Fuel Consumption Prediction Research Articles

Fuel Consumption Prediction Based on Whale Algorithm Optimized Support Vector Machine

In order to improve the prediction accuracy of vehicle fuel consumption, this paper proposes a support vector machine model based on whale optimization algorithm for prediction. First, SPSS software is used to analyze the relationship between the influencing factors of fuel consumption. On this basis, the correlation between various influencing factors and fuel consumption is analyzed. Then, the whale optimization algorithm is used to optimize the penalty parameter C and the kernel parameter g in the support vector machine kernel function, and it is brought into the model for simulation. Taking the determination coefficient and mean square error as the evaluation indexes, the prediction results show that the model has high accuracy.

An application of augmented Lagrangian differential evolution algorithm for optimizing the speed of inland ships sailing on the Yangtze River

As an effective way to achieve energy conservation and emission reduction, the speed optimization of inland ships has received increasing attention. In this study, a modeling method for the speed optimization of inland ships sailing on the Yangtze River and a heuristic algorithm called Augmented Lagrangian Differential Evolution (ALDE) method are proposed. The ALDE introduces adaptive control parameters, an elite strategy, and opposition-based learning technology for optimization. To validate the performance of ALDE, it is compared with six other algorithms to solve several well-known constrained optimization problems, and the results show that the ALDE outperforms these six algorithms. To construct a speed optimization model, a discretization model for the Yangtze River waterway is established by segmenting the waterway. Considering the relationship among the environment, ship, engine, and propeller, a fuel consumption prediction model is built, and then, the objective optimization function for inland ship speed optimization is established by integrating the discrete model as well as the fuel consumption model. Finally, numerical simulation results show that the ALDE can realize less fuel consumption for an inland ship sailing from Yibin Port to Jiangyin Port, verifying the feasibility of the ALDE for speed optimization of inland ships.

Prediction of Specific Fuel Consumption of 60 HP 2WD Tractor Using Artificial Neural Networks

In this study, specific fuel consumption was determined at different axle load, tire pressure and drawbar force for a 60 HP tractor. In addition, the results were also estimated with the help of Artificial Neural Networks (ANN), which is one of the machine learning methods. The testings were carried out on the Tractor Draft Test Track. In the study; three different drive tire inner pressures (P) (160 kPa, 120 kPa and 80 kPa), four different dynamic axle loads (W) (1796 daN, 2076 daN, 2276 daN, 2476 daN) and four different traction forces (500 daN,1000 daN, 1500 daN, 2000 daN) were tested. Specific fuel consump-tion values varied between 290.7-542.1 g/kWh depending on the draft force in the testings Despite the 38% increase in the axle load, a 3.5% decrease occurred in the specific fuel consumption values. Specific fuel consumption values in-creased when increased tire inner pressure. Specific fuel consumption in-creased by 1.03% when a 100% increase in tire pressure. In ANN, the most suc-cessful model was determined by trying different training algorithms, transfer functions and the number of neurons in the hidden layer. In the most successful network model, MAE, RMSE values for the prediction of specific fuel consump-tion were found to be 0.005331, 0.007551 respectively.

Time-Aware and Feature Similarity Self-Attention in Vessel Fuel Consumption Prediction

An accurate vessel fuel consumption prediction is essential for constructing a ship route network and vessel management, leading to efficient sailings. Besides, ship data from monitoring and sensing systems accelerate fuel consumption prediction research. However, the ship data consist of three properties: sequential, irregular time interval, and feature importance, making the predicting problem challenging. In this paper, we propose Time-aware Attention (TA) and Feature-similarity Attention (FA) applied to bi-directional Long Short-Term Memory (LSTM). TA acquires time importance by nonlinear function from irregular time intervals in each sequence and emphasizes data depending on the importance. FA emphasizes data based on similarities of features in the sequence by estimating feature importance with learnable parameters. Finally, we propose the ensemble model of TA and FA-based BiLSTM. The ensemble model, which consists of fully connected layers, is capable of simultaneously capturing different properties of ship data. The experimental results on ship data showed that the proposed model improves the performance in predicting fuel consumption. In addition to model performance, visualization results of attention maps and feature importance help to understand data properties and model characteristics.

A Comparative Study on Fuel Consumption Prediction Methods of Heavy-Duty Diesel Trucks Considering 21 Influencing Factors

With increasingly prominent environmental problems, controlling automobile exhaust has become essential to the environment. The fuel consumption of transportation is the critical factor that determines exhaust gas. By analyzing the naturalistic driving data of heavy-duty diesel trucks (HDDTs), this paper explored the influence of engine technical state, road features, weather, and temperature conditions on fuel consumption during driving. The detailed process is as follows: Firstly, we collected 1153 naturalistic driving data from 34 HDDTs and made a specific analysis and summary description of the data; secondly, by establishing a binary Logistic regression model, we quantitatively explored the influence of significant factors on the fuel consumption; meanwhile, based on quantitative analysis of factor’s effectiveness, this research used several machine learning algorithms (back-propagation neural network, decision tree, and random forest) to build fuel consumption predictors, and compared the prediction performance of different algorithms. The results showed that the prediction accuracy of the decision tree, back-propagation (BP) neural network, and random forest is 81.38%, 83.98%, and 86.58%, respectively. The random forest showed the best performance in predicting. The conclusions can assist transportation companies in formulating driving training strategies and contribute to reducing energy consumption and emissions.

An Arctic ship performance model for sea routes in ice-infested waters

Global warming is creating significant change in the Arctic environment, with widespread reduction in ice extent, thickness and compactness. This has opened numerous shipping routes through the Arctic and provides the opportunity to reduce the distance, time and emissions of voyages. However, for a ship to operate in such polar routes, ice conditions need to be properly accounted for to accurately estimate fuel consumption. To meet this key challenge, this paper presents a ship performance model designed to incorporate a set of ice resistance algorithms alongside the calculation of open-water ship resistance and fuel consumption. In particular, a novel method is proposed to calculate ship resistance in ice-floe fields, other than the traditional level-ice condition. Subsequently, the model has been used to simulate a voyage in the Northern Sea Route and the fuel consumption prediction agrees well with corresponding full-scale measurement data. Overall, the work provides a practical tool for the emerging Arctic shipping industry to carry out fuel analysis and voyage planning.

Fuel Consumption Prediction for Construction Trucks: A Noninvasive Approach Using Dedicated Sensors and Machine Learning

Decision support and optimization tools to be used in construction often require an accurate estimation of the cost variables to maximize their benefit. Heavy machinery is traditionally one of the greatest costs to consider mainly due to fuel consumption. These typically diesel-powered machines have a great variability of fuel consumption depending on the scenario of utilization. This paper describes the creation of a framework aiming to estimate the fuel consumption of construction trucks depending on the carried load, the slope, the distance, and the pavement type. Having a more accurate estimation will increase the benefit of these optimization tools. The fuel consumption estimation model was developed using Machine Learning (ML) algorithms supported by data, which were gathered through several sensors, in a specially designed datalogger with wireless communication and opportunistic synchronization, in a real context experiment. The results demonstrated the viability of the method, providing important insight into the advantages associated with the combination of sensorization and the machine learning models in a real-world construction setting. Ultimately, this study comprises a significant step towards the achievement of IoT implementation from a Construction 4.0 viewpoint, especially when considering its potential for real-time and digital twins applications.

Data analytics for fuel consumption management in maritime transportation: Status and perspectives

The shipping industry is associated with approximately three quarters of all world trade. In recent years, the sustainability of shipping has become a public concern, and various emissions control regulations to reduce pollutants and greenhouse gas (GHG) emissions from ships have been proposed and implemented globally. These regulations aim to drive the shipping industry in a low-carbon and low-pollutant direction by motivating it to switch to more efficient fuel types and reduce energy consumption. At the same time, the cyclical downturn of the world economy and high bunker prices make it necessary and urgent for the shipping industry to operate in a more cost-effective way while still satisfying global trade demand. As bunker fuel bunker (e.g., heavy fuel oil [HFO], liquified natural gas [LNG]) consumption is the main source of emissions and bunker fuel costs account for a large proportion of operating costs, shipping companies are making unprecedented efforts to optimize ship energy efficiency. It is widely accepted that the key to improving the energy efficiency of ships is the development of accurate models to predict ship fuel consumption rates under different scenarios. In this study, ship fuel consumption prediction models presented in the literature (including the academic literature and technical reports as a typical type of “grey literature”) are reviewed and compared, and models that optimize ship operations based on fuel consumption prediction results are also presented and discussed. Current research challenges and promising research questions on ship performance monitoring and operational optimization are identified.

Analysis and prediction of ship energy efficiency based on the MRV system

ABSTRACT To reduce CO2 emissions from shipping activities to, from, and within the European Union (EU) area, a system of monitoring, reporting, and verification (MRV) of CO2 emissions from ships are implemented in 2015 by the EU. Although the MRV records in 2018 and 2019 have been published, there are scarce studies on the MRV system especially from a quantitative perspective, which restrains the potential of the MRV. To bridge this gap, this paper first analyzes and compares MRV records in 2018 and 2019, and then develops machine learning models for annual average fuel consumption prediction for each ship type combining ship features from an external database. The performance of the prediction models is accurate, with the mean absolute percentage error (MAPE) on the test set no more than 12% and the average R-squared of all the models at 0.78. Based on the analysis and prediction results, model meanings, implications, and extensions are thoroughly discussed. This study is a pioneer to analyze the emission reports in the MRV system from a quantitative perspective. It also develops the first fuel consumption prediction models from a macro perspective using the MRV data. It can contribute to the promotion of green shipping strategies.

Prediction of Main Engine Speed and Fuel Consumption of Inland Ships Based on Deep Learning

The accurate fuel consumption forecasting system is for shipping companies to carry out fuel management more effectively in order to improve economic benefits. This article aims to use deep learning algorithms to improve the accuracy of ship engine speed and fuel consumption prediction. First, the main factors affecting the engine speed and fuel consumption of inland watercrafts are analyzed, and the input parameters of the neural network model are determined. Secondly, based on the dynamic time series characteristics of inland water vessels, the LSTM (Long short-term memory) algorithm with “time parameters” was selected to establish a neural network prediction model. Finally, the difference between the LSTM deep neural network model and the traditional machine learning model is compared, and the obtained prediction data is compared with the actual data. The experimental results show the superiority of deep learning in the prediction of main engine speed and fuel consumption of inland watercraft.

Research on Optimization of Main Engine Speed of Inland Ship Based on Genetic Algorithm

This paper aims at reduced total fuel consumption of voyage by obtaining navigation data of inland ships, and established the fuel consumption prediction model of the main engine in the segment through studying the navigation route by sections according to the characteristics of the navigation environment, and solved the problem with the help of genetic algorithm to obtain the best main engine for each segment main engine speed. The results show that: (1) The error of the neural network model verification data set is mostly within 6%, and the model performance is good; (2) The experimental ship from Nanjing to Jingzhou from June to September adopts the optimized main engine speed to save 0.6232t of fuel, which is 4.06% lower than the original total fuel consumption. This method has practical and guiding significance for reducing the cost of inland ship navigation and improving the ecological environment of the Yangtze River.

An adaptive hyper parameter tuning model for ship fuel consumption prediction under complex maritime environments

An accurate prediction of ship fuel consumption is critical for speed, trim, and voyage optimisation etc. While previous studies have focused on predicting ship fuel consumption with respect to a variety of factors, research on the impact of environmental factors on fuel consumption has been lacking. In addition, although recent research efforts have widely focused on machine learning methods to predict fuel consumption, studies on hyperparameter values that are suitable for these prediction models are limited. To compensate for this deficiency in existing literature, an adaptive hyperparameter tuning method is proposed, and the effects of maritime environmental factors on fuel consumption are taken into account. Through experimentation, the proposed adaptive hyperparameter tuning method was validated via artificial neural network (ANN), support vector regression (SVR), random forest (RF), and least absolute shrinkage and selection operator (Lasso). The hyperparameter tuning proportionally increased the amplitudes of the coefficients of determination (R2) of these algorithms. The increase of the amplitude demonstrated the following trend, in the order of the largest increase to the lowest increase: ANN, Lasso, SVM, and RF. The rates of increase were between 0.0773% and 2.1653%. Furthermore, after the environmental factors were considered, the prediction accuracies of the ANN and Lasso increased; however, the opposite was observed for the SVR and RF. As such, we confirmed that the use of Bayesian optimisation for hyperparameter tuning can effectively improve the fuel consumption prediction accuracy, and our proposed model can therefore serve as a significant reference for calculating fuel consumption.

A Novel Hybrid Fuel Consumption Prediction Model for Ocean-Going Container Ships Based on Sensor Data

Accurate, reliable, and real-time prediction of ship fuel consumption is the basis and premise of the development of fuel optimization; however, ship fuel consumption data mainly come from noon reports, and many current modeling methods have been based on a single model; therefore they have low accuracy and robustness. In this study, we propose a novel hybrid fuel consumption prediction model based on sensor data collected from an ocean-going container ship. First, a data processing method is proposed to clean the collected data. Secondly, the Bayesian optimization method of hyperparameters is used to reasonably set the hyperparameter values of the model. Finally, a hybrid fuel consumption prediction model is established by integrating extremely randomized tree (ET), random forest (RF), Xgboost (XGB) and multiple linear regression (MLR) methods. The experimental results show that data cleaning, the size of the dataset, marine environmental factors, and hyperparameter optimization can all affect the accuracy of the model, and the proposed hybrid model provides better predictive performance (higher accuracy) and greater robustness (smaller standard deviation) as compared with a single model. The proposed hybrid model should play a significant role in ship fuel consumption real-time monitoring, fault diagnosis, energy saving and emission reduction, etc.

In-Vehicle Validation of Heavy-Duty Vehicle Fuel Savings via a Hierarchical Predictive Online Controller

<div class="section abstract"><div class="htmlview paragraph">This paper presents the evolution of a series of connected, automated vehicle technologies from simulation to in-vehicle validation for the purposes of minimizing the fuel usage of a class-8 heavy duty truck. The results reveal that an online, hierarchical model-predictive control scheme, implemented via the use of extended horizon driver advisories for velocity and gear, achieves fuel savings comparable to predictions from software-in-the-loop (SiL) simulations and engine-in-the-loop (EiL) studies that operated with a greater degree of powertrain and chassis automation. The work of this paper builds on prior work that presented in detail this predictive control scheme that successively optimizes vehicle routing, arrival and departure at signalized intersections, speed trajectory planning, platooning, predictive gear shifting, and engine demand torque shaping. This paper begins by outlining the controller development progression from a previously published engine-in-the-loop study to the in-vehicle driver-in-the-loop testing highlighted in this work. The purpose of this field testing is to quantify the level of agreement between field data and prior results, particularly noting the influence of noise, disturbances and unmodeled dynamics. Also detailed are the steps to effectively implement the predictive horizons in experimentation with human rather than automated driver inputs by replacing the conventional speedometer with an augmented driver display. The driver display seeks to maintain driver awareness and route preview simultaneously. Additional features of the instrumented truck testing protocol include detailed route maps on speed limits, intersections, traffic and road grade; highly accurate fuel flowrate measurements calibrated via gravimetric methods; and closed-loop control between the predictive controller, driver, and vehicle controller area network (CAN). These combined capabilities allow for model validation, investigation of the effects of human-in-the-loop on controller performance, and identification of unmodeled dynamics and disturbances. The results show that the driver advisory implementation matches fuel consumption predictions from software-in-the-loop simulations and engine-in-the-loop studies to within 2-9%. Following the study of in-vehicle performance and variability, these results also validate, using real road tests, the up to 21% fuel savings achieved by using the hierarchical control scheme with a combination of multiple chassis and powertrain optimization technologies.</div></div>

An Innovative Methodology to Take into Account Traffic Information on WLTP Cycle for Hybrid Vehicles

The most efficient energy management strategies for hybrid vehicles are the “Optimization-Based Strategies”. These strategies require a preliminary knowledge of the driving cycle, which is not easy to predict. This paper aims to combine Worldwide Harmonized Light-Duty Vehicles Test Cycle (WLTC) low section short trips with real traffic levels for vehicle energy and fuel consumption prediction. Future research can focus on implementing a new strategy for Hybrid Electric Vehicle (HEV) energy optimization, taking into account WLTC and Google Maps traffic levels. First of all, eight characteristic parameters are extracted from real speed profiles, driven in urban road sections in the city of Messina at different traffic conditions, and WLTC short trips as well. The minimum distance algorithm is used to compare the parameters and assign the three traffic levels (heavy, average, and low traffic level) to the WLTC short trips. In this way, for each route assigned from Google maps, vehicle’s energy and fuel consumption are estimated using WLTC short trips remodulated with distances and traffic levels. Moreover, a vehicle numerical model was implemented and used to test the accuracy of fuel consumption and energy prediction for the proposed methodology. The results are promising since the average of the percentage errors’ absolute value between the experimental driving cycles and forecast ones is 3.89% for fuel consumption, increasing to 6.80% for energy.

Comparative analysis on the fuel consumption prediction model for bulk carriers from ship launching to current states based on sea trial data and machine learning technique

MARPOL 73/78, Chapter IV is considered to reveal the emission control engineering on ships. The probability model of the fuel oil consumption is established based on the machine learning technique. The proposed methods are applied into this research in order to establish the probability model of fuel oil consumption. The combination of Monte Carlo (MC) simulation method with Artificial Neural Networks (ANNs) is an optimal solution to deal the fuel consumption of marine main diesel engine. The sample data has been established based on the Monte Carlo simulation method. The model of fuel oil consumption is designed by Artificial Neural Network method. The proposed prediction model of fuel oil consumption is based on a back-propagation training algorithm of ANNs method. The research results of proposed model have been verified with the actual operation data that have been collected from a certain bulk carrier of VINIC shipping transportation company in Vietnam. The collected data is the actual operation parameters from the noon-log report of voyage during two years of the ship. The probability model of fuel oil consumption for main diesel engine is very useful in the field of ships energy efficiency management with higher accurancy than the other previous models.

Development of a Fuel Consumption Prediction Model Based on Machine Learning Using Ship In-Service Data

As interest in eco-friendly ships increases, methods for status monitoring and forecasting using in-service data from ships are being developed. Models for predicting the energy efficiency of a ship in real time need to effectively process the operational data and be optimized for such an application. This paper presents models that can predict fuel consumption using in-service data collected from a 13,000 TEU class container ship, along with statistical and domain-knowledge methods to select the proper input variables for the models. These methods prevent overfitting and multicollinearity while providing practical applicability. To implement the prediction model, either an artificial neural network (ANN) or multiple linear regression (MLR) were applied, where the ANN-based models showed the best prediction accuracy for both variable selection methods. The goodness of fit of the models based on ANN ranged from 0.9709 to 0.9936. Furthermore, sensitivity analysis of the draught under normal operating conditions indicated an optimal draught of 14.79 m, which was very close to the design draught of the target ship, and provides the optimal fuel consumption efficiency. These models could provide valuable information for ship operators to support decision making to maintain efficient operating conditions.

Modified Deep Belief Network Model and Its Application

Aiming at the problem that the traditional deep belief network (DBN) model has long training time and it is difficult to find the global optimum because of the fixed step. A modified DBN model was proposed. In the DBN pre-training phase, the adaptive step (AS) was introduced into the Contrastive Divergence (CD) algorithm to solve the problem of finding the global optimum difficulty because of the fixed step, improving the accuracy and oscillation resistance. In the fine-tuning phase, the BFGS quasi-newton method (QNM) was used to speed up the convergence and reduced the training time. The modified DBN model was applied to the fuel consumption prediction of the descending section of the aircraft. The experimental outcomes prove that compared with the traditional DBN and its variant model, the modified DBN model improves the convergence speed and the prediction accuracy.

MACHINE LEARNING METHODS IN MONITORING OPERATING BEHAVIOUR OF MARINE TWO-STROKE DIESEL ENGINE

The aim of this article is to enhance performance monitoring of a two-stroke electronically controlled ship propulsion engine on the operating envelope. This is achieved by setting up a machine learning model capable of monitoring influential operating parameters and predicting the fuel consumption. Model is tested with different machine learning algorithms, namely linear regression, multilayer perceptron, Support Vector Machines (SVM) and Random Forests (RF). Upon verification of modelling framework and analysing the results in order to improve the prediction accuracy, the best algorithm is selected based on standard evaluation metrics, i.e. Root Mean Square Error (RMSE) and Relative Absolute Error (RAE). Experimental results show that, by taking an adequate combination and processing of relevant sensory data, SVM exhibit the lowest RMSE 7.1032 and RAE 0.5313%. RF achieve the lowest RMSE 22.6137 and RAE 3.8545% in a setting when minimal number of input variables is considered, i.e. cylinder indicated pressures and propulsion engine revolutions. Further, article deals with the detection of anomalies of operating parameters, which enables the evaluation of the propulsion engine condition and the early identification of failures and deterioration. Such a time-dependent, self-adopting anomaly detection model can be used for comparison with the initial condition recorded during the test and sea run or after survey and docking. Finally, we propose a unified model structure, incorporating fuel consumption prediction and anomaly detection model with on-board decision-making process regarding navigation and maintenance.

Hybrid electric buses fuel consumption prediction based on real-world driving data

Estimating fuel consumption by hybrid diesel buses is challenging due to its diversified operations and driving cycles. In this study, long-term transit bus monitoring data were utilized to empirically compare fuel consumption of diesel and hybrid buses under various driving conditions. Artificial neural network (ANN) based high-fidelity microscopic (1 Hz) and mesoscopic (5–60 min) fuel consumption models were developed for hybrid buses. The microscopic model contained 1 Hz driving, grade, and environment variables. The mesoscopic model aggregated 1 Hz data into 5 to 60-minute traffic pattern factors and predicted average fuel consumption over its duration. The prediction results show mean absolute percentage errors of 1–2% for microscopic models and 5–8% for mesoscopic models. The data were partitioned by different driving speeds, vehicle engine demand, and road grade to investigate their impacts on prediction performance.