Search

The extreme ultraviolet (EUV) photolithography process is a cornerstone of semiconductor manufacturing and operates under demanding precision standards realized via nanometer-level overlay (OVL) error modeling. This procedure allows the machine to anticipate and correct OVL errors before impacting the wafer, thereby facilitating near-optimal image exposure while simultaneously minimizing the overall OVL error. Such models are usually high dimensional and exhibit rigorous statistical phenomena such as collinearities that play a crucial role in the process of tuning their parameters. Ordinary least squares (OLS) is the most widely used method for parameters tuning of overlay models, but in most cases it fails to compensate for such phenomena. In this paper, we propose the usage of ridge regression, a widely known machine learning (ML) algorithm especially suitable for datasets that exhibit high multicollinearity. The proposed method was applied in perturbed data from a 300 mm wafer fab, and the results show reduced residuals when ridge regression is applied instead of OLS. [ABSTRACT FROM AUTHOR]

Aiming at the problem of poor uniformity of maize sowing caused by ground wheel slip, an electronic control seed-metering system (ECSMS) for maize single seed precision sowing was designed and a mathematical model for motor control of the ECSMS was determined. The PID parameters were set by Z-N method and fuzzy control. The fuzzy PID control design and Simulink simulation were completed by MATLAB, which reduced response time of the system by 0.23 s and improved the control accuracy. Experiments on the JPS-12 test bench show that the qualification index (QI) of maize seed-metering device with the ECSMS increases by 4.47%, the multiples index (MI) decreases by 1.96%, the miss index (MIX) decreases by 2.81%, and the coefficient of variation (CV) of qualified seed spacing decreases by 5.06%, and the sowing uniformity has been greatly improved. Test results of the soil-tank test bench show that the system has good sowing uniformity and stability. And the QI is 96.74%, the MI is 2.15%, the MIX is 1.10%, and the CV of qualified seed spacing is 16.24%. Under different setting seed spacing and different sowing operation speed, the change range of seeding quality index was within 10%. The results of field sowing test show that the QI was 84.21%, the MI was 2.63%, the MIX was 7.89%, and the CV of qualified seed spacing was 22.15%, which meet the requirements of JB/T 10293-2013 ‘Specification for single seed planters (precision planters)’ and the agronomic requirements for maize precision sowing. The system runs stably and reliably in practical operation and has good operation performance. [ABSTRACT FROM AUTHOR]

Multiple-point geostatistics (MPS) is an established tool for the uncertainty quantification of Earth systems modeling, particularly when dealing with the complexity and heterogeneity of geological data. This study presents a novel pixel-based MPS method for modeling spatial data using advanced machine-learning algorithms. Pixel-based multiple-point simulation implies the sequential modeling of individual points on the simulation grid, one at a time, by borrowing spatial information from the training image and honoring the conditioning data points. The developed methodology is based on the mapping of the training image patterns database using the t-Distributed Stochastic Neighbor Embedding (t-SNE) algorithm for dimensionality reduction, and the clustering of patterns by applying the Density-based Spatial Clustering of Applications with Noise (DBSCAN) algorithm, as an efficient unsupervised classification technique. For the automation, optimization, and input parameter tuning, multiple stages are implemented, including entropy-based determination of the template size and a k-nearest neighbors search for clustering parameter selection, to ensure the proposed method does not require the user's interference. The proposed model is validated using synthetic two- and three-dimensional datasets, both for conditional and unconditional simulations, and runtime information is provided. Finally, the method is applied to a case study gold mine for stochastic orebody modeling. To demonstrate the computational efficiency and accuracy of the proposed method, a two-dimensional training image with 101 by 101 pixels is simulated for 100 conditional realizations in 453 s (~4.5 s per realization) using only 361 hard data points (~3.5% of the simulation grid), and the resulting average simulation has a good visual match and only an 11.8% pixel-wise mismatch with the training image. [ABSTRACT FROM AUTHOR]

The extreme ultraviolet (EUV) photolithography process is a cornerstone of semiconductor manufacturing and operates under demanding precision standards realized via nanometer-level overlay (OVL) error modeling. This procedure allows the machine to anticipate and correct OVL errors before impacting the wafer, thereby facilitating near-optimal image exposure while simultaneously minimizing the overall OVL error. Such models are usually high dimensional and exhibit rigorous statistical phenomena such as collinearities that play a crucial role in the process of tuning their parameters. Ordinary least squares (OLS) is the most widely used method for parameters tuning of overlay models, but in most cases it fails to compensate for such phenomena. In this paper, we propose the usage of ridge regression, a widely known machine learning (ML) algorithm especially suitable for datasets that exhibit high multicollinearity. The proposed method was applied in perturbed data from a 300 mm wafer fab, and the results show reduced residuals when ridge regression is applied instead of OLS.

Multiple-point geostatistics (MPS) is an established tool for the uncertainty quantification of Earth systems modeling, particularly when dealing with the complexity and heterogeneity of geological data. This study presents a novel pixel-based MPS method for modeling spatial data using advanced machine-learning algorithms. Pixel-based multiple-point simulation implies the sequential modeling of individual points on the simulation grid, one at a time, by borrowing spatial information from the training image and honoring the conditioning data points. The developed methodology is based on the mapping of the training image patterns database using the t-Distributed Stochastic Neighbor Embedding (t-SNE) algorithm for dimensionality reduction, and the clustering of patterns by applying the Density-based Spatial Clustering of Applications with Noise (DBSCAN) algorithm, as an efficient unsupervised classification technique. For the automation, optimization, and input parameter tuning, multiple stages are implemented, including entropy-based determination of the template size and a k-nearest neighbors search for clustering parameter selection, to ensure the proposed method does not require the user’s interference. The proposed model is validated using synthetic two- and three-dimensional datasets, both for conditional and unconditional simulations, and runtime information is provided. Finally, the method is applied to a case study gold mine for stochastic orebody modeling. To demonstrate the computational efficiency and accuracy of the proposed method, a two-dimensional training image with 101 by 101 pixels is simulated for 100 conditional realizations in 453 s (~4.5 s per realization) using only 361 hard data points (~3.5% of the simulation grid), and the resulting average simulation has a good visual match and only an 11.8% pixel-wise mismatch with the training image.

This paper proposes a method that combines active disturbance rejection control (ADRC) and adaptive fuzzy sliding mode control (AFSMC), which is beneficial for the optoelectronic platform to enhance the target tracking capability. A servo system model based on the LuGre friction is first established. The AFSMC controller estimates the unknown part of the platform, and the fuzzy approximator can reduce chattering. Then, an ADRC controller based on Back-Propagation neural network tuning is designed and compared with the empirical method, which makes the system's tracking accuracy higher. Lyapunov's theorem and Barbara's lemma are forceful methods to prove engineering stability. Simulations illustrate that the influence of external disturbances on the optoelectronic platform can be suppressed, thereby enhancing the disturbance isolation of the controller. • The control scheme with double speed loop can improve tracking performance. • The adaptive fuzzy sliding mode controller can estimate model nonlinearities and torque disturbances. • The ADRC controller using neural network tuning can suppress external disturbances faster than traditional tuning. • The simulations show that the control method is effective. [ABSTRACT FROM AUTHOR]

Background Spatial transcriptomics (ST) combines stained tissue images with spatially resolved high-throughput RNA sequencing. The spatial transcriptomic analysis includes challenging tasks like clustering, where a partition among data points (spots) is defined by means of a similarity measure. Improving clustering results is a key factor as clustering affects subsequent downstream analysis. State-of-the-art approaches group data by taking into account transcriptional similarity and some by exploiting spatial information as well. However, it is not yet clear how much the spatial information combined with transcriptomics improves the clustering result. Results We propose a new clustering method, Stardust , that easily exploits the combination of space and transcriptomic information in the clustering procedure through a manual or fully automatic tuning of algorithm parameters. Moreover, a parameter-free version of the method is also provided where the spatial contribution depends dynamically on the expression distances distribution in the space. We evaluated the proposed methods results by analyzing ST data sets available on the 10x Genomics website and comparing clustering performances with state-of-the-art approaches by measuring the spots' stability in the clusters and their biological coherence. Stability is defined by the tendency of each point to remain clustered with the same neighbors when perturbations are applied. Conclusions Stardust is an easy-to-use methodology allowing to define how much spatial information should influence clustering on different tissues and achieving more stable results than state-of-the-art approaches. [ABSTRACT FROM AUTHOR]

The quad tilt rotor (QTR) has complex dynamics characteristics, and the environmental factors have a great influence on it. This increases the difficulty of its control especially in transition mode. To solve this problem, the design of the controller based on active disturbance rejection control (ADRC) with a novel tuning algorithm is proposed in this paper. The aerodynamic models of propeller, wing, vertical tail and fuselage are build respectively by using the idea of component modeling. The pitch channel of the linearized flight dynamic model is provided for the control system. A simple tuning method of active disturbance rejection control for the quad tilt rotor in transition process that achieves high performance and good robustness is presented. The proposed method makes ADRC become easy to tune and more practical. The problem of parameter tuning is turned into a multi-objective optimization problem, and using radial basis function (RBF) neural network with particle swarm optimization algorithm (PSO) and bacterial foraging optimization algorithm (BFO) hybrid algorithm tuning method (NBPO) to fit the tuning rules. We introduce the control input, the overshoot and rise time of the system into the fitness function. The local and global optimal solutions are introduced into the chemotaxis and migration to improve the information exchange ability of BFO. The correctness and effectiveness of the NBPO tuning method were verified by numerical simulation results, compared with the one parameter tuning method (OPM) and PID, the results showed that the NBPO tuning method can greatly improve the control performance of the system.

The continuous penetration of renewable energy resources (RES) into the energy mix and the transition of the traditional electric grid towards a more intelligent, flexible and interactive system, has brought electrical load forecasting to the foreground of smart grid planning and operation. Predicting the electric load is a challenging task due to its high volatility and uncertainty, either when it refers to the distribution system or to a single household. In this paper, a novel methodology is introduced which leverages the advantages of the state-of-the-art deep learning algorithms and specifically the Convolution Neural Nets (CNN). The main feature of the proposed methodology is the exploitation of the statistical properties of each time series dataset, so as to optimize the hyper-parameters of the neural network and in addition transform the given dataset into a form that allows maximum exploitation of the CNN algorithm's advantages. The proposed algorithm is compared with the LSTM (Long Short Term Memory) technique which is the state of the art solution for electric load forecasting. The evaluation of the algorithms was conducted by employing three open-source, publicly available datasets. The experimental results show strong evidence of the effectiveness of the proposed methodology. [ABSTRACT FROM AUTHOR]

The frequency stability in the sending-end network with renewable energy generating units should be paid attention especially for an asynchronous interconnection system. When the DC blocking fault occurs in the system, the problem of high frequency in the sending-end network can be suppressed by using the frequency limit controller (FLC). This study introduces the structure and the parameters tuning method of the FLC, and the parameters optimisation method is proposed to avoid FLC acting repeatedly. In the study, a case is studied, and the results show that the FLC is important in the sending-end network. Finally, aiming at the coordination problem between multiple FLCs, a control strategy based on FLC influence coefficient is proposed in the study, and the strategy can suppress the over regulation effectively.

This paper presents a predator-prey pigeon-inspired optimization (PPPIO) algorithm for an automatic landing system of a fixed-wing unmanned aerial vehicle (UAV) in the longitudinal plane. A pitch command autopilot and two types of approach power compensators are presented and evaluated in automatic landing. The design of a glide slope command system and an approach power compensator are converted to a finite-dimensional optimization problem. The longitudinal flight controller parameters are optimized by the proposed PPPIO algorithm. The differences between the two proposed approach power compensators are also analyzed. The proposed optimization process can guarantee the control gains vector converge to an approximate optimal solution, and it is computationally much more efficient. Simulation results are presented to demonstrate that this approach helps solve control system optimization problems for different criteria and attains a rather accurate result. As verified by these simulation results, the proposed automatic landing system can help improve UAV's performance during the landing phase. [ABSTRACT FROM AUTHOR]

The frequency stability in the sending-end network with renewable energy generating units should be paid attention especially for an asynchronous interconnection system. When the DC blocking fault occurs in the system, the problem of high frequency in the sending-end network can be suppressed by using the frequency limit controller (FLC). This study introduces the structure and the parameters tuning method of the FLC, and the parameters optimisation method is proposed to avoid FLC acting repeatedly. In the study, a case is studied, and the results show that the FLC is important in the sending-end network. Finally, aiming at the coordination problem between multiple FLCs, a control strategy based on FLC influence coefficient is proposed in the study, and the strategy can suppress the over regulation effectively. [ABSTRACT FROM AUTHOR]

This study focused on the use of the non‐dominated sorting genetic algorithm (NSGA‐II) to achieve an optimal power system stabiliser (PSS) parameters for a given operating point with a renewable source of energy so as to increase the system damping and guarantee enough stability margin. The parameters tuning was formulated using an eigenvalue‐based multi‐objective function. In recent years, the changeability and fluctuation of the wind power injected into the network have led to new challenges in small signal stability (SSS). These wind speed change and load demand change perturbations take place routinely and cause the variation of the operating conditions. However, as the conventional PSS is conceived for a fixed operating point in order to obtain the linearised transfer function model, it cannot yield good results when the operating range is too wide. To this end, the adaptive neuro‐fuzzy inference system was proposed to estimate the stabiliser parameters in real time after a learning phase. The nine‐bus Western System Coordinating Council and the obtained simulations results were assessed using Matlab/Simulink package. The validity of the proposed methodology was checked through the PSS parameters evolution simulation for daily load forecast curves and monthly wind speed prediction curves. [ABSTRACT FROM AUTHOR]

The continuous penetration of renewable energy resources (RES) into the energy mix and the transition of the traditional electric grid towards a more intelligent, flexible and interactive system, has brought electrical load forecasting to the foreground of smart grid planning and operation. Predicting the electric load is a challenging task due to its high volatility and uncertainty, either when it refers to the distribution system or to a single household. In this paper, a novel methodology is introduced which leverages the advantages of the state-of-the-art deep learning algorithms and specifically the Convolution Neural Nets (CNN). The main feature of the proposed methodology is the exploitation of the statistical properties of each time series dataset, so as to optimize the hyper-parameters of the neural network and in addition transform the given dataset into a form that allows maximum exploitation of the CNN algorithm’s advantages. The proposed algorithm is compared with the LSTM (Long Short Term Memory) technique which is the state of the art solution for electric load forecasting. The evaluation of the algorithms was conducted by employing three open-source, publicly available datasets. The experimental results show strong evidence of the effectiveness of the proposed methodology.

Owing to its significant roles in computer vision applications, human face tracking has drawn extensive attention in recent years. Most researchers solve face tracking using particle filter, meanshift and their derivatives. Unlike the traditional methods, in this study, face tracking is treated as an optimisation problem and a new meta‐heuristic optimisation algorithm, differential harmony search (DHS), is introduced to solve face tracking problems. We compare the speed and accuracy of the proposed method with particle filter, meanshift and improved harmony search. Experimental results show that DHS‐based tracker is faster and more accurate and it is easy to handle the parameters tuning. Furthermore, to improve the reliability of tracking, multiple visual cues are applied to DHS‐based tracking system and experimental results demonstrate the increased robustness achieved by fusing multiple cues.

Abstract This paper presents a parameters tuning method based on the genetic algorithm (GA) for an active disturbance rejection control (ADRC) of a three-axis inertially stabilized platform (ISP) with imaging sensors. To improve the stabilization accuracy and robustness of an aerial ISP under multi-source disturbances environment, an ADRC control scheme is first proposed. Then, to accurately identify and tune the parameters in the ADRC controller, a GA-based parameters tuning method is proposed. In this way, the performance of the ADRC is superior to the empirical method. To validate the proposed method, the simulations and experiments are carried out. The results show that the proposed ADRC with GA-based parameters tuning method has significant disturbance rejection ability which can improve the stabilization accuracy obviously. Compared with the ADRC with empirically tuning method, the stabilization error (RMS) under movable base is decreased up to 50.09%. Highlights • An ADRC control scheme is proposed to improve the stabilization accuracy and disturbance rejection ability of an ISP. • A GA-based parameters tuning method is proposed to accurately identify and tune the parameters of the ADRC controller. • The empirical parameters tuning method and the proposed GA-based method are both applied in the work as a comparison. • Simulations and experiments are carried out to validate the proposed method. [ABSTRACT FROM AUTHOR]

Modern scientific applications are composed of various methods, techniques and models to solve complicated problems. Such composite applications commonly are represented as workflows. Workflow scheduling is a well-known optimization problem, for which there is a great amount of solutions. Most of the algorithms contain parameters, which affect the result of a method. Thus, for the efficient scheduling it is important to tune parameters of the algorithms. Moreover, performance models, which are used for the estimation of obtained solutions, are crucial parts of workflow scheduling. In this work we present a combined approach for automatic parameters tuning and performance models construction in the background of the WMS lifecycle. Algorithms tuning is provided by hyper-heuristic genetic algorithm, whereas models construction is performed via symbolic regression methods. Developed algorithm was evaluated using CLAVIRE platform and is applicable for any distributed computing systems to optimize the execution of composite applications. [ABSTRACT FROM AUTHOR]

This paper presents an adaptive PSO algorithm whose numerical parameters can be updated following a scheduled protocol respecting some known criteria of convergence in order to enhance the chances to reach the global optimum of a hard combinatorial optimization problem, such those encountered in global optimization problems of composite laminates. Some examples concerning hard design problems are provided, showing the effectiveness of the approach.

The process of tuning the parameters that characterize evolutionary algorithms is difficult and can be time consuming. This paper presents a self-tuning algorithm for dynamically updating the crossover and mutation probabilities during a run of genetic programming. The genetic operators that are considered in this work are the geometric semantic genetic operators introduced by Moraglio et al. Differently from other existing self-tuning algorithms, the proposed one works by assigning a (different) crossover and mutation probability to each individual of the population. The experimental results we present show the appropriateness of the proposed self-tuning algorithm: on seven different test problems, the proposed algorithm finds solutions of a quality that is better than, or comparable to, the one achieved using the best known values for the geometric semantic crossover and mutation rates for the same problems. Also, we study how the mutation and crossover probabilities change during the execution of the proposed self-tuning algorithm, pointing out an interesting insight: mutation is basically the only operator used in the exploration phase, while crossover is used for exploitation, further improving good quality solutions. [ABSTRACT FROM AUTHOR]

Abstract: This paper pretends to offer design rules for the parameters adjustment of the Dynamic Matrix Control (DMC) to allow an easier starting up. The effect on the time response of each algorithm parameter that can be tuned by the user is studied in an unconstrained system. To this aim, the position of the closed loop poles of the equivalent system is calculated. To simplify the study and to obtain more direct conclusions the number of poles will be limited using a First Order Plus Death Time simplification of the real plant. Design rules proposed in this study are tested in some simulated benchmarks and in a real plant. [Copyright &y& Elsevier]

Abstract: The traditional integer order proportional-integral-differential (IO-PID) controller is sensitive to the parameter variation or/and external load disturbance of permanent magnet synchronous motor (PMSM). And the fractional order proportional-integral-differential (FO-PID) control scheme based on robustness tuning method is proposed to enhance the robustness. But the robustness focuses on the open-loop gain variation of controlled plant. In this paper, an enhanced robust fractional order proportional-plus-integral (ERFOPI) controller based on neural network is proposed. The control law of the ERFOPI controller is acted on a fractional order implement function (FOIF) of tracking error but not tracking error directly, which, according to theory analysis, can enhance the robust performance of system. Tuning rules and approaches, based on phase margin, crossover frequency specification and robustness rejecting gain variation, are introduced to obtain the parameters of ERFOPI controller. And the neural network algorithm is used to adjust the parameter of FOIF. Simulation and experimental results show that the method proposed in this paper not only achieve favorable tracking performance, but also is robust with regard to external load disturbance and parameter variation. [Copyright &y& Elsevier]

Abstract: Forest fires cause important losses around the world every year. A good prediction of fire propagation is a crucial point to minimize the devastating effects of these hazards. Several models that represent this phenomenon and provide a prediction of its spread have been developed. These models need input parameters which are usually difficult to know or even estimate. A two- stage prediction methodology was proposed to improve the quality of these parameters. In this methodology, such parameters are calibrated according to real observations and then, used in the prediction step. However, there are several parameters, which are not uniform along the map, but vary according to the topography of the terrain. Besides, these parameters are not constant along time but they are strongly dynamic. In such cases, it is necessary to introduce complementary models that overcome both restrictions. In the former case, the need of a spatial distribution model of a given variable is needed to be able to provide a spatial distribution for a given variable along the whole terrain by starting from the measured values of that parameter in certain points of the terrain. In the case of time variability, a complementary model such as weather forecasting model, could enable the capability of dealing with dynamic behavior of these parameters along time. In this paper, we describe an enhanced two-stage prediction scheme, where both type of complementary models a wind field model and a weather prediction model are coupled to the prediction scheme by enabling the system to dynamically adapts to complex terrains and dynamic conditions. [Copyright &y& Elsevier]