Neural Network-Based Electric Vehicle Range Prediction for Smart Charging Optimization

2021 ◽  
Author(s):  
Matthew J. Eagon ◽  
Daniel Kindem ◽  
Harish Panneer Selvam ◽  
William Northrop
Keyword(s):  
Fuel Consumption ◽  
Variance Estimation ◽  
Interval Estimation ◽  
Training Data ◽  
Bounded Interval ◽  
Fuel Consumption Rate ◽  
Novel Approach ◽  
Smart Charging ◽  
Physics Model ◽  
Discharge Profile

Abstract Range prediction is a standard feature in most modern road vehicles, allowing drivers to make informed decisions about when to refuel. Most vehicles make range predictions through data- or model-driven means, monitoring the average fuel consumption rate or using a tuned vehicle model to predict fuel consumption. The uncertainty of future driving conditions makes the range prediction problem challenging, particularly for less pervasive battery electric vehicles (BEV). Most contemporary machine learning-based methods attempt to forecast the battery SOC discharge profile to predict vehicle range. In this work, we propose a novel approach using two recurrent neural networks (RNNs) to predict the remaining range of BEVs and the minimum charge required to safely complete a trip. Each RNN has two outputs which can be used for statistical analysis to account for uncertainties; the first loss function leads to mean and variance estimation (MVE), while the second results in bounded interval estimation (BIE). These outputs of the proposed RNNs are then used to predict the probability of a vehicle completing a given trip without charging, or if charging is needed, the remaining range and minimum charging required to finish the trip with high probability. Training data was generated using a low-order physics model to estimate vehicle energy consumption from historical drive cycle data collected from medium-duty last-mile delivery vehicles. The proposed method demonstrated high accuracy in the presence of day-to-day route variability, with the root-mean-square error (RMSE) below 6% for both RNN models.

scholarly journals Autonomous flight cycles and extreme landings of airliners beyond the current limits and capabilities using artificial neural networks

2021 ◽  
Author(s):  
Haitham Baomar ◽  
Peter J. Bentley
Keyword(s):  
Neural Networks ◽  
Artificial Neural Networks ◽  
Weather Conditions ◽  
Extreme Weather ◽  
Training Data ◽  
Aviation Industry ◽  
Control Models ◽  
Novel Approach ◽  
Network Modules ◽  
Artificial Neural

AbstractWe describe the Intelligent Autopilot System (IAS), a fully autonomous autopilot capable of piloting large jets such as airliners by learning from experienced human pilots using Artificial Neural Networks. The IAS is capable of autonomously executing the required piloting tasks and handling the different flight phases to fly an aircraft from one airport to another including takeoff, climb, cruise, navigate, descent, approach, and land in simulation. In addition, the IAS is capable of autonomously landing large jets in the presence of extreme weather conditions including severe crosswind, gust, wind shear, and turbulence. The IAS is a potential solution to the limitations and robustness problems of modern autopilots such as the inability to execute complete flights, the inability to handle extreme weather conditions especially during approach and landing where the aircraft’s speed is relatively low, and the uncertainty factor is high, and the pilots shortage problem compared to the increasing aircraft demand. In this paper, we present the work done by collaborating with the aviation industry to provide training data for the IAS to learn from. The training data is used by Artificial Neural Networks to generate control models automatically. The control models imitate the skills of the human pilot when executing all the piloting tasks required to pilot an aircraft between two airports. In addition, we introduce new ANNs trained to control the aircraft’s elevators, elevators’ trim, throttle, flaps, and new ailerons and rudder ANNs to counter the effects of extreme weather conditions and land safely. Experiments show that small datasets containing single demonstrations are sufficient to train the IAS and achieve excellent performance by using clearly separable and traceable neural network modules which eliminate the black-box problem of large Artificial Intelligence methods such as Deep Learning. In addition, experiments show that the IAS can handle landing in extreme weather conditions beyond the capabilities of modern autopilots and even experienced human pilots. The proposed IAS is a novel approach towards achieving full control autonomy of large jets using ANN models that match the skills and abilities of experienced human pilots and beyond.

scholarly journals Removal of Cochannel Interference Using Probabilistic Latent Component Analysis in Passive Bistatic Radar

2021 ◽  
Vol 2021 ◽  
pp. 1-10
Author(s):  
Tao Ying ◽  
Xuebao Wang ◽  
Wei Tian ◽  
Cheng Zhou
Keyword(s):  
Target Detection ◽  
Reference Signal ◽  
Component Analysis ◽  
Cochannel Interference ◽  
Training Data ◽  
Bistatic Radar ◽  
Frequency Channel ◽  
Novel Approach ◽  
Passive Bistatic Radar

This paper examines the problem of cancellation of cochannel interference (CCI) present in the same frequency channel as the signal of interest, which may bring a reduction in the performance of target detection, in passive bistatic radar. We propose a novel approach based on probabilistic latent component analysis for CCI removal. The highlight is that removing CCI is considered as reconstruction, and extraction of Doppler-shifted and time-delayed replicas of the reference signal exploited fully as training data. The results of the simulation show that the developed method is effective.

scholarly journals Design of new clean and efficient combustion mode and thermodynamics research using NSGA-II algorithm

2020 ◽  
Vol 24 (5 Part A) ◽  
pp. 2699-2706
Author(s):  
Guoqing Shen
Keyword(s):  
Natural Gas ◽  
Combustion Chamber ◽  
Fuel Consumption ◽  
Consumption Rate ◽  
Thermodynamic Characteristics ◽  
Dual Fuel ◽  
Combustion Mode ◽  
Nsga Ii ◽  
Fuel Consumption Rate ◽  
Efficient Combustion

In order to study a new clean and efficient combustion mode, which can relieve the pressure of traditional energy and ensure low emissions, in this study, a diesel/natural gas dual fuel engine is designed by non-dominant sorting genetic algorithm (NSGA-?), and its thermodynamic characteristics are studied. The WP10.290 Diesel engine is modified into a diesel/natural gas dual fuel engine. The emissions of harmful substances and thermal efficiency of the modified engine under different working conditions are compared. The combustion chamber structure and adaptability between combustion chamber and injection parameters are optimized by using NSGA-II algorithm and CFD software. The results show that the emission of NOx and CH4 and the fuel consumption rate can be reduced simultaneously by using the composite combustion model compared with the original engine. When the CH4 emission is close to zero, the fuel consumption rate decreases obviously, and NOx slightly increases. When the angle between the injection holes is 141.57? the amount of NOx in the cylinder is large. When the injection advance angle is 21.91?CA, the pressure in the cylinder is the highest, the CH4 production is the lowest, the NOx production is higher, and the oxygen content in the combustion mixture is less. The NOx production is the lowest. diesel/natural gas dual fuel engine can ensure efficient combustion while reducing emissions. In this study, the performance of the dual fuel engine at various speeds can be further studied, which can provide theoretical support for the design of diesel/natural gas dual fuel engine.

scholarly journals Discovering Latent Class Labels for Multi-Label Learning

2020 ◽  
Author(s):  
Jun Huang ◽  
Linchuan Xu ◽  
Jing Wang ◽  
Lei Feng ◽  
Kenji Yamanishi
Keyword(s):  
Large Scale ◽  
Latent Class ◽  
Training Data ◽  
Data Sets ◽  
Robust Learning ◽  
Large Scale Data ◽  
Novel Approach ◽  
Fixed Set ◽  
Class Labels ◽  
Scale Data

Existing multi-label learning (MLL) approaches mainly assume all the labels are observed and construct classification models with a fixed set of target labels (known labels). However, in some real applications, multiple latent labels may exist outside this set and hide in the data, especially for large-scale data sets. Discovering and exploring the latent labels hidden in the data may not only find interesting knowledge but also help us to build a more robust learning model. In this paper, a novel approach named DLCL (i.e., Discovering Latent Class Labels for MLL) is proposed which can not only discover the latent labels in the training data but also predict new instances with the latent and known labels simultaneously. Extensive experiments show a competitive performance of DLCL against other state-of-the-art MLL approaches.

Performance of biomass combustor based drying system for ginger drying

2021 ◽  
Vol 45 (01) ◽  
pp. 19-25
Author(s):  
D. K. Vyas ◽  
N. Seth ◽  
J. J. Chavda
Keyword(s):  
Fuel Consumption ◽  
Flow Rate ◽  
Consumption Rate ◽  
Air Flow ◽  
Air Flow Rate ◽  
Saw Dust ◽  
Fuel Consumption Rate ◽  
Hot Air ◽  
Drying System ◽  
Maize Cobs

A biomass combustor based dryer was evaluated with different biomass for drying of ginger. Biomass combustor based dryer consists of fuel hopper, combustion chamber, heat exchanger, grate for proper combustion of the combustible gas, chimney, ambient air inlet, hot air outlet and drying chamber. The system was evaluated at five fuel consumption rate (1 to 5 kg.h–1) and five air flow rate (100, 150, 200, 300 and 400 m3.h–1) using maize cobs, sized wood and saw dust briquettes for ginger drying. The experimental performances show that the hot air temperature inside the dryer vary between 36 to 81ºC for maize cobs, 53 to 85ºC for sized wood and 49 to 87ºC for biomass briquettes at tested air flow rate and fuel consumption rate in the system. The maximum efficiency of the system was found at the fuel consumption rate of 1 kg.h–1 and 400 m3.h–1 air flow rate using maize cobs, sized wood and saw dust briquettes as fuel respectively. The cost of operation of ginger drying at 1 kg.h–1 fuel consumption rate and 400 m3/h air flow rate was Rs. 32.76, 34.26, 34.76 and 55 per hour using maize cobs, sized wood, saw dust briquettes and mechanical drying system, respectively. Hence, the drying of ginger in biomass combustor based dryer using maize cobs at 1 kg.h–1 fuel consumption rate and 400 m3/h air flow rate resulted in better performance.

Prediction of Gas Turbine Trip: a Novel Methodology Based on Random Forest Models

2021 ◽  
Author(s):  
Enzo Losi ◽  
Mauro Venturini ◽  
Lucrezia Manservigi ◽  
Giuseppe Fabio Ceschini ◽  
Giovanni Bechini ◽  
...  
Keyword(s):  
Machine Learning ◽  
Random Forest ◽  
Gas Turbine ◽  
Gas Turbines ◽  
Remaining Useful Life ◽  
Training Data ◽  
The Novel ◽  
Novel Approach ◽  
Forest Models ◽  
Random Forest Models

Abstract A gas turbine trip is an unplanned shutdown, of which the most relevant consequences are business interruption and a reduction of equipment remaining useful life. Thus, understanding the underlying causes of gas turbine trip would allow predicting its occurrence in order to maximize gas turbine profitability and improve its availability. In the ever competitive Oil & Gas sector, data mining and machine learning are increasingly being employed to support a deeper insight and improved operation of gas turbines. Among the various machine learning tools, Random Forests are an ensemble learning method consisting of an aggregation of decision tree classifiers. This paper presents a novel methodology aimed at exploiting information embedded in the data and develops Random Forest models, aimed at predicting gas turbine trip based on information gathered during a timeframe of historical data acquired from multiple sensors. The novel approach exploits time series segmentation to increase the amount of training data, thus reducing overfitting. First, data are transformed according to a feature engineering methodology developed in a separate work by the same authors. Then, Random Forest models are trained and tested on unseen observations to demonstrate the benefits of the novel approach. The superiority of the novel approach is proved by considering two real-word case-studies, involving filed data taken during three years of operation of two fleets of Siemens gas turbines located in different regions. The novel methodology allows values of Precision, Recall and Accuracy in the range 75–85 %, thus demonstrating the industrial feasibility of the predictive methodology.

scholarly journals Experimental Analysis of Calculation of Fuel Consumption Rate by On-Road Mileage in a 2.0 L Gasoline-Fueled Passenger Vehicle

2018 ◽  
Vol 8 (12) ◽  
pp. 2390 ◽  
Author(s):  
Jaehyuk Lim ◽  
Yumin Lee ◽  
Kiho Kim ◽  
Jinwook Lee
Keyword(s):  
Fuel Consumption ◽  
Fuel Economy ◽  
Fuel Efficiency ◽  
Co2 Concentration ◽  
The United States ◽  
Traffic Conditions ◽  
Fuel Consumption Rate ◽  
Driving Cycles ◽  
The Difference ◽  
Consumption Value

The five-driving test mode is vehicle driving cycles made by the Environment Protection Association (EPA) in the United States of America (U.S.A.) to fully reflect actual driving environments. Recently, fuel consumption value calculated from the adjusted fuel consumption formula has been more effective in reducing the difference from that experienced in real-world driving conditions, than the official fuel efficiency equation used in the past that only considered the driving environment included in FTP and HWFET cycles. There are many factors that bring about divergence between official fuel consumption and that experienced by drivers, such as driving pattern behavior, accumulated mileage, driving environment, and traffic conditions. In this study, we focused on the factor of causing change of fuel efficiency value, calculated according to how many environmental conditions that appear on the real-road are considered, in producing the fuel consumption formula, and that of the vehicle’s accumulated mileage in a 2.0 L gasoline-fueled vehicle. So, the goals of this research are divided into four major areas to investigate divergence in fuel efficiency obtained from different equations, and what factors and how much CO2 and CO emissions that are closely correlated to fuel efficiency change, depending on the cumulative mileage of the vehicle. First, the fuel consumption value calculated from the non-adjusted formula, was compared with that calculated from the corrected fuel consumption formula. Also, how much CO2 concentration levels change as measured during each of the three driving cycles was analyzed as the vehicle ages. In addition, since the US06 driving cycle is divided into city mode and highway mode, how much CO2 and CO production levels change as the engine ages during acceleration periods in each mode was investigated. Finally, the empirical formula was constructed using fuel economy values obtained when the test vehicle reached 6500 km, 15,000 km, and 30,000 km cumulative mileage, to predict how much fuel consumption of city and highway would worsen, when mileage of the vehicle is increased further. When cumulative mileage values set in this study were reached, experiments were performed by placing the vehicle on a chassis dynamometer, in compliance with the carbon balance method. A key result of this study is that fuel economy is affected by various fuel consumption formula, as well as by aging of the engine. In particular, with aging aspects, the effect of an aging engine on fuel efficiency is insignificant, depending on the load and driving situation.

Sign in / Sign up

Export Citation Format

Share Document