Search Results

In this paper, methods of identifying the structural features of fibers and cracks in GDL images based on artificial intelligence are proposed. The block probabilistic Hough transform and the quadric voting based on the weighted K-means algorithm are programmed to realize the fiber feature extraction, and the crack feature extraction is realized by the regional connectivity algorithm and the geometric feature calculation based on the circumscribed graph of the crack region. ...The image processing technology based on artificial intelligence can capture the microstructural features of GDL images and extract feature parameters, which provides a reliable tool for GDL image analysis and has guiding significance for further research on GDL.

Therefore, we intend to make a step forward with these data-driven artificial intelligence algorithms. We applied four data-driven artificial intelligence algorithms to diagnose three common faults of PEMFC (each fault type has two severity levels, slight and severe). ...With the development of artificial intelligence, performing fault diagnosis with the massive sampling data of the fuel cell system has become a popular research topic. ...But few people have successfully verified the diagnosis performance of these artificial intelligence algorithms on a real high power on-board PEMFC system. Therefore, we intend to make a step forward with these data-driven artificial intelligence algorithms.

Abstract Artificial intelligence-based computing systems like artificial neural networks (ANN) have recently found increasing applications in predicting complex chemical phenomena like combustion properties.

This paper describes the applicable and capability of neural network as an artificial intelligence tool to determine the performance and emissions in a compressed natural gas direct injection (CNG-DI) engine. ...A feed-forward back-propagation artificial neural network (BPANN) approach is explored to predict the combustion performance in the term of indicated power and emissions in the appearance of CO and NO emissions level. ...The data for combustion process under various engine operating parameters at the fixed speed at 1000 rpm were obtained to train the developed artificial neural network (ANN). The operating conditions employed to represent the combustion parameters for controlling the injection and ignition event are start of injection (SOI), end of injection (EOI) and spark advance (SA) timing, which affects to the combustion processes, performance as well as emissions formation.

This paper describes the application and capability of neural network as an artificial intelligence tool to determine the performance and emissions in a compressed natural gas direct injection (CNG-DI) engine. ...A feed-forward back-propagation artificial neural network (BPANN) approach is explored to predict the combustion performance in terms of indicated power and emissions in the appearance of CO and NO emissions level. ...The data for combustion process under various engine operating parameters at the fixed speed at 1000 rpm were obtained to train the developed artificial neural network (ANN). The operating conditions employed to represent the combustion parameters for controlling the injection and ignition event are start of injection (SOI), end of injection (EOI) and spark advance (SA) timing, which affects to the combustion processes, performance as well as emissions formation.

The modelling methods that are used in this context are artificial neural networks. The key parameters for modelling of in-cylinder pressure are identified as inputs and a network structure is trained with data generated from an engine model that is validated against data from a real medium-duty diesel engine.

In the present study, this new approach is investigated by loading a bare filter substrate with submicron alumina particles generated by an atomizer to fabricate an “artificial ash” coating. The substrate backpressure showed only a mild increase after loading with artificial ash. ...The substrate backpressure showed only a mild increase after loading with artificial ash. The FTP weighted average filtration efficiency increased from ~75% for a blank substrate to ~90% after loading with 1.5 g/L of artificial ash. ...The FTP weighted average filtration efficiency increased from ~75% for a blank substrate to ~90% after loading with 1.5 g/L of artificial ash. Tests over the WLTC with the ash coated filter showed much reduced soot emission during the cold start and nearly negligible emissions thereafter.

This new control approach uses an artificial neural network to replace the conventional multiple mode approach. The desired engine emission and temperature limits, for example based on SCR conversion efficiency, are sent to the artificial neural network, which controls the actuators to meet the desired limits. ...The desired engine emission and temperature limits, for example based on SCR conversion efficiency, are sent to the artificial neural network, which controls the actuators to meet the desired limits. It also enables the lowest possible fuel consumption or, for example, the highest possible exhaust gas enthalpy within the given system state. ...Since this method allows semi-automatic calibration and training of the artificial neural network, the manual calibration effort can be reduced compared to the conventional approach as an additional advantage.

With the information obtained, a Dynamic Artificial Neural Network is trained and modelling of the vehicle emissions becomes possible.

A well-known auto-ignition model for gasoline, which was proposed by Halstead et al, is automatically optimized on computers by using a simple artificial brain including genetic algorithm as learning theory and an intuition model. Arbitrary constants inside the mathematical equations of highly-nonlinear chemical reaction processes can be fitted by using the experimental time-evolutions of several components.

This work introduces a new kind of model to calculate thermal behavior of combustion engines using an artificial neural network (ANN) which is highly accurate and extremely fast. The neural network is a multi-layered feed-forward network; it is trained by data generated with a validated semi-physical model. ...The training of the artificial neural network is described and the quality of the training is shown based on several statistical parameters.

The purpose of the current work is to overcome these limitations and to present how Artificial Neural Networks expand the capability of utilizing engine speed signal for fuel identification by using main combustion characteristics such as firing peak cylinder pressure and peak pressure rise rate.

A methodology is proposed to 1) choose in-cycle features of the pressure trace that highly correlate with knock events and 2) train artificial neural networks to predict in-cycle knock events before knock onset. The methodology was validated at different operating conditions and different levels of generalization.

This paper will describe the Army's Vehicle Intelligence Program and discuss some of the VI technologies being considered for use within the Army's Tactical Wheeled Vehicle fleet.

In this article, CFRP and metal body vehicles were tested under artificial lightning. The electric discharging caused by the artificial lightning in the vehicles was investigated under different grounding conditions. ...A CFRP roof plate and a CFRP box mimicking vehicle cabin were also examined with artificial lightning to study generic cases, which did not depend on vehicle body shapes. The comparative study showed no significant difference between the CFRP and metal vehicles in lighting-strike performance.

The objective of this paper is the prediction of efficiency and NOx emission of a Spark Ignition engine based on engine design and operational parameters using artificial neural networks (ANN). This paper deals with quasi-dimensional, two-zone thermodynamic simulation of four-stroke SI engine fueled with biogas. ...For real time computations in electronic control unit (ECU) an artificial neural network (ANN) model has been suggested as an alternative to the engine simulation model.

Thus, a new concepts of operating a direct injection diesel engine with artificial air intake, which is composed of oxygen, nitrogen, water vapor, carbon dioxide, and argon, are investigated both theoretically and experimentally.

This study focuses on analyzing the path planning problem for intelligent vehicles in complex terrains by utilizing the optimization evaluation function of the artificial bee colony (ABC) algorithm. Additionally, the impact of turning radius at different speeds is considered during the planning process.

Towards this end artificial neural-network (ANN) technique using Levenberg-Marquardt (LM) training algorithm is used to train and optimize the fuel injection timing of a single cylinder, four-stroke gasoline engine.

To optimize the flow performance of intake port, a new optimization method combining genetic algorithm (GA) and artificial neural network (ANN) was proposed. First, an automatic system for generating the geometry of the tangential intake port was constructed to create various port geometries through inputting the 18 pre-defined structural parameters.