Accurate Demand Prediction Research Articles

The Demand Prediction of Water Capacity for Drinking Water Plant by Artificial Neural Network

With the constant progress of the times and the rapid economic and social development, the demand for water consumption in the Hexi area of Xiangtan City is continuously rising. Ensuring the safety of urban water supply has become a crucial task. In this paper, we explore the effectiveness of the artificial neural network model in predicting water demand, leveraging the operational data from a water plant in Xiangtan. Thirty-three parameters are employed in this study to forecast water capacity. The results of our analysis reveal that the back propagation (BP) neural network model offers a more accurate and reliable prediction of water demand. This model, through its iterative learning process, is able to capture the complex relationships and patterns inherent in the water demand data. By adjusting its weights and thresholds based on the error between predicted and actual values, the BP neural network continuously improves its predictive accuracy. The application of the BP neural network in water demand prediction not only enhances the precision of forecasts but also contributes to better water resource management and allocation. It enables authorities to make informed decisions regarding water supply planning, ensuring the reliability and sustainability of the urban water system. The BP neural network model demonstrates its potential in accurately predicting water demand in Xiangtan’s Hexi area, thus contributing to the safety and efficiency of the urban water supply system.

An applied study of a technique incorporating machine learning algorithms to optimize water demand prediction

Abstract In water resource management, accurate water demand prediction is essential for developing effective water supply strategies and optimizing resource allocation. This study aims to investigate machine learning algorithms, particularly echo state network (ESN) models, to improve the accuracy and efficiency of water demand prediction. ESN models are selected for their excellent nonlinear time series processing capabilities, which address the challenges of traditional prediction methods when dealing with complex water resource systems. By optimizing the parameters of the ESN model, the study hopes to provide a more scientific and efficient method for residential domestic and agricultural water demand forecasting, thus supporting more refined water resources planning and management decisions. Residential water demand prediction and crop water demand prediction are the two parts of this study. In the prediction of residential water demand, based on the actual data of City Z, the optimized ESN model predicts the water demand in 2025, and the total water demand in the baseline scenario is 790.9 million m3, and the expected values of water demand in different scenarios combined with the economic growth rate and the change of water price range from 659.4708 million m3 to 730.448 million m3. The article’s accuracy analysis of crop water demand prediction indicates that the model’s relative errors in predicting the water demand of the three major crops are limited to 10%. The ESN model optimized using the machine learning algorithm in this paper has good potential for water demand prediction and is an efficient and accurate prediction tool for managing water resources.

Analysis of High-frequency Occurrence Areas of Beijing Taxis based on the K-means Method

As the economy grows and of the pace of life accelerates, travelling by taxi has gradually become one of the most convenient and fastest choices for residents to go out. However, because of the randomness of people's travelling and the mobility of taxi driving, there exist an imbalance between the travelling demand of regional taxi users and the supply of taxis. In order to avoid the problem of uneven distribution of taxis and to achieve accurate prediction of taxi demand, this paper selects the vehicle number, latitude and longitude, and carrying status data of Beijing taxis as the basis of the study, and researches to explore the spatial distribution characteristics of the travelling demand of taxi users. The main contents of the paper are as follows. First of all, this paper initializes the GPS data of taxis, and obtains the boarding and alighting locations. The distribution of taxi regional travel demand is described by using K-means clustering method. Finally, the paper shows five high frequency locations. It is of great significance for taxi dispatching management, reducing empty rate, saving energy consumption and maximizing passenger travel demand.

Demand Prediction of Shared Bicycles Based on Graph Convolutional Network-Gated Recurrent Unit-Attention Mechanism

Shared bicycles provide a green, environmentally friendly, and healthy mode of transportation that effectively addresses the “final mile” problem in urban travel. However, the uneven distribution of bicycles and the imbalance of user demand can significantly impact user experience and bicycle usage efficiency, which makes it necessary to predict bicycle demand. In this paper, we propose a novel shared-bicycle demand prediction method based on station clustering. First, to address the challenge of capturing patterns in station-level bicycle demand, which exhibits significant fluctuations, we employ a clustering method that combines graph information from the bicycle transfer graph and potential energy. This method aggregates closely related stations into corresponding prediction regions. Second, we use the GCN-CRU-AM (Graph Convolutional Network-Gated Recurrent Unit-Attention Mechanism) model to predict bicycle demand in each region. This model extracts the spatial information and correlation between regions, integrates time feature data and local weather data, and assigns weights to the input features. Finally, experimental results based on the data from Citi Bike System in New York City demonstrate that the proposed model achieves a more accurate demand prediction.

Enhancing hourly heat demand prediction through artificial neural networks: A national level case study

Meeting the goal of zero emissions in the energy sector by 2050 requires accurate prediction of energy consumption, which is increasingly important. However, conventional bottom-up model-based heat demand forecasting methods are not suitable for large-scale, high-resolution, and fast forecasting due to their complexity and the difficulty in obtaining model parameters. This paper presents an artificial neural network (ANN) model to predict hourly heat demand on a national level, which replaces the traditional bottom-up model based on extensive building simulations and computation. The ANN model significantly reduces prediction time and complexity by reducing the number of model input types through feature selection, making the model more realistic by removing non-essential inputs. The improved model can be trained using fewer meteorological data types and insufficient data, while accurately forecasting the hourly heat demand throughout the year within an acceptable error range. The model provides a framework to obtain accurate heat demand predictions for large-scale areas, which can be used as a reference for stakeholders, especially policymakers, to make informed decisions.

Short-Term Electricity Demand Forecasting Using Deep Neural Networks: An Analysis for Thai Data

Electricity demand forecasting plays a significant role in energy markets. Accurate prediction of electricity demand is the key factor in optimizing power generation and consumption, saving energy resources, and determining energy prices. However, integrating energy mix scenarios, including solar and wind power, which are highly nonlinear and seasonal, into an existing grid increases the uncertainty of generation, creating additional challenges for precise forecasting. To tackle such challenges, state-of-the-art methods and algorithms have been implemented in the literature. Artificial Intelligence (AI)-based deep learning models can effectively handle the information of long time-series data. Based on patterns identified in datasets, various scenarios can be developed. In this paper, several models were constructed and tested using deep AI networks in two different scenarios: Scenario1 used data for weekdays, excluding holidays, while Scenario2 used the data without exclusion. To find the optimal configuration, the models were trained and tested within a large space of alternative hyperparameters. We used an Artificial Neural Network (ANN)-based Feedforward Neural Network (FNN) to show the minimum prediction error for Scenario1 and a Recurrent Neural Network (RNN)-based Gated Recurrent Network (GRU) to show the minimum prediction error for Scenario2. From our results, it can be concluded that the weekday dataset in Scenario1 prepared by excluding weekends and holidays provides better forecasting accuracy compared to the holistic dataset approach used in Scenario2. However, Scenario2 is necessary for predicting the demand on weekends and holidays.

Improved Giza pyramids construction algorithm for Modify the deep neural network-based method for energy demand forecasting

Accurate prediction of energy demand is crucial for improving services, reducing costs, and optimizing operations in energy systems. Deep neural networks (DNNs) have emerged as a popular method for energy demand forecasting. However, the performance of DNNs can be affected by data quality and hyperparameter selection. To address these concerns, this study proposes a novel energy demand forecasting technique that combines DNNs with an enhanced Giza pyramid construction methodology. The aim of this study is to provide a more reliable and effective approach for forecasting energy demand. The DNNs are employed to capture the complex relationships between input and output variables, while the Giza pyramids algorithm is utilized to optimal selection of hyperparameters of the network. Real-world energy demand data is used to evaluate the proposed approach, comparing it with state-of-the-art baseline models. The research methodology involves assessing the suggested approach using real-world energy demand information and conducting a comparative analysis with cutting-edge baseline models, including modified BP neural network (MBPNN), Neural Network based Genetic Algorithm (NNGA), and reinforcement learning and deep neural network (RLDNN). The IGPCA/CNN method outperforms other methods in energy prediction accuracy across short-term, medium-term, and long-term time scales. It achieves an MSE score of 0.564, lower than MBPNN, NNGA, and RLDNN. In medium-term prediction, it achieves an MSE score of 0.587, better than MBPNN, NNGA, and RLDNN. In long-term prediction, it achieves an MSE score of 0.629, lower than MBPNN and RLDNN. Further analysis and validation experiments are needed to ensure robustness and generalizability. Comparing the method with other state-of-the-art approaches can provide a comprehensive understanding of its superiority. The performance of the models is evaluated based on reliability and effectiveness in energy demand forecasting. The major conclusion of this study is that the proposed approach outperforms the initial models in accurately forecasting energy demand. The combination of DNNs and the improved Giza pyramid construction methodology results in enhanced performance, demonstrating superior reliability and effectiveness compared to other models. The study highlights the significance of accurate energy demand prediction for optimizing energy systems and reducing costs.

Modelling the Impact of Adverse Weather on Airport Peak Service Rate with Machine Learning

Accurate prediction of traffic demand and airport capacity plays a crucial role in minimising ground delays and airborne holdings. This paper focuses on the latter aspect. Adverse weather conditions present significant challenges to airport operations and can substantially reduce capacity. Consequently, any predictive model, regardless of its complexity, should account for weather conditions when estimating the airport capacity. At present, the sole shared platform for airport capacity information in Europe is the EUROCONTROL Public Airport Corner, where airports have the option to voluntarily report their capacities. These capacities are presented in tabular form, indicating the maximum number of hourly arrivals and departures for each possible runway configuration. Additionally, major airports often provide a supplementary table showing the impact of adverse weather in a somewhat approximate manner (e.g., if the visibility is lower than 100 m, then arrival capacity decreases by 30%). However, these tables only cover a subset of airports, and their generation is not harmonised, as different airports may use different methodologies. Moreover, these tables may not account for all weather conditions, such as snow, strong winds, or thunderstorms. This paper presents a machine learning approach to learn mapping from weather conditions and runway configurations to the 99th percentile of the delivered throughput from historical data. This percentile serves as a capacity proxy for airports operating at or near capacity. Unlike previous attempts, this paper takes a novel approach, where a single model is trained for several airports, leveraging the generalisation capabilities of cutting-edge machine learning algorithms. The results of an experiment conducted using 2 years of historical traffic and weather data for the top 45 busiest airports in Europe demonstrate better alignment in terms of mean pinball error with the observed departure and arrival throughput when compared to the operational capacities reported in the EUROCONTROL Public Airport Corner. While there is still room for improvement, this system has the potential to assist airports in defining more reasonable capacity values, as well as aiding airlines in assessing the impact of adverse weather on their flights.

Power Load Forecasting System Based on Deep Hybrid Learning Model

Accurate prediction of electricity demand is crucial to ensure the stability and dependability of local power grids. Numerous scholars have put forth comprehensive prediction systems; However, the majority of these models fail to capture the inherent global features present within the data. This study introduces a novel integrated prediction system that leverages the synergistic capabilities of Time Convolutional Network (TCN) and Long Short-Term Memory (LSTM) architectures, aiming to enhance the accuracy of short-term electric load forecasting. The initial step is to establish separate prediction models for LSTM and TCN, with a focus on electric load data. After combining the output results of these models, the reciprocal of the error square ratio was used as a weighting factor. By using this approach, the LSTM-TCN model for combined prediction is created. This research paper performs an exhaustive examination of the case study by employing authentic data sourced from the Australian Energy Management Authority. The study's findings support that the LSTM-TCN model outperforms both single prediction models and traditional network models in terms of performance. The results indicate that the LSTM-TCN model exhibits greater accuracy in predicting short-term energy demand.

Developing a hybrid model for accurate short-term water demand prediction under extreme weather conditions: a case study in Melbourne, Australia

Accurate prediction of short-term water demand, especially, in the case of extreme weather conditions such as flood, droughts and storms, is crucial information for the policy makers to manage the availability of freshwater. This study develops a hybrid model for the prediction of monthly water demand using the database of monthly urban water consumption in Melbourne, Australia. The dataset consisted of minimum, maximum, and mean temperature (°C), evaporation (mm), rainfall (mm), solar radiation (MJ/m2), maximum relative humidity (%), vapor pressure (hpa), and potential evapotranspiration (mm). The dataset was normalized using natural logarithm and denoized then by employing the discrete wavelet transform. Principle component analysis was used to determine which predictors were most reliable. Hybrid model development included the optimization of ANN coefficients (its weights and biases) using adaptive guided differential evolution algorithm. Post-optimization ANN model was trained using eleven different leaning algorithms. Models were trained several times with different configuration (nodes in hidden layers) to achieve better accuracy. The final optimum learning algorithm was selected based on the performance values (regression; mean absolute, relative and maximum error) and Taylor diagram.

Influence of heating, air conditioning and vehicle automation on the energy and power demand of electromobility

With the increasing number of electric vehicles in the transport sector, the relevance of accurate energy and power demand predictions of electromobility is growing. Thereby, different vehicle functions, especially heating and air conditioning and vehicle automation, have a significant influence. In accordance with the upcoming Euro-7 emissions standard, the energy consumption for heating even has to be contained in the manufacturer’s consumption data in the future. To increase the accuracy of energy and power demand predictions of electromobility, the energy consumption of vehicle functions such as heating and air conditioning as well as the energy savings through vehicle automation must be considered.This paper presents approaches for modeling and simulation the energy consumption of heating, air conditioning and vehicle automation which can be used as an extension of electric vehicles WLTP (Worldwide Harmonized Light-Duty Vehicles Test Procedure) consumption simulation on the level of vehicle classes. The Germany-wide results of the electric vehicles energy demand for heating and air conditioning on the level of NUTS3-areas (Nomenclature of territorial units for statistics) and vehicle classes show regionally different results and confirm the relevance of the research approach. Vehicle automation results are described on the level of the five SAE automation levels (Society of Automotive Engineers automation levels) and the vehicle classes. The approaches and results can be used for single vehicles or assumed vehicle fleets.

Method for Dynamic Prediction of Oxygen Demand in Steelmaking Process Based on BOF Technology

Oxygen is an important energy medium in the steelmaking process. The accurate dynamic prediction of oxygen demand is needed to guarantee molten steel quality, improve the production rhythm, and promote the collaborative optimization of production and energy. In this work, a analysis of the mechanism and of industrial big data was undertaken, and we found that the characteristic factors of Basic Oxygen Furnace (BOF) oxygen consumption were different in different modes, such as duplex dephosphorization, duplex decarbonization, and the traditional mode. Based on this, a dynamic-prediction modeling method for BOF oxygen demand considering mode classification is proposed. According to the characteristics of BOF production organization, a control module based on dynamic adaptions of the production plan was researched to realize the recalculation of the model predictions. A simulation test on industrial data revealed that the average relative error of the model in each BOF mode was less than 5% and the mean absolute error was about 450 m3. Moreover, an accurate 30-minute-in-advance prediction of dynamic oxygen demand was realized. This paper provides the method support and basis for the long-term demand planning of the static balance and the short-term real-time scheduling of the dynamic balance of oxygen.

MagicScaler: Uncertainty-Aware, Predictive Autoscaling

Predictive autoscaling is a key enabler for optimizing cloud resource allocation in Alibaba Cloud's computing platforms, which dynamically adjust the Elastic Compute Service (ECS) instances based on predicted user demands to ensure Quality of Service (QoS). However, user demands in the cloud are often highly complex, with high uncertainty and scale-sensitive temporal dependencies, thus posing great challenges for accurate prediction of future demands. These in turn make autoscaling challenging---autoscaling needs to properly account for demand uncertainty while maintaining a reasonable trade-off between two contradictory factors, i.e., low instance running costs vs. low QoS violation risks. To address the above challenges, we propose a novel predictive autoscaling framework MagicScaler , consisting of a Multi-scale attentive Gaussian process based predictor and an uncertainty-aware scaler. First, the predictor carefully bridges the best of two successful prediction methodologies---multi-scale attention mechanisms, which are good at capturing complex, multi-scale features, and stochastic process regression, which can quantify prediction uncertainty, thus achieving accurate demand prediction with quantified uncertainty. Second, the scaler takes the quantified future demand uncertainty into a judiciously designed loss function with stochastic constraints, enabling flexible trade-off between running costs and QoS violation risks. Extensive experiments on three clusters of Alibaba Cloud in different Chinese cities demonstrate the effectiveness and efficiency of MagicScaler , which outperforms other commonly adopted scalers, thus justifying our design choices.

Edge-Based Short-Term Energy Demand Prediction

The electrical grid is gradually transitioning towards being an interconnected area of the smart grid, where embedded devices operate in an autonomous manner without any human intervention. An important element for this transition is the energy demand prediction, since the needs for energy have substantially increased due to the introduction of new and heavy consumption sources, such as electric vehicles. Accurate energy demand prediction, especially for short-term durations (i.e., minutes to hours), allows grid operators to produce the substantial amount needed to satisfy the demand–response equilibrium and avoid peak electricity load conditions that may also lead to blackouts in densely populated areas. However, to achieve such an accuracy level, machine learning (ML) models require extensive training with historical measurements, which is usually resource intensive (e.g., in memory and processing power). Hence, deriving accurate predictions for short-term energy demands is challenging due to the absence of external factors such as environmental data from different regions and seasons and categorical values such as bank/bridging holidays in the ML model. Additionally, existing work focuses on ML model execution on Cloud platforms, which usually does not satisfy the real-time requirements of grid operators for short-term energy demand predictions. To address these challenges, this article presents a new method that considers environmental factors and categorical values to build an energy profile for each consumer in a multi-access edge computing (MEC) framework. The method is also based on the Temporal Fusion Transformer (TFT) ML model, which allows it to learn the temporal dependencies of the gathered historical measurements and predict energy demands with satisfying accuracy. The method is applied to a home energy management system testbed containing photovoltaic systems, smart meters, sensors and actuators for detecting environmental factors (i.e., temperature, humidity and radiation) as well as energy storage systems as an additional energy supply source. The MEC framework is deployed in data concentrator devices where the TFT ML model is executed with low resource requirements, ensuring additional security as the data do not leave the location where they are produced.

Viability and accessibility of the Great Lakes microclimate data over current TMY weather data for accurate energy demand predictions

As climate change continues to evolve with alarming increases in extreme weather conditions, its impact on building cooling and heating energy needs, in order to maintain acceptable occupant comfort becomes increasingly beyond the design constraints. To reduce such mismatch, it is important to improve the accuracy of weather files used for building energy simulations (BES). The widely used simulation software, EnergyPlus (EP) provides Typical Meteorological Year (TMY) weather data, which are not designed to detect extreme conditions and have limited data locations. There is a greater number of Microclimate (MC) stations that can be incorporated into BES for higher resolution data. The challenge is the time-consuming data preparation process to match the EP format. To address this issue, the co-authors developed a program, Virtual Information Fabric Infrastructure (VIFI), which automatically prepares MC data for more accurate BES. To demonstrate the need for higher resolution data, this study analyzed locations in the Great Lakes region and found that July's cooling demand for Rogers City MI and Alpena MI, which are located 35 mi (56 km) apart, differed by 19%. The results emphasize the importance of considering MC data in BES and provide a solution for their efficient integration into BES.

Implementasi Algoritma Monte Carlo Untuk Memprediksi Permintaan Aksesoris Mobil

Toko ADS Variasi dan AC Mobil is a shop engaged in the sale of car accessories, repairing and replacing car air conditioning spare parts in Kota Palembang, Toko ADS Variasi dan AC Mobil faces challenges in predicting the demand for the right car accessories to sell so that there is no excess or shortage of stock items. So that a method is needed that can be implemented into the inventory system and sales and purchase records. In this research, the author uses the Monte Carlo Algorithm which will generate predictions of demand for car accessories by considering variations in factors that affect demand, taking historical data on car accessories sales from the relevant period, then conducting Monte Carlo simulations using the data as input. The simulation will be repeated as many times as needed and provide results in the form of probability distributions of demand for car accessories that may occur. By taking the average value of the probability distribution, an accurate demand prediction can be generated. The result obtained from this research is a more accurate prediction of demand for car accessories based on the latest historical data so that stores can determine the right amount of stock and avoid overstocking or understocking. In addition, by considering the factors that affect demand, stores can plan more effective marketing strategies and increase store profits

Developing a multiview spatiotemporal model based on deep graph neural networks to predict the travel demand by bus

The accurate prediction of travel demand by bus is crucial for effective urban mobility demand management. However, most models of travel demand prediction by bus tend to focus on the bus’s spatiotemporal dependencies, while ignoring the interactions between buses and other transportation modes, such as metros and taxis. We propose a Multiview Spatiotemporal Graph Neural Network (MSTGNN) model to predict short-term travel demand by bus. It emphasizes the ability to capture the interaction dependencies among the travel demand of buses, metros, and taxis. Firstly, a multiview graph consisting of bus, metro, and taxi views is constructed, with each view containing both a local and global graph. Secondly, a multiview attention-based temporal graph convolution module is developed to capture spatiotemporal and cross-view interaction dependencies among different transport modes. Especially, to address the uneven spatial distributions of features in multiview learning, the cross-view spatial feature consistency loss is introduced as an auxiliary loss. Finally, we conduct intensive experiments using a real-world dataset from Shenzhen, China. The results demonstrate that our proposed MSTGNN model performs better than the existing models. Ablation experiments validate the contributions of various modes of transportation to the improvement of the model’s performance.

A Short-Term Parking Demand Prediction Framework Integrating Overall and Internal Information

With the development of smart cities and smart transportation, cities can gradually provide people with more information to facilitate their life and travel, and parking is also inseparable from both of them. Accurate on-street parking demand prediction can improve parking resource utilization and parking management efficiency, as well as potentially improve urban traffic conditions. Previous parking demand prediction methods seldom consider the correlation between the parking demand of a road section and its surroundings. Therefore, in order to capture the correlation of parking demand in the temporal and spatial dimensions as carefully as possible and enrich the relevant features in the prediction model so as to achieve more accurate prediction results, we designed a parking demand prediction structure that considers different features from two perspectives: overall and internal. We used gated recurrent units (GRU) to extract demand influences in the temporal dimension. The GRU is used in combination with a graph convolutional neural network (GCN) to extract demand influencing factors in the spatial dimension. Additionally, a more detailed representation is designed to express spatial dimensional features. Then, based on the historical parking demand features extracted using encoder–decoder, we fuse the extracted spatio-temporal features with them to finally obtain an on-street parking demand prediction model combining the overall and the internal information. By combining them, we can integrate more correlation factors to achieve a more accurate parking demand prediction. The performance of the model is evaluated by real parking data in Xiufeng District of Guilin. The results show that the proposed model achieves good prediction performance compared with other baselines. In addition, we also design feature ablation experiments. Through the comparison of the results, we find that each feature considered in the proposed model is important in parking demand prediction.

Forecast energy demand, CO2 emissions and energy resource impacts for the transportation sector

Managing energy demand and reducing greenhouse gas emissions are among the most significant challenges ahead for many countries. Accurate prediction of energy demand and CO2 emissions is an important method to tackle the challenge. In addition, as a major source of energy consumption and CO2 emissions, the transportation sector is at the center of the problem. This study employs a hybrid approach integrating a multi-objective mathematical model with data-driven machine learning algorithms to predict energy demand and CO2 emissions in the transportation sector with improved accuracy. Sensitivity analyses are conducted to evaluate the impact of individual and joint consumptions of different energy resources on CO2 emissions. Case studies are conducted with countrywide statistics in Canada where the prediction indicates that energy demand and CO2 emissions will increase by 34.72% and 50.02% from 2019 to 2048 in the Canadian transportation sector. Results also indicated the varied impacts of different energy resources such that a 5% increase in energy demand in oil, gas, electricity, and renewable energy will result in + 4.8%, +0.315%, +0.28%, and −0.51% changes in CO2 emissions respectively. The proposed framework with integrated prediction model and the sensitivity analyses can help identify critical factors impacting CO2 emissions in the transportation sector and provide quantitative references for energy demand management and CO2 emissions reduction.

Secure-Enhanced Federated Learning for AI-Empowered Electric Vehicle Energy Prediction

Although AI-empowered schemes bring some sound solutions to stimulate more reasonable energy distribution schemes between charging stations (CSs) and CS providers, frequent data sharing between them is possible to incur many security and privacy breaches. To solve these problems, federated learning (FL) is an ideal solution that only requires CSs to upload local models instead of detailed data. Although the CSs’ electricity consumption need not to be exposed to the server directly, FL-based schemes still have been excavated several security threats such as information exploiting attacks, data poisoning attacks, model poisoning attacks, and free-riding attacks. Hence, in this article, both the effectiveness of energy management and the potential risks of FL for electric vehicle infrastructures (EVIs) are considered, we propose a lightweight authentication FL-based energy demand prediction for EVIs with premium-penalty mechanism. Security analysis and performance evaluation prove that our proposed framework can generate an accurate electricity demand prediction framework to defend multiple FL attacks for EVIs.