.51401, surpassing the other standalone and relevant models. The attained findings emphasize the merits of incorporating recurrent and convolutional structures to obtain more effective and accurate predictions of energy consumption time series

Household power load forecasting plays an important role in the operation and planning of power grids. To address the prediction issue of household power consumption in power grids, this paper chooses a time series of historical power consumption as the feature variables and uses landmark-based spectral clustering (LSC) and a deep learning model to cluster and predict the power consumption dataset, respectively. Firstly, the investigated data are reshaped into a matrix and all missing entries are recovered by matrix completion. Secondly, the data samples are divided into three clusters by the LSC method according to the periodicity and regularity of power consumption. Then, all samples in each cluster are expanded via bootstrap aggregating technique. Subsequently, a combination of a convolutional neural network (CNN) and a long short-term memory (LSTM) is employed to predict power consumption. The goal of CNN is to extract the features from input data in sequence learning, and LSTM aims to train and predict the power consumption. Finally, the forecasting performance of the LSC–CNN–LSTM is compared with several other deep learning models to verify its reliability and effectiveness in the field of household power load. The experimental results show that the proposed hybrid method is superior to other state-of-the-art deep learning techniques in forecasting performance.

Maintaining the balance of active power in the United Power System of Russia is carried out by the System Operator, which uses different lead intervals when planning electric power modes. In technologically isolated territories, maintaining the balance of active power is a special task, for the reason that the consumption of these territories is provided by their own generation facilities without using significant external reserves, so high accuracy of prediction of power consumption (electricity) in such power systems is important. Operational forecast of power consumption (electricity) allows realizing an optimal distribution of loads between generation facilities, taking into account the provision of the required reserves of generating power to maintain the frequency within acceptable limits. The field tests carried out on the isolated operation of the power system of the Kaliningrad region from the United Power System of Russia revealed imperfections in the existing method of operational power forecasting. The paper presents the results of studies of the structure of electricity consumption (power), daily graphs of power consumption and the rate of change in power consumption, as well as studies of the influence of weather factors on the amount of power consumption in the power system of the Kaliningrad region. A mathematical model of hourly consumption of the power system is developed on the basis of the conducted research, and the result of its practical application in System Operator is shown.

MuMMI (Multiple Metrics Modeling Infrastructure) environment is an infrastructure that facilitates systematic measurement, modeling, and prediction of performance, power consumption and performance-power tradeoffs for parallel systems. MuMMI builds upon three existing frameworks: Prophesy for performance modeling and prediction of parallel applications, PAPI for hardware performance counter monitoring, and PowerPack for power measurement and profiling. In this paper, we present the MuMMI framework, which consists of an Instrumentor, Databases and Analyzer. The MuMMI Instrumentor provides automatic performance and power data collection and storage with low overhead. The MuMMI Databases extend the databases of Prophesy to store power and energy consumption and hardware performance counters' data with different CPU frequency settings. The MuMMI Analyzer extends the data analysis component of Prophesy to support power consumption and hardware performance counters, and it entails performance and power modeling, performance-power tradeoff and optimizations, and web-based automated modeling system. Currently, our MuMMI online automated performance and power modeling system uses four modeling techniques: curve fitting, parameterization, kernel coupling and performance-counters-based, we discuss the effort to automate the process of developing performance and power models for scientific applications online, and focus on exploring performance-counters-based performance and power modeling. The MuMMI environment is able to aid in performance and power data measurement, storage, modeling and prediction of scientific applications on XSEDE resources in XSEDE community.

Reaming operation is a commonly used finishing phase for already drilled hole. Finishing is required because surface roughness of hole plays a significant role towards the functionality of the component. Surface roughness is a critical parameter for fatigue life of the component. Cutting forces are important indicator for power consumption required for cutting task. An artificial neural network (ANN) based surface roughness and power consumption model was established for Al 6061 under reaming operation. Back propagation neural networks were utilized for prediction of surface roughness and power consumption. Reaming test data was used to train and test the ANN network. In this presented study comparative investigation has been performed between the actual experimental values and neural network outputs to achieve good agreement.

The widespread use of the Unmanned Aerial Vehicles (UAVs) in many fields of industry such as defense, entertainment and transportation in recent years has led researchers to carry out many studies on UAV controls. One of the most popular of these study topics is the prediction of the instantaneous power consumption of UAVs during flight. In this study, a modified version of the Temporal Convolutional Network (TCN) architecture, which has recently been widely used for time series prediction, is used for prediction the UAVs' power consumption. As a result of the simulation processes performed using the flight data obtained from the DJI Matrice M100 drone, it was observed that the modified TCN model, which has an RMSE value of 0.0635, achieves better predictions compared to the recurrent neural networks and the classical TCN model.

Development and validation of intelligent load control for VRF air-conditioning system with deep learning based load forecasting

On the path to Exascale, the goal of High Performance Computing (HPC) to achieve maximum performance becomes the goal of achieving maximum performance under strict power constraint. Novel approaches to hardware and software co-design of modern HPC systems have to be developed to address such challenges. In this paper, we study prediction of power consumption of HPC systems using metrics obtained from hardware performance counters. We argue that this methodology is portable across different micro architecture implementations and compare results obtained on Intel 64, IBMR and Cavium ThunderXR ARMv8 microarchitectures.We discuss optimal number and type of hardware performance counters required to accurately predict power consumption.We compare accuracy of power predictions provided by models based on Linear Regression (LR) and Neural Networks (NN). We find that the NN-based model provides better accuracy of predictions than the LR model. We also find, that presently it is not yet possible to predict power consumption on a given microarchitecture using data obtained on a different microarchitecture. Results of our work can be used as a starting point for developing unified, cross-architectural models for predicting power consumption.

Skid-Steer Mobile Robots (SSMRs) provide a robust and simple mechanical drive platform making them useful in many applications. Power consumption is an important consideration in the design of any mobile robot and particularly important for SSMRs because of the slipping and corresponding friction that induce large loads on the drive system. The slipping behavior is generally characterized through Instantaneous Centers of Rotation (ICR) of the contact patches and it has been established that these are functions of the system dynamics. However, the existing SSMR power models generally treat these constraints at the kinematic level by assuming constant slip rates taken from empirical data. This paper demonstrates a method to evaluate SSMR power consumption based on slip parameters that are calculated as differential equations extracted from the equations of motion. The dynamic power model is validated and then implemented on two practical manufacturing applications in which a mobile robot is climbing on steel surfaces with primary power consumption due to turning and overcoming gravity. The applications show that the dynamic ICR model plays a significant role in estimating power requirements. The results further demonstrate that the power and energy requirements for a given task depend on the payload and motion along the task in a non-linear fashion, for example showing that the minimum payload does not necessarily correspond to the minimum energy use. This outcome suggests that dynamic effects can be used to find optimal trajectories to minimize power or energy requirements.

Finding the best energy-performance tradeoffs for High Performance Computing (HPC) applications is a major challenge for many modern supercomputing centers. With the increased focus on data center energy efficiency and the emergence of possible data center power constraints, making the right decision at a given time is becoming more important. A real-world situation like can a given 1000 compute node application be executed at a maximum of 2.7 GHz CPU frequency without going over the energy provider defined power band, or the available monthly energy limit? is just one example of the types of decisions HPC data centers will face. The previously developed Adaptive Energy and Power Consumption Prediction (AEPCP) model answers this question for the case of a fixed CPU frequency. This paper will extend the AEPCP process to enable the development of analytical models for estimating application execution time, power, and energy consumptions as functions of the number of compute nodes and maximum operating CPU frequency. Based on these analytical models, an adaptive model (Lightweight Adaptive Consumption Prediction (LACP)) is presented that implements the extended prediction process. This information allows for improved estimation of potential energy-performance costs and tradeoffs of applications and thus identifies the optimal resource configuration for specific data center boundary conditions.

Modern vending machines operate to serve the consumer easy-access demand for daily items such as food, drinks, or books. Purchasing processes are carried out via automation from item selection, payment, up to the item dispensing are done without interaction to the human and this can be carried out for 24 hours a day and 7 days a week. However, the vending machines operates for 24/7, consumes the same electrical load regardless of if there is a transaction. Despite the advancements of Industrial Revolution 4.0 (IR4.0) in many sectors, electrical power of vending machine is yet to be optimized. This research aims to optimize vending machine power consumption by using camera to detect human near to the vending machine and relays to control the power usage. Human can be detected via camera via Artificial Intelligence (AI) and this becomes the input to the Smart Relay. Abnormality such as electrical leakage that draws more power can be detected as well. IoE dashboard is developed for vending operator to monitor the power usage, leakage potential, optimization, and prediction of power consumption. The system saves about to 50 percent of electrical power usage in the vending machine and low energy wastage while maintaining the same vending machine capabilities.

Research was carried out within the framework of the state task of the Ministry of Science and Higher Education of the Russian Federation (topic Development of wavelet analysis models for non-stationary modes of electrical networks to improve the reliability and efficiency of power supply to consumers, topic code: FENG-2023-0005).
 Subject of research: predictive neural network model for a daily load schedule.
 Purpose of the study: prediction of power consumption based on the graph of electrical loads using a hybrid neural network.
 Object of research: methods for predicting the parameters of the power supply system.
 Main results of research: the results of forecasting the parameters of the power supply system based on daily load curves are presented. The simulation was carried out in the MATLAB software package. As a forecasting tool, a neural network was used, for the training of which load values averaged over half-hour time intervals and coefficients characterizing daily load schedules were used. The results obtained showed that the hybrid network gives a fairly accurate result, thereby confirming the adequacy of using a neural network to predict power or electricity consumption. The results of the study can be used for short-term forecasting and other parameters of the power supply system.

This paper describes the experimental cooling of a greenhouse in Madrid (Spain) using a radiant heated floor (RHF) coupled to an air-water heat pump (HP). Two cooling scenarios were studied over the summers of 2005 and 2006: natural ventilation + a shading screen (control system), and natural ventilation + a shading screen + an RHF (concrete) coupled to an air-water heat pump (i.e., in cooling mode; nominal power, 34.1 W m–2). Using the difference between the outside and inside air temperatures, it was concluded that at 0.5 m above the floor the RHF system reduced the temperature by 1.1°C in 2005 and 0.8°C in 2006. Both cooling scenarios were compared with other cooling technologies: the use of the natural ventilation + shading + RHF gave a smaller air temperature difference than fogging at a height equal to or lower than 0.5 m. A model based on the heat pump performance curves was constructed to predict its power consumption and good predictions of the variation over the day were obtained. The power consumption of the heat pump was 104.8 Wh m–2 d–1 (from 13:00 to 18:00 h) under our test conditions in Madrid. The RHF-HP system may only be appropriate for cooling greenhouses under certain circumstances, e.g., when growing high value crops or when cost is not a limiting factor, as its initial investment cost is about € 38 m–2.