Advanced searches left 3/3

Fuel Efficiency - DOAJ

Summarized by Plex Scholar
Last Updated: 15 December 2022

* If you want to update the article please login/register

Achieving fuel efficiency of harbour craft vessel via combined time-series and classification machine learning model with operational data

This paper discusses recent work on forecasting the fuel consumption rate of a harbour craft vessel through the combination time series and classification prediction models. The noisy readings in the fuel flow rate measurements are reduced by the Haar wavelet transform filters. Wind data is converted into wind effect, and the vessel speed is increased by transforming GPS coordinates of vessel location to vessel distance traveled over time. The k-means clustering groups received operational data from the same operations for the design of the classification model, which was later published. Both the time-series and classification algorithms are used in parallel to produce prediction results.

Source link: https://doi.org/10.1016/j.martra.2022.100073


Data fusion and machine learning for ship fuel efficiency modeling: Part II – Voyage report data, AIS data and meteorological data

When voyage report data is used as the primary data source for ship fuel efficiency analysis, it is often unreliable, especially when voyage report results are used as the main data source for ship fuel efficiency analysis. These more reliable data about weather and sea conditions that ship sails through is mixed with voyage report results in order to improve ship fuel consumption rate estimates' accuracy. Experimental results showed the benefits of combining voyage report data, AIS data, and meteorological results in improving the fit of machine learning models of forecasting ship fuel consumption rate. These results are marginally better than our previous analysis, which shows the benefits of adopting the actual geographic locations of the ship identified by AIS data, relative to the estimated geographical positions derived from the great circle route's estimates, in retrieving weather and sea conditions the ship sails through.

Source link: https://doi.org/10.1016/j.commtr.2022.100073


Data fusion and machine learning for ship fuel efficiency modeling: Part I – Voyage report data and meteorological data

The theoretical basis of these recommendations is a model that can accurately forecast a ship's bunker fuel consumption rate based on its sailing speed, displacement/draft, trim, weather, and sea conditions. The Voyage survey is a vital data source for ship fuel efficiency modeling, but its information quality on weather and sea conditions is limited by a snapshotting approach with eye inspection. This research develops nine datasets based on this data fusion technology to help overcome this problem. This research provides a way to fuse voyage report results and publicly available meteorological data in this region, and builds nine datasets based on this data fusion technique in order to solve this problem. The best datasets discovered reveal the benefits of fusing voyage report results and meteorological results, as well as the more widely accepted quality of voyage report results. Their fit errors on daily bunker fuel use are usually between 0. 5 and 4. 0 ton/day. These models have a good deal of indicating the relative importance of various determinants to a ship's fuel consumption rate. These models have a good interpretability in expressing the relative importance of various determinants to a ship's fuel consumption rate.

Source link: https://doi.org/10.1016/j.commtr.2022.100074


Data fusion and machine learning for ship fuel efficiency modeling: Part III – Sensor data and meteorological data

Sensors installed on a ship provide high quality results that can be used for ship bunker fuel efficiency analysis. The benefits of fusing sensor data and meteorological data for ship fuel consumption rate quantification can be demonstrated by the best dataset found. Given the best data from data fusion, their R2 values over the training set are 0. 999 or 1. 000, with R2 values for the test set all above 0. 966. These promising results go well beyond the expectations of most industry applications for ship fuel efficiency analysis. The applicability of the selected datasets and ML models is also tested in a rolling horizon approach, resulting in a declaration that a rolling horizon strategy of u201c5-month training + 1-month test/applicatoinu201d could be effective in practice, but sensor results of fewer than five months may be insufficient to develop ML models.

Source link: https://doi.org/10.1016/j.commtr.2022.100072


Optimizing Fuel Efficiency on an Islanded Microgrid under Varying Loads

Microgrids have been studied in the past and are based on measurements of fuel use by generators under static load. We design a mixed-integer linear optimization scheme to schedule generator and fuel storage system operation to satisfy known demands in order to help determine the effects of time-varying loads on optimal generator performance and fuel consumption. Regardless of the imposed penalty placed on the generator, our results show that the change in fuel efficiency between scenarios with the introduction of ESS is minimal. Because the ESS helps the generator reduce power output fluctuation, the generator can reduce power output fluctuation, the ESS helps reduce fuel consumption.

Source link: https://doi.org/10.3390/en15217943


Eco-Driving and Its Impacts on Fuel Efficiency: An Overview of Technologies and Data-Driven Methods

Eco-driving is a multidimensional term that encompasses driving habits, route selection, and any other choices or behaviours related to the vehicles'u2019 fuel consumption. The results reveal a series of shortcomings and challenges that will be further explored in the context of system wide implementations of deriving policies that raise driver awareness while also improving system quality.

Source link: https://doi.org/10.3390/su13010226

* Please keep in mind that all text is summarized by machine, we do not bear any responsibility, and you should always check original source before taking any actions

* Please keep in mind that all text is summarized by machine, we do not bear any responsibility, and you should always check original source before taking any actions