Search results

An abundance of techniques has been presented so forth for waste classification but, they deliver inefficient results with low accuracy. Their achievement on various repositories is different and also, there is insufficiency of high-scale databases for training. The purpose of the study is to provide high security.

In this research, optimization-assisted federated learning (FL) is introduced for thermoplastic waste segregation and classification. The deep learning (DL) network trained by Archimedes Henry gas solubility optimization (AHGSO) is used for the classification of plastic and resin types. The deep quantum neural networks (DQNN) is used for first-level classification and the deep max-out network (DMN) is employed for second-level classification. This developed AHGSO is obtained by blending the features of Archimedes optimization algorithm (AOA) and Henry gas solubility optimization (HGSO). The entities included in this approach are nodes and servers. Local training is carried out depending on local data and updations to the server are performed. Then, the model is aggregated at the server. Thereafter, each node downloads the global model and the update training is executed depending on the downloaded global and the local model till it achieves the satisfied condition. Finally, local update and aggregation at the server is altered based on the average method. The Data tag suite (DATS_2022) dataset is used for multilevel thermoplastic waste segregation and classification.

By using the DQNN in first-level classification the designed optimization-assisted FL has gained an accuracy of 0.930, mean average precision (MAP) of 0.933, false positive rate (FPR) of 0.213, loss function of 0.211, mean square error (MSE) of 0.328 and root mean square error (RMSE) of 0.572. In the second level classification, by using DMN the accuracy, MAP, FPR, loss function, MSE and RMSE are 0.932, 0.935, 0.093, 0.068, 0.303 and 0.551.

The multilevel thermoplastic waste segregation and classification using the proposed model is accurate and improves the effectiveness of the classification.

This study aims to present a numerical solution for the analysis of the influence of surface roughness as presented by a sinusoidal ripple of different amplitude and wavelength on the performance of transient elastohydrodynamic lubrication at motion start-up under different operational parameters of entraining speed and load as well as different acceleration rates.

A statistical asperity micro-contact model represented by a sinusoidal ripple expressed by two parameters (wavelength and undeformed amplitude) is considered. The ball equation of motion is used to calculate the force on the ball as it starts to move. The time-dependent Reynolds equation is solved together with surface deformation and statistical asperity models using the Newton–Raphson technique with the Gauss–Seidel iteration method.

The behaviour of the film thickness was found to be strongly influenced by the acceleration rate for different ripple amplitude and wavelength parameters. The effect of increasing the final entraining speed will eventually lead to rapid film thickness build-up and increase the film thickness jump at the moment of motion start-up. The effect of increasing applied load is to reduce the deviation of the minimum film thickness jump at the start-up of motion, making its value approximately equal to the steady-state value over the entire run-time period.

Influence of surface roughness for various wavelength and undeformed amplitude on the performance of transient elastohydrodynamic lubrication at motion start-up is presented at different acceleration rates as well as for different operating parameters of entraining speed and load. Ball equation of motion is used to calculate the force on the ball as it starts to move.

This paper aims to propose a learning exponential jerk trajectory planning to suppress the residual vibrations of industrial robots.

Based on finite impulse response filter technology, a step signal with a proper amplitude first passes through two linear filters and then performs exponential filter shaping to obtain an exponential jerk trajectory and cancel oscillation modal. An iterative learning strategy designed by gradient descent principle is used to adjust the parameters of exponential filter online and achieve the maximum vibration suppression effect.

By building a SCARA robot experiment platform, a series of contrast experiments are conducted. The results show that the proposed method can effectively suppress residual vibration compared to zero vibration shaper and zero vibration and derivative shaper.

The idea of the adopted iterative leaning strategy is simple and reduces the computing power of the controller. A cheap acceleration sensor is available because it just needs to measure vibration energy to feedback. Therefore, the proposed method can be applied to production practice.

This paper aims to study the startup performance of thrust bearing. The effects of acceleration scenarios, roughness, the area ratio of texture and texture depth on the transient startup performance of the thrust bearing were analyzed.

The lubrication model is solved by the Reynolds equation with the mass-conservation boundary condition. The Greenwood and Tripp contact model is used to predict asperity contact load. The finite volume method is used to discretize the governing equations.

By studying the bearing performance with different acceleration functions, it was found that the higher the acceleration at the beginning of the startup, the faster the thrust bearing operates under the hydrodynamic lubrication regime in the start stage. It appears that the friction and contact time of asperity increase with the increasing roughness. The optimal area ratio of texture is within 30%–50%. The depth of texture ranging from 1 to 2 is the best.

This paper proposes a transient mixed lubrication analysis model of the thrust bearing. This model can be used to analyze the variations of tribological performance and lubrication regime of the thrust bearing under different acceleration scenarios.

The peer review history for this article is available at: https://publons.com/publon/10.1108/ILT-09-2022-0268/

This paper aims to propose an online local trajectory planner for safe and fast trajectory generation that combines the jerk-limited trajectory (JLT) generation algorithm and the particle swarm optimization (PSO) algorithm. A trajectory switching algorithm is proposed to improve the trajectory tracking performance. The proposed system generates smooth and safe flight trajectories online for quadrotors.

First, the PSO algorithm method can obtain the optimal set of target points near the path points obtained by the global path searching. The JLT generation algorithm generates multiple trajectories from the current position to the target points that conform to the kinetic constraints. Then, the generated multiple trajectories are evaluated to pick the obstacle-free trajectory with the least cost. A trajectory switching strategy is proposed to switch the unmanned aerial vehicle (UAV) to a new trajectory before the UAV reaches the last hovering state of the current trajectory, so that the UAV can fly smoothly and quickly.

The feasibility of the designed system is validated through online flight experiments in indoor environments with obstacles.

The proposed trajectory planning system is integrated into a quadrotor platform. It is easily implementable onboard and computationally efficient.

The proposed local planner for trajectory generation and evaluation combines PSO and JLT generation algorithms. The proposed method can provide a collision-free and continuous trajectory, significantly reducing the required computing resources. The PSO algorithm locally searches for feasible target points near the global waypoint obtained by the global path search. The JLT generation algorithm generates trajectories from the current state toward each point contained by the target point set. The proposed trajectory switching strategy can avoid unnecessary hovering states in flight and ensure a continuous and safe flight trajectory. It is especially suitable for micro quadrotors with a small payload and limited onboard computing power.

In order to improve the computation efficiency of the four-point rainflow algorithm, a new fast four-point rainflow cycle counting algorithm (FFRA) using a novel loop iteration mode is proposed.

In this new algorithm, the loop iteration mode is simplified by reducing the number of iterations, tests and deletions. The high efficiency of the new algorithm makes it a preferable candidate in fatigue life online estimation of structural health monitoring systems.

The extensive simulation results show that the extracted cycles by the new FFRA are the same as those by the four-point rainflow cycle counting algorithm (FRA) and the three-point rainflow cycle counting algorithm (TRA). Especially, the simulation results indicate that the computation efficiency of the FFRA has improved an average of 12.4 times compared to the FRA and an average of 8.9 times compared to the TRA. Moreover, the equivalence of cycle extraction results between the FFRA and the FRA is proved mathematically by utilizing some fundamental properties of the rainflow algorithm. Theoretical proof of the efficiency improvement of the FFRA in comparison to the FRA is also given.

This merit makes the FFRA preferable in online monitoring systems of structures where fatigue life estimation needs to be accomplished online based on massive measured data. It is noticeable that the high efficiency of the FFRA attributed to the simple loop iteration, which provides beneficial guidance to improve the efficiency of existing algorithms.

In the digital assembly system of large aircraft components (LAC), the docking trajectory of LAC is an important factor affecting the docking accuracy and stability of the LAC. The main content of docking trajectory planning is how to move the LAC from the initial posture and position to the target posture and position (TPP). This paper aims to propose a trajectory planning method of LAC based on measured data.

First, the posture and position error model of the wing is constructed according to the measured data of the measurement points (MPs) and the fork lug joints. Second, the particle swarm optimization algorithm based on the dynamic inertia factor is used to optimize the TPP of the wing. Third, to ensure the efficiency and stability of posture adjustment, the S-shaped curve is used as the motion trajectory of LAC, and the parameters of the trajectory are solved by the generalized multiplier method. Finally, a series of docking experiments are carried out.

During the process of posture adjustment, the motion of the numerical control locator (NCL) is stable, and the interaction force between the NCLs is always within a reasonable range. After the docking, the MPs are all within the tolerance range, and the coaxiality error of the fork lug hole is less than 0.2 mm.

In this paper, the measured data rather than the theoretical design model is used to solve the TPP, which improves the docking accuracy of LAC. Experiment results show that the proposed trajectory method can complete the LAC docking effectively and improve the docking accuracy.

In this paper, the smoothed point interpolation method (SPIM) is used to model the slope deformation. However, the computational efficiency of SPIM is not satisfying when modeling the large-scale nonlinear deformation problems of geological bodies.

In this paper, the SPIM is used to model the slope deformation. However, the computational efficiency of SPIM is not satisfying when modeling the large-scale nonlinear deformation problems of geological bodies.

A simple slope model with different mesh sizes is used to verify the performance of the efficient face-based SPIM. The first accelerating strategy greatly enhances the computational efficiency of solving the large-scale slope deformation. The second accelerating strategy effectively improves the convergence of nonlinear behavior that occurred in the slope deformation.

The designed efficient face-based SPIM can enhance the computational efficiency when analyzing large-scale nonlinear slope deformation problems, which can help to predict and prevent potential geological hazards.

The transient loads on the spherical hybrid sliding bearings (SHSBs) rotor system during the process of accelerating to stable speed are related to time, which exhibits a complex transient response of the rotor dynamics. The current study of the shaft center trajectory of the SHSBs rotor system is based on the assumption that the rotational speed is constant, which cannot truly reflect the trajectory of the rotor during operation. The purpose of this paper truly reflects the trajectory of the rotor and further investigates the stability of the rotor system during acceleration of SHSBs.

The model for accelerated rotor dynamics of SHSBs is established. The model is efficiently solved based on the fourth-order Runge–Kutta method and then to obtain the shaft center trajectory of the rotor during acceleration.

Results show that the bearing should choose larger angular acceleration in the acceleration process from startup to the working speed; rotor system is more stable. With the target rotational speed increasing, the changes in the shaft trajectory of the acceleration process are becoming more complex, resulting in more time required for the bearing stability. When considering the stability of the rotor system during acceleration, the rotor equations of motion provide a feasible solution for the simulation of bearing rotor system.

The study can simulate the running stability of the shaft system from startup to the working speed in this process, which provides theoretical guidance for the stability of the rotor system of the SHSBs in the acceleration process.

The purpose of this study is to further advance the multiple space/time subdomain framework with model reduction. Existing linear multistep (LMS) methods that are second-order time accurate, and useful for practical applications, have a significant limitation. They do not account for separable controllable numerical dissipation of the primary variables. Furthermore, they have little or no significant choices of altogether different algorithms that can be integrated in a single analysis to mitigate numerical oscillations that may occur. In lieu of such limitations, under the generalized single-step single-solve (GS4) umbrella, several of the deficiencies are circumvented.

The GS4 framework encompasses a wide variety of LMS schemes that are all second-order time accurate and offers controllable numerical dissipation. Unlike existing state-of-art, the present framework permits implicit–implicit and implicit–explicit coupling of algorithms via differential algebraic equations (DAE). As further advancement, this study embeds proper orthogonal decomposition (POD) to further reduce model sizes. This study also uses an iterative convergence check in acquiring sufficient snapshot data to adequately capture the physics to prescribed accuracy requirements. Simple linear/nonlinear transient numerical examples are presented to provide proof of concept.

The present DAE-GS4-POD framework has the flexibility of using different spatial methods and different time integration algorithms in altogether different subdomains in conjunction with the POD to advance and improve the computational efficiency.

The novelty of this paper is the addition of reduced order modeling features, how it applies to the previous DAE-GS4 framework and the improvement of the computational efficiency. The proposed framework/tool kit provides all the needed flexibility, robustness and adaptability for engineering computations.