Search results

The purpose of this study is to develop and test a new deep learning model to predict aircraft fuel consumption. For this purpose, real data obtained from different landings and take-offs were used. As a result, a new hybrid convolutional neural network (CNN)-bi-directional long short term memory (BiLSTM) model was developed as intended.

The data used are divided into training and testing according to the k-fold 5 value. In this study, 13 different parameters were used together as input parameters. Fuel consumption was used as the output parameter. Thus, the effect of many input parameters on fuel flow was modeled simultaneously using the deep learning method in this study. In addition, the developed hybrid model was compared with the existing deep learning models long short term memory (LSTM) and BiLSTM.

In this study, when tested with LSTM, one of the existing deep learning models, values of 0.9162, 6.476, and 5.76 were obtained for R², root mean square error (RMSE), and mean absolute percentage error (MAPE), respectively. For the BiLSTM model when tested, values of 0.9471, 5.847 and 4.62 were obtained for R², RMSE and MAPE, respectively. In the proposed hybrid model when tested, values of 0.9743, 2.539 and 1.62 were obtained for R², RMSE and MAPE, respectively. The results obtained according to the LSTM and BiLSTM models are much closer to the actual fuel consumption values. The error of the models used was verified against the actual fuel flow reports, and an average absolute percent error value of less than 2% was obtained.

In this study, a new hybrid CNN-BiLSTM model is proposed. The proposed model is trained and tested with real flight data for fuel consumption estimation. As a result of the test, it is seen that it gives much better results than the LSTM and BiLSTM methods found in the literature. For this reason, it can be used in many different engine types and applications in different fields, especially the turboprop engine used in the study. Because it can be applied to different engines than the engine type used in the study, it can be easily integrated into many simulation models.

One of the challenges encountered in the design of guided projectiles is their prohibitive cost. To diminish it, an appropriate avenue many researchers have explored is the use of the non-actuator method for guiding the projectile to the target. In this method, biologically inspired by the flying concept of the single-winged seed, for instance, that of maple and ash trees, the projectile undergoes a helical motion to scan the region and meet the target in the descent phase. Indeed, the projectile is a decelerator device based on the autorotation flight while it attempts to resemble the seed’s motion using two wings of different spans. There exists a wealth of studies on the stability of the decelerators (e.g. the mono-wing, samara and pararotor), but all of them have assumed the body (exclusive of the wing) to be symmetric and paid no particular attention to the scanning quality of the region. In practice, however, the non-actuator-guided projectiles are asymmetric owing to the presence of detection sensors. This paper aims to present an analytical solution for stability analysis of asymmetric decelerators and apprise the effects of design parameters to improve the scanning quality.

The approach of this study is to develop a theoretical model consisting of Euler equations and apply a set of non-dimensionalized equations to reduce the number of involved parameters. The obtained governing equations are readily applicable to other decelerator devices, such as the mono-wing, samara and pararotor.

The results show that the stability of the body can be preserved under certain conditions. Moreover, pertinent conclusions are outlined on the sensitivity of flight behavior to the variation of design parameters.

The analytical solution and sensitivity analysis presented here can efficiently reduce the design cost of the asymmetric decelerator.

This article aims to examine the role of capital assets in rural household poverty transitions of poverty escape and poverty descent over periods of 2014–2016 and 2016–2018.

Based on the sustainable livelihood approach, this paper uses binary logit model to explore the influence of multidimensional capital assets on poverty transitions and use instrumental variable estimation to solve the endogeneity between total net asset and poverty transitions.

Capital assets have significant impacts on household poverty transitions. The role of capital assets in households' poverty escape and poverty descent are not symmetrical. The authors verify that rural households with rich total net asset are more likely to escape poverty and less likely to descend into poverty by using instrumental variable estimation. The authors verify that there is a mediation effect that total net asset can help households' escaping poverty and prevent them from falling into poverty through promoting rural households to engage in business activities.

This paper is the first to explore how capital assets affect poverty transitions in rural China based on the sustainable livelihood approach. The findings of this research can provide valuable policy implications for the pursuit of common prosperity in China and references for other developing countries.

To authenticate the existence and principles of the adhesion recovery phenomenon under water pollution conditions, an innovative circumferential rail–wheel adhesion test rig was used. The study conducted extensive tests on the adhesion characteristics under large sliding conditions.

Experiments were conducted to investigate the influence of speed, axle load and slip on adhesion recovery. Based on the experimental results, the adhesion recovery transition function was re-fitted.

The study reveals that the adhesion recovery phenomenon truly exists under water conditions. The adhesion coefficient shows an increasing trend with the growth of the slip ratio. Moreover, at the current speed and axle load levels, the adhesion recovery is directly proportional to the square of the slip ratio and inversely proportional to the axle load.

The phenomenon of adhesion recovery and the formulated equations in this study can serve as an experimental and theoretical foundation for the design of braking and anti-skid control algorithms for trains.

The peer review history for this article is available at: https://publons.com/publon/10.1108/ILT-11-2023-0379/

The purpose of this paper is to examine recovery through lived experience. It is part of a series that explores candid accounts of addiction and recovery to identify important components in the recovery process.

The G-CHIME model comprises six elements important to addiction recovery (growth, connectedness, hope, identity, meaning in life and empowerment). It provides a standard against which to consider addiction recovery, having been used in this series, as well as in the design of interventions that improve well-being and strengthen recovery. In this paper, a first-hand account is presented, followed by a semi-structured e-interview with the author of the account. Narrative analysis is used to explore the account and interview through the G-CHIME model.

This paper shows that addiction recovery is a remarkable process that can be effectively explained using the G-CHIME model. The significance of each component in the model is apparent from the account and e-interview presented.

Each account of recovery in this series is unique and, as yet, untold.

To ensure the motion attitude and stable contact force of massage robot working on unknown human tissue environment, this study aims to propose a robotic system for autonomous massage path planning and stable interaction control.

First, back region extraction and acupoint recognition based on deep learning is proposed, which provides a basis for determining the working area and path points of the robot. Second, to realize the standard approach and movement trajectory of the expert massage, 3D reconstruction and path planning of the massage area are performed, and normal vectors are calculated to control the normal orientation of robot-end. Finally, to cope with the soft and hard changes of human tissue state and body movement, an adaptive force tracking control strategy is presented to compensate the uncertainty of environmental position and tissue hardness online.

Improved network model can accomplish the acupoint recognition task with a large accuracy and integrate the point cloud to generate massage trajectories adapted to the shape of the human body. Experimental results show that the adaptive force tracking control can obtain a relatively smooth force, and the error is basically within ± 0.2 N during the online experiment.

This paper incorporates deep learning, 3D reconstruction and impedance control, the robot can understand the shape features of the massage area and adapt its planning massage path to carry out a stable and safe force tracking control during dynamic robot–human contact.

GraphQL is a new Open API specification that allows clients to send queries and obtain data flexibly according to their needs. However, a high-complexity GraphQL query may lead to an excessive data volume of the query result, which causes problems such as resource overload of the API server. Therefore, this paper aims to address this issue by predicting the response data volume of a GraphQL query statement.

This paper proposes a GraphQL response data volume prediction approach based on Code2Vec and AutoML. First, a GraphQL query statement is transformed into a path collection of an abstract syntax tree based on the idea of Code2Vec, and then the query is aggregated into a vector with the fixed length. Finally, the response result data volume is predicted by a fully connected neural network. To further improve the prediction accuracy, the prediction results of embedded features are combined with the field features and summary features of the query statement to predict the final response data volume by the AutoML model.

Experiments on two public GraphQL API data sets, GitHub and Yelp, show that the accuracy of the proposed approach is 15.85% and 50.31% higher than existing GraphQL response volume prediction approaches based on machine learning techniques, respectively.

This paper proposes an approach that combines Code2Vec and AutoML for GraphQL query response data volume prediction with higher accuracy.

The urban environment is perceived through multiple senses in parallel, which means that visual understanding of space is aided and complemented by auditory, basic-orienting, and haptic stimuli – although mainly unconsciously. Sensory conditions are inherent attributes of urban places, but are often overlooked in research. To include these aspects in any way in analysis of the urban landscape, they need to be understood as properties of urban space, to be translated from attributes of the perceiver to attributes of the perceived. Using the relation between a designed garden and its suburban context in Bad Oeynhausen (DE) as an example, I will explore an alternative analytical methodology that takes the first-hand perspective view of the subject moving through the city as the starting point. The human body explores space by moving through it; walking is the most direct way to access, study, and research the physical qualities of the (urban) landscape, involving not only visual experience but also sound, rhythm, kinaesthesia, balance, and so forth. A notation technique that discloses the interrelation between visual qualities and their perception over time is the technique of ‘scoring’. Scores are symbolisations of processes, which extend over time. They can objectively represent non-visual qualities of space, communicating the relation between such processes and their spatial context to others in other places and other moments. These representations of movement expose the qualities of the surroundings that change as one moves through them, thus communicating the experiential aspects of urban landscape.