Load Time Series Research Articles

IMPLEMENTATION OF PROPHET IN AMERICAN ELECTRICITY FORECASTING WITH AND WITHOUT PARAMETER TUNING

Prophet is one of the machine learning approximation methods that accommodate trends, seasonality, and holiday impacts in time series data. Generally, the performance of machine learning models can be improved by implementing hyperparameter tuning. This study investigates whether hyperparameter tuning can improve the model's performance. To show its effectiveness, the Prophet model constructed by parameter tuning is compared to the one with fixed parameter values (namely the default model) for both the original series and the Box-Cox transformed series in terms of mean absolute percentage error (MAPE). Based on the experimental results of the twenty-four daily electricity load time series in American Electric Power (AEP). This shows that parameter tuning successfully reduces the MAPE of the default model in the range of about 3-8% for training data. However, there is no guarantee for testing data. Although, in some cases, parameter tuning can reduce the MAPE value of the default model by up to 38%, in other cases, it actually increases the MAPE of the default model by almost 15%. The experiments on testing data also show that models built from transformed data do not necessarily produce more accurate forecast values than those built from the original data.

Reliability assessment of distribution networks with distributed generation considering different types of user loads

Abstract This paper meticulously evaluates the reliability of distribution networks integrated with distributed generation (DG), considering various user classifications. Initially, we delve into the daily load profiles of residential, commercial, and industrial sectors, establishing a nuanced time-series load model that encapsulates the distinct characteristics of these customer segments. Subsequently, we formulate a model for the harmonious operation of energy storage systems, tailored to the interplay between DG output and load requirements. Utilizing the sequential Monte Carlo simulation algorithm, we compute the reliability metrics for the IEEE RBTSBUS6 feeder system, revealing that DG integration significantly enhances distribution network reliability. Notably, identical DG output yields varied impacts on the reliability of distinct customer sectors, underscoring the need for tailored strategies in DG deployment.

Fatigue Damage and Reliability Assessment of Wind Turbine Structure During Service Utilizing Real-Time Monitoring Data

Under the action of wind load, a wind turbine tower will produce alternating stress, which leads to fatigue failure. According to the mean wind speed at the wind turbine impeller collected from the SCADA system, the mean wind speed of the simulation point is calculated by using the wind speed exponential model formula. Davenport spectra are used to simulate the pulsating wind speed time series. The wind spectrum is obtained using the harmonic superposition method. Subsequently, the wind speed time series and wind load time series at the simulation point are calculated. Structural modeling of a 5 MW wind turbine tower is performed in ABAQUS 2021. The modal shape and natural frequency are obtained by modal analysis to verify the rationality of the model. Subsequently, wind loads are applied to the model, and structural stress time history is obtained by transient modal dynamics analysis. The stress time history of the maximum stress area of the tower structure is extracted, and the rain flow counting method is applied to it to obtain the stress spectrum. The Weibull distribution of the stress spectrum is fitted, the mean and variance of the total damage in one day are calculated, and the fatigue reliability analysis of the maximum stress area of the tower structure is carried out. And the nonlinear fatigue cumulative damage analysis of the region is carried out. This work has implications for fatigue reliability studies for approximate operating conditions.

A Well-Tested Pump Card Classification and Identification Algorithm Based on Time Series Analysis Techniques

Abstract To accurately identify the operational conditions of oil pumping wells, promptly monitor the working status of oil wells, expedite responses to sudden pumping well accidents, and ensure orderly production at the oil extraction site, this paper introduces a time segment-based oil pumping well condition identification algorithm. This method no longer relies on drawn dynamometer cards for condition identification. Instead, it solely utilizes sensor-provided polished rod displacement and load time series, aligning more closely with the data’s intrinsic nature. The algorithm is sufficiently lightweight, highly responsive, and does not require GPU resources, making it suitable for edge deployment. Furthermore, this approach provides evidence of the time continuity and numerical continuity of time series data. Compared to traditional identification methods, this approach maintains a high level of accuracy, offers a streamlined model, and is interpretable, presenting a novel perspective for oil pumping well condition identification.

Deep learning based photovoltaic generation on time series load forecasting

Deep learning based photovoltaic generation on time series load forecasting

Wind turbine foundation ring fatigue reliability analysis under wind-load using SCADA system

Under the action of wind load, the foundation ring of the fan will generate stress concentration and alternating stress, leading to fatigue failure. Based on the average wind speed obtained from the supervisory control and data acquisition (SCADA) system, the orthogonal expansion method of random pulsating wind field and the number theory point selection method are used to calculate and simulate the corresponding pulsating wind speed time series, and then calculate the wind speed time series and wind load time series. Based on the ABAQUS finite element software, a model of a 5 MW wind turbine is established. The modal analysis is used to obtain the modal vibration pattern and the intrinsic frequency to verify the reasonableness of the structural modeling. Subsequently, the wind load time series is applied to the wind turbine model to obtain the stress time series using instantaneous modal dynamic analysis (Modal dynamics). The stress time series at the location of maximum stress concentration in the foundation ring is extracted, and the stress time series with the same stress amplitude is obtained through conversion, which has Wiener characteristics. After obtaining the stress amplitude using the rainflow counting method, the fatigue reliability is calculated based on the structural fatigue reliability analysis method considering the threshold crossing duration of the random process. This research holds reference significance for the calculation of fatigue reliability under approximate working conditions.

Allometric comparison of viral dynamics in the nasal cavity-nasopharyngeal mucus layer of human and rhesus monkey by CFD-HCD approach

Background and objectiveViral respiratory infections stand as a considerable global health concern, presenting significant risks to the health of both humans and animals. This study aims to conduct a preliminary analysis of the time series of viral load in the nasal cavity-nasopharynx (NC-NP) of the human and rhesus macaque (RM). MethodsTaking into account the random uniform distribution of virus-laden droplets with a diameter of 10 μm in the mucus layer, this study applies the computational fluid dynamics-host cell dynamics (CFD-HCD) method to 3D-shell NC-NP models of human and RM, analyzing the impact of initial distribution of droplets on the viral dynamics of severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2), estimating parameters in the HCD model based on experimental data, integrating them into simulations to predict the time series of viral load and cell counts, and being visualized. The reproductive number (R0) are calculated to determine the occurrence of infection. The study also considers cross-parameter combinations and cross-experimental datasets to explore potential correlations between the human and RM. ResultsThe research findings indicate that the uniform distribution of virus-laden droplets throughout the whole NC-NP models of human and RM is reasonable for simulating and predicting viral dynamics. The visualization results offer dynamic insights into virus infection over a period of 20 days. Studies involving parameter and dataset exchanges between the two species underscore certain similarities in predicting virus infections between the human and RM. ConclusionsThis study lays the groundwork for further exploration into the parallels and distinctions in respiratory virus dynamics between humans and RMs, thus aiding in making more informed decisions in research and experimentation.

Time-Series Prediction of Electricity Load for Charging Piles in a Region of China Based on Broad Learning System

This paper introduces a novel electricity load time-series prediction model, utilizing a broad learning system to tackle the challenge of low prediction accuracy caused by the unpredictable nature of electricity load sequences in a specific region of China. First, a correlation analysis with mutual information is utilized to identify the key factors affecting the electricity load. Second, variational mode decomposition is employed to obtain different mode information, and then a broad learning system is utilized to build a prediction model with different mode information. Finally, particle swarm optimization is used to fuse the prediction models under different modes. Simulation experiments using real data validate the efficiency of the proposed method, demonstrating that it offers higher accuracy compared to advanced modeling techniques and can assist in optimal electricity-load scheduling decision-making. Additionally, the R2 of the proposed model is 0.9831, the PRMSE is 21.8502, the PMAE is 17.0097, and the PMAPE is 2.6468.

Short-term Load Forecasting and Regime Switching Detection Based on Normalized Causal Entropy Boosting

Abstract This paper introduces a method based on normalized causal entropy boosting (NCEBoosting) to identify regime switches in load patterns and improve the adaptability of prediction models. Load time-series exhibit multidimensional, nonlinear changes influenced by factors such as human behavior and meteorological environments. To detect regime switches, we calculate the sparsified causal entropy matrix using a small batch of load flow data, which corrects the original prediction model. Subsequently, the corrected model is employed to predict the load requirements after the switch. By comparing the causal entropy across different time periods, we effectively detect conceptual drift, phase changes, and unexpected events in the data, enabling the determination of load mode switches. Given that load timing data is typically generated in batches, calculating the causal entropy of successive data batches provides a robust metric. In this paper, we determine the load mode change by calculating the integrated causal entropy of the batch data streams and calibrating the prediction model accordingly. Extensive computational experiments on real building load datasets are conducted to validate the proposed method. The results demonstrate that the normalized causal entropy boosting effectively detects load mode switches compared to traditional approaches. Furthermore, the method demonstrates efficient adaptive learning based on online data streams, resulting in improved adaptability and prediction accuracy.

Analysis of Carbon Emission Reduction with Using Low-Carbon Demand Response: Case Study of North China Power Grid

The power sector is the single industry with the largest carbon emission in China, the carbon emission of which accounts for more than 40% of China’s total carbon emissions. In relevant research on the simulation of power system operation, current studies focus more on energy conservation and economical operation, while few consider the low-carbon optimization of the power system from the perspective of carbon emissions. In addition, in relevant research on carbon reduction in the power system, current studies focus more on controlling the direct carbon emission of the source side and less on the indirect carbon emissions of the load side, which focus on the reverse effect of a user’s electricity consumption behavior on the carbon reduction goals of the power system. This article delved into a deterministic simulation model of power system operation based on time series load curves and proposed a carbon reduction mechanism called the low-carbon demand response mechanism, which guides users to actively respond and reduce the carbon emission of power systems. In addition, this article conducted an empirical analysis based on the planning data of the North China Power Grid. To minimize carbon emissions, a simulation of low-carbon optimization operation for the North China Power Grid in 2040 was carried out. Then, based on the simulation results, an analysis of the carbon reduction benefits of low-carbon demand response was carried out. Ultimately, the empirical analysis verified that low-carbon optimization operation and low-carbon demand response technology possess significant carbon reduction potential for the power system.

Data-driven forecasting of FOWT dynamics and load time series using lidar inflow measurements

This study focuses on forecasting the fairlead tension and floater dynamics time series of floating offshore wind turbines (FOWTs) using a data-driven approach that incorporates onboard sensor measurements and lidar inflow data. Sensors on FOWTs can provide data on turbine dynamics, such as rotational and translational movements, and load metrics like mooring line loads. However, these sensors are limited to current state measurements and do not provide future signal projections. In this research, we investigate a data-driven forecasting methodology using a simulated dataset. This dataset encompasses FOWT responses to diverse environmental conditions and the associated lidar measurements. Utilizing a Long Short-Term Memory (LSTM) sequence-to-sequence model, this study forecasts the time series of fairlead tension, surge, and pitch for forecasting horizons of 20, 40, and 60 seconds, considering lidar ranges from 100m to 500m. The performance of these forecasting models is benchmarked against a simple persistence model. The results indicate that incorporating lidar inflow measurements significantly improves the forecasts of fairlead tensions and platform motions. The enhancement for pitch motion forecasts is observed across all forecasting horizons. For fairlead tension and surge motion, the enhancement is observed for the longer horizons of 40s and 60s. These findings underscore the value of lidar data in accurate forecasting and emphasize the need to account for the interplay between lidar range, wind speed, and forecasting horizon to achieve optimal forecast accuracy.

Dynamics of the virtual center of wind pressure: An approach for the estimation of wind turbine loads

A new stochastic method for the reconstruction of time series of bending moments at the main shaft of a wind turbine is presented. The method is based on the characterization of the dynamics of the virtual center of wind pressure (CoWP). The CoWP, calculated purely from the incoming wind field, provides an approach for correlating the bending moments at the shaft with large-scale turbulent structures of the wind acting on the rotor plane. The proposed method allows a fast procedure for generating time series of loads on the shaft. As a result, the computational and time requirements for maximum and fatigue load assessment could be reduced.

Optimizing dimension selection for rain flow counting in fatigue assessment of large-scale lattice wind turbine support structures: A comprehensive study and design guidance

The selection of dimension (d) for the rain flow counting matrix significantly influences the reliable prediction of fatigue life in wind turbine support structures. However, there are limited public reports on the impact of dimensions on the fatigue assessment of wind turbine structures. This study explores the influence of dimensions on the fatigue assessment process and outcomes, focusing on a large-scale ultra-high lattice wind turbine support structure from an engineering project. Through integrated load simulation, fatigue load time series for different design load cases (DLCs) are obtained for the overall structure. The rain flow counting method is utilized to derive the overall rain flow counting matrix (Markov_n) with varying dimensions. Subsequently, employing a multi-scale fatigue assessment method, the damage matrix (Markov_D), cumulative fatigue damage (D), and fatigue life are computed. A detailed exploration examines the impact of rain flow counting matrix dimensions on Markov_n, Markov_D, D, fatigue life, and calculation time (t). Furthermore, guidance is provided for selecting the optimal dimension for engineering design. The findings indicate that selecting dimensions that are too small results in the loss and merging of smaller mean and amplitude load values, thereby failing to accurately represent the load history. Moreover, a small dimension yields an excessively cautious estimation of cumulative fatigue damage (D) and underestimates the fatigue life, consequently inflating construction expenses.

Emotion-Driven Energy Load Forecasting: An Ensemble Leveraging Insights from News

In modern times, system energy load forecasting is an extremely important process in a variety of contexts. Moreover, energy load time series fluctuations are influenced by a wide range of factors, ranging, inter alia, from environmental conditions, natural events, and demographics to both regional and global geopolitical contexts, economic conditions, energy sources, policies, and regulations. Given these, this paper examines the integration of news information from the global scene into Greek energy load forecasting schemes through the use of sentiment analysis. Investigating the ways the general emotional footprint of news worldwide affects and can be used in an energy modeling context, we benchmark possible ensemble configurations incorporating a multitude of 31 emotion polarities. Building on our previous work, an ensemble method that exploits specific outputs from a multi-label sentiment classifier and a sentiment analysis procedure under a multivariate forecasting scheme is presented. It is shown, through an empirical evaluation of the results, that the integration of emotion representations related to non-Greek news concerning global current affairs improves the predictions.

PCA-TANN with model-based transfer learning for predicting blast load time series on structures

A comprehensive understanding of blast load information is crucial for evaluating structural response. Machine learning has demonstrated its efficiency in predicting blast loads. However, the high costs associated with acquiring blast load data limit the development of machine learning. A novel method, PCA-TANN, is proposed for predicting blast load time series on structures and improving the problem of insufficient training data by integrating transfer learning concepts and combining artificial neural networks (ANN) with principal component analysis (PCA). The main idea of this model involves preprocessing the time series data using PCA to extract essential features, employing artificial neural networks based on transfer learning (TANN) for knowledge transfer from the previous task to the new task, and integrating a few preprocessed data of the new task for training. The blast load data obtained from a single square column is utilized to assess the performance of PCA-TANN and PCA-ANN models. The findings demonstrate that in situations with limited data availability, PCA-TANN reasonably replicates the blast load characteristic, resulting in accurate time series predictions of overpressure with R2 scores exceeding 0.85. This performance significantly surpasses that of the used PCA-ANN model, which achieves R2 scores below 0.4.

Using conditional Invertible Neural Networks to perform mid‐term peak load forecasting

AbstractMeasures for balancing the electrical grid, such as peak shaving, require accurate peak forecasts for lower aggregation levels of electrical loads. Thus, the Big Data Energy Analytics Laboratory (BigDEAL) challenge—organised by the BigDEAL—focused on forecasting three different daily peak characteristics in low aggregated load time series. In particular, participants of the challenge were asked to provide long‐term forecasts with horizons of up to 1 year in the qualification. The authors present the approach of the KIT‐IAI team from the Institute for Automation and Applied Informatics at the Karlsruhe Institute of Technology. The approach to the challenge is based on a hybrid generative model. In particular, the authors use a conditional Invertible Neural Network (cINN). The cINN gets the forecast of a sliding mean as representative of the trend, different weather features, and calendar information as conditioning input. By this, the proposed hybrid method achieved second place overall and won two out of three tracks of the BigDEAL challenge.

Short-term Electricity Load Forecasting Based on Improved Seagull Algorithm Optimized Gated Recurrent Unit Neural Network

INTRODUCTION: The complexity of the power network, changes in weather conditions, diverse geographical locations, and holiday activities comprehensively affect the normal operation of power loads. Power load changes have characteristics such as non stationarity, randomness, seasonality, and high volatility. Therefore, how to construct accurate short-term power load forecasting models has become the key to the normal operation and maintenance of power.OBJECTIVES: Accurate short-term power load forecasting helps to arrange power consumption planning, optimize power usage and largely reduce power system losses and operating costs.METHODS: A hybrid decomposition-optimization-integration load forecasting method is proposed to address the problems of low accuracy of current short-term power load forecasting methods.RESULTS: The original power load time series is decomposed using the complete ensemble empirical modal decomposition method, while the correlation of power load influencing factors is analysed using Pearson correlation coefficients. The seagull optimisation algorithm is overcome to fall into local optimality by using the random adaptive non-linear adjustment strategy of manipulated variables and the differential variational Levy flight strategy, which improves the search efficiency of the algorithm. Then, the The gated cyclic unit hidden layer parameters are optimised by the improved seagull optimisation algorithm to construct a short-term electricity load forecasting model.The effectiveness of the proposed method is verified by simulation experimental analysis. The results show that the proposed method has improved the accuracy of the forecasting model.CONCLUSION: The CEEMD method is used to decompose the original load time series, which improves the accuracy of the measurement model. The GRU prediction model based on improved SOA optimization not only has better prediction accuracy than other prediction models, but also consumes the least amount of time compared to other prediction models.

Investigation of Load, Solar and Wind Generation as Target Variables in LSTM Time Series Forecasting, Using Exogenous Weather Variables

Accurately forecasting energy metrics is essential for efficiently managing renewable energy generation. Given the high variability in load and renewable energy power output, this represents a crucial area of research in order to pave the way for increased adoption of low-carbon energy solutions. Whilst the impact of different neural network architectures and algorithmic approaches has been researched extensively, the impact of utilising additional weather variables in forecasts have received far less attention. This article demonstrates that weather variables can have a significant influence on energy forecasting and presents methodologies for using these variables within a long short-term memory (LSTM) architecture to achieve improvements in forecasting accuracy. Moreover, we introduce the use of the seasonal components of the target time series, as exogenous variables, that are also observed to increase accuracy. Load, solar and wind generation time series were forecast one hour ahead using an LSTM architecture. Time series data were collected in five Spanish cities and aggregated for analysis, alongside five exogenous weather variables, also recorded in Spain. A variety of LSTM architectures and hyperparameters were investigated. By tuning exogenous weather variables, a 33% decrease in mean squared error was observed for solar generation forecasting. A 22% decrease in mean absolute squared error (MASE), compared to 24-h ahead forecasts made by the Transmission Service Operator (TSO) in Spain, was also observed for solar generation. Compared to using the target variable in isolation, utilising exogenous weather variables decreased MASE by approximately 10%, 15% and 12% for load, solar and wind generation, respectively. By using the seasonal component of the target variables as an exogenous variable itself, we demonstrated decreases in MASE of 19%, 12% and 8% for load, solar and wind generation, respectively. These results emphasise the significant benefits of incorporating weather and seasonal components into energy-related time series forecasts.

Generating Synthetic Electricity Load Time Series at District Scale Using Probabilistic Forecasts

Thanks to various European directives, individuals are empowered to share and trade electricity within Renewable Energy Communities, enhancing the operational efficiency of local energy systems. The digital transformation of the energy market enables the integration of decentralized energy resources using cloud computing, the Internet of Things, and artificial intelligence. In order to assess the feasibility of new business models based on data-driven solutions, various electricity consumption time series are necessary at this level of aggregation. Since these are currently not yet available in sufficient quality and quantity, and due to data privacy reasons, synthetic time series are essential in the strategic planning of smart grid energy systems. By enabling the simulation of diverse scenarios, they facilitate the integration of new technologies and the development of effective demand response strategies. Moreover, they provide valuable data for assessing novel load forecasting methodologies that are essential to manage energy efficiently and to ensure grid stability. Therefore, this research proposes a methodology to synthesize electricity consumption time series by applying the Box–Jenkins method, an intelligent sampling technique for data augmentation and a probabilistic forecast model. This novel approach emulates the stochastic nature of electricity consumption time series and synthesizes realistic ones of Renewable Energy Communities concerning seasonal as well as short-term variations and stochasticity. Comparing autocorrelations, distributions of values, and principle components of daily sequences between real and synthetic time series, the results exhibit nearly identical characteristics to the original data and, thus, are usable in designing and studying efficient smart grid systems.

Load Balance Forecasting Based on Hybrid Deep Neural Network

Load forecasting is the foundation of utility design, and it is a fundamental business problem in the utility industry. Load forecasting, mainly referring to forecasting electricity demand and energy, is being used throughout all segments of the electric power industry, including generation, transmission, distribution, and retail. In this paper, a long short-term memory network with a hybrid approach is improved with a dense algorithm and proposed for electricity load forecasting. A long short-term memory network is designed to effectively exhibit the dynamic behavior of load time series. The proposed model is tested for Panama study including historical data and weather variables. The prediction accuracy is validated by performance metrics, and the best of the metrics are attained when mean absolute error is 5.262, mean absolute percentage error 0.0000376, and root mean square error 18.243. The experimental results show a high prediction rate for load balance forecasting of electric power consumption.