The prediction of electricity sale market is directly related to the balance of electricity supply and demand. The high-precision prediction method of electricity sale market will effectively avoid ...the imbalance of system supply and demand, and improve the security and economy of the system. In the past, it was difficult to obtain the information of users' electricity consumption, so the prediction methods of electricity sale market in the past were based on the overall electricity sale. The advantage of this method is that it can get the forecast value of the future electricity sales quickly and intuitively, while the disadvantage is that it fails to consider the electricity development trend of all walks of life that compose the electricity sales in detail.
This paper presents a new distributed smart charging strategy for grid integration of plug-in electric vehicles (PEVs). The main goal is to smooth the daily grid load profile while ensuring that each ...PEV has a desired state of charge level at the time of departure. Communication and computational overhead, and PEV user privacy are also considered during the development of the proposed strategy. It consists of two stages: 1) an offline process to estimate a reference operating power level based on the forecasted mobility energy demand and base loading profile, and 2) a real-time process to determine the charging power for each PEV so that the aggregated load tracks the reference loading level. Tests are carried out both on primary and secondary distribution networks for different heuristic charging scenarios and PEV penetration levels. Results are compared to that of the optimal solution and other state-of-the-art techniques in terms of variance and peak values, and shown to be competitive. Finally, a real vehicle test implementation is done using a commercial-of-the-shelf charging station and an electric vehicle.
•Daily load profiles of fourteen academic buildings are classified using k-means.•Three methods are compared using different data collection time-steps and timeframes.•Two distinct groups of ...buildings are identified regarding power demand patterns.•A seasonal effect is observed using six-month and one-year timeframes.•A two-cluster classification is confirmed for building stock aggregated profiles.
We investigated clustering techniques on time series of daily electric load profiles of fourteen higher education buildings on the same campus. A k-means algorithm is implemented, and three different methods are compared: time-series features extraction with Manhattan distance and raw time series with Euclidian distance and Dynamic Time Warping. The impact of data characteristics with data collection time-steps and timeframes is studied using a database of more than 6,500 daily electric load profiles. We show that Euclidian distance applied to electric demand time series with three-month timeframes and ten-minute time-step provides the most consistent clustering results. In addition, useful insights are highlighted for non-residential buildings electric demand modeling and forecasting. Two groups of buildings can be distinguished regarding electric load profile patterns. On one hand, teaching, research, libraries, and gymnasium buildings show similar patterns distributed in two clusters corresponding to business days and closing days load profiles. On the other hand, campus office buildings present a larger number of clusters inconsistent with day-type dependent load profiles. A seasonal effect is also observed using six-month and one-year timeframes. Finally, a two-cluster distribution is obtained when aggregating all buildings load profiles.
Display omitted
•Methodology for 48-hour multi-step heat demand forecasting in a DH system.•Gaussian process regression outperforms considered machine learning methods.•Accurate temperature forecasts ...are important, solar irradiation forecasts are not.•Forecasting errors for 48 h ahead below 3% of the max. heating power.•Proposed forecasting solution can be fitted to different DH systems.
Short-term heat demand forecasting in district heating (DH) systems is essential for a sufficient heat supply and optimal operation of the DH. In this study, a machine learning based multi-step short-term heat demand forecasting approach using the data of the largest Slovenian DH system is considered. The proposed approach involved feature extraction and comparative analysis of different representative machine learning based forecasting models. Nonlinear models performed better than linear models, and the best forecasting results were obtained by Gaussian process regression (GPR), where the mean absolute normalized error was 2.94% of the maximum heating power of the DH system. The analysis confirmed the importance of accurate temperature forecasts but did not confirm the relevance of using future solar irradiation forecasts. The optimal length of training data is shown to be 3 years, and past data of up to 4 days can be used as input to increase the forecasting accuracy. The forecasting model (GPR) proposed in this study can be fitted to different DH systems. In the presented case study, it was selected to implement the online forecasting solution for the DH of Ljubljana and has been generating forecasts with a mean absolute normalized error of 2.70% since November 2019.
Abstract
The COVID-19 lockdown has instigated significant changes in household behaviours across a variety of categories including water consumption, which in the south and east regions of England is ...at an all-time high. We analysed water consumption data from 11,528 households over 20 weeks from January 2020, revealing clusters of households with distinctive temporal patterns. We present a data-driven household water consumer segmentation characterising households’ unique consumption patterns and we demonstrate how the understanding of the impact of these patterns of behaviour on network demand during the COVID-19 pandemic lockdown can improve the accuracy of demand forecasting. Our results highlight those groupings with the highest and lowest impact on water demand across the network, revealing a significant quantifiable change in water consumption patterns during the COVID-19 lockdown period. The implications of the study to urban water demand forecasting strategies are discussed, along with proposed future research directions.
Winter wheat (Triticum aestivum L.) is one of the most important cereal crops, supplying essential food for the world population. Because the United States is a major producer and exporter of wheat ...to the world market, accurate and timely forecasting of wheat yield in the United States (U.S.) is fundamental to national crop management as well as global food security. Previous studies mainly have focused on developing empirical models using only satellite remote sensing images, while other yield determinants have not yet been adequately explored. In addition, these models are based on traditional statistical regression algorithms, while more advanced machine learning approaches have not been explored. This study used advanced machine learning algorithms to establish within-season yield prediction models for winter wheat using multi-source data to address these issues. Specifically, yield driving factors were extracted from four different data sources, including satellite images, climate data, soil maps, and historical yield records. Subsequently, two linear regression methods, including ordinary least square (OLS) and least absolute shrinkage and selection operator (LASSO), and four well-known machine learning methods, including support vector machine (SVM), random forest (RF), Adaptive Boosting (AdaBoost), and deep neural network (DNN), were applied and compared for estimating the county-level winter wheat yield in the Conterminous United States (CONUS) within the growing season. Our models were trained on data from 2008 to 2016 and evaluated on data from 2017 and 2018, with the results demonstrating that the machine learning approaches performed better than the linear regression models, with the best performance being achieved using the AdaBoost model (R2 = 0.86, RMSE = 0.51 t/ha, MAE = 0.39 t/ha). Additionally, the results showed that combining data from multiple sources outperformed single source satellite data, with the highest accuracy being obtained when the four data sources were all considered in the model development. Finally, the prediction accuracy was also evaluated against timeliness within the growing season, with reliable predictions (R2 > 0.84) being able to be achieved 2.5 months before the harvest when the multi-source data were combined.
To reach the carbon emission reduction targets set by the European Union, the building sector has embraced multiple strategies such as building retrofit, demand side management, model predictive ...control and building load forecasting. All of which require knowledge of the building dynamics in order to effectively perform. However, the scaling-up of building modelling approaches is still, as of today, a recurrent challenge in the field. The heterogeneous building stock makes it tedious to tailor interpretable approaches in a scalable way. This work puts forward an automated and scalable method for stochastic model identification of building heat dynamics, implemented on a set of 247 Dutch residential buildings. From established models and selection approach, automation extensions were proposed along with a novel residual auto-correlation indicator, i.e., normalized Cumulated Periodogram Boundary Excess Sum (nCPBES), to classify obtained model fits. Out of the available building stock, 93 building heat dynamics models were identified as good fits, 95 were classified as close and 59 were designed as poor. The identified model parameters were leveraged to estimate thermal characteristics of the buildings to support building energy benchmarking, in particular, building envelope insulation performance. To encourage the dissimination of the work and assure reproducibility, the entire code base can be found on Github along with an example data set of 3 anonymized buildings. The presented method takes an important step towards the automation of building modeling approaches in the sector. It allows the development of applications at large-scale, enhancing building performance benchmarks, boosting city-scale building stock scenario modeling and assisting end-use load identifications as well as building energy flexibility potential estimation.
A robust machine learning methodology is used to generate a site-specific power-curve of a full-scale isolated wind turbine operating in an atmospheric boundary layer to drastically improve the power ...predictions, and, thus, the forecasting of the monthly energy production estimates. The study has important implication in measuring the financial feasibility of wind farms by improving the accuracy of monthly energy estimates. The significance of the study is that atmospheric stability and air-density are accounted in the power predictions of the wind turbine. Artificial Neural Networks (ANN) machine learning approach is used to generate multi-parameter input models to estimate the power produced by the wind turbine. The ANN model in this study uses Feed Forward Back Propagation (FFBP) algorithm. The power- and wind-data is obtained from a 2.5 MW wind turbine that has a Meteorological tower located 900 m Southwest of the wind turbine in Kirkwood, Iowa, USA. The study investigates the role of atmospheric boundary-layer metrics – Wind Speed, Density (a measure of stratification), Richardson Number, turbulence intensity, and wind shear as input parameters into the ANN model. The study investigates the influence of FFBP ANN hyper-parameters on the power prediction accuracy. Comparison of the FFBP ANN model to other power curve correction techniques demonstrated an improvement in the Mean Absolute Error (MAE) of 40% when compared to the density correction (the next closest). The five-parameter 4-layer FFBP ANN has an average energy production error of 0.4% for the nine months while the IEC this error is −3.7% and for the air density correction the error is −1.9%, respectively. Finally, the study determines the performance of the FFBP ANN model for different atmospheric stability regimes (Unstable, Stable, Strongly Stable, Strongly Unstable and Neutral) classified using two criterions - Richardson number and Turbulence intensity. The largest MAE occurs during the strongly stable regime of the atmospheric boundary layer for both criteria.
The major highlights of the manuscript are.Atmospheric Inputs improve ANN performance for long term energy forecasting.The optimal number of hidden layers was 4 (100, 50, 20,10 neurons).The 4 layer ANN outperformed RBF, RF, SVR and GP machine learning algorithms.The model improved the energy production forecasting error to 0.4%.The results were dependent on atmospheric stability.
Forecasting the burden of COVID-19 has been impeded by limitations in data, with case reporting biased by testing practices, death counts lagging far behind infections, and hospital census reflecting ...time-varying patient access, admission criteria, and demographics. Here, we show that hospital admissions coupled with mobility data can reliably predict severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2) transmission rates and healthcare demand. Using a forecasting model that has guided mitigation policies in Austin, TX, we estimate that the local reproduction number had an initial 7-d average of 5.8 (95% credible interval CrI: 3.6 to 7.9) and reached a low of 0.65 (95% CrI: 0.52 to 0.77) after the summer 2020 surge. Estimated case detection rates ranged from 17.2% (95% CrI: 11.8 to 22.1%) at the outset to a high of 70% (95% CrI: 64 to 80%) in January 2021, and infection prevalence remained above 0.1% between April 2020 and March 1, 2021, peaking at 0.8% (0.7-0.9%) in early January 2021. As precautionary behaviors increased safety in public spaces, the relationship between mobility and transmission weakened. We estimate that mobility-associated transmission was 62% (95% CrI: 52 to 68%) lower in February 2021 compared to March 2020. In a retrospective comparison, the 95% CrIs of our 1, 2, and 3 wk ahead forecasts contained 93.6%, 89.9%, and 87.7% of reported data, respectively. Developed by a task force including scientists, public health officials, policy makers, and hospital executives, this model can reliably project COVID-19 healthcare needs in US cities.