Differential evolution (DE) is a simple and efficient population-based stochastic algorithm for solving global numerical optimization problems. DE largely depends on algorithm parameter values and search strategy. Knowledge on how to tune the best values of these parameters is scarce. This paper aims to present a consistent methodology for tuning optimal parameters. At the heart of the methodology is the use of an artificial neural network (ANN) that learns to draw links between the algorithm performance and parameter values. To do so, first, a data-set is generated and normalized, then the ANN approach is performed, and finally, the best parameter values are extracted. The proposed method is evaluated on a set of 24 test problems from the Black-Box Optimization Benchmarking (BBOB) benchmark. Experimental results show that three distinct cases may arise with the application of this method. For each case, specifications about the procedure to follow are given. Finally, a comparison with four tuning rules is performed in order to verify and validate the proposed method’s performance. This study provides a thorough insight into optimal parameter tuning, which may be of great use for users.

The computational cost and memory demand required by computational fluid dynamics (CFD) codes simulations can become very high. Therefore, the application of convolutional neural networks (CNN) in this field has been studied owing to its capacity to learn patterns from sets of input data, which can considerably approximate the results of the CFD simulations with relative low errors. DeepCFD code has been taken as a basis and with some slight variations in the parameters of the CNN, while the net is able to solve the Navier–Stokes equations for steady turbulent flows with variable input velocities to the domain. In order to acquire extensive input data to the CNN, a data augmentation technique, which considers the similarity principle for fluid dynamics, is implemented. As a consequence, DeepCFD is able to learn the velocities and pressure fields quite accurately, speeding up the time-consuming CFD simulations.

The application of computational fluid dynamics (CFD) to turbulent flow has been a considerable topic of research for many years. Nonetheless, using CFD tools results in a large computational cost, which implies that, for some applications, CFD may be unviable. To date, several authors have carried out research applying deep learning (DL) techniques to CFD-based simulations. One of the main applications of DL with CFD is in the use of convolutional neural networks (CNNs) to predict which samples will have the desired magnitude. In this study, a CNN which predicts the streamwise and vertical velocities and the pressure fields downstream of a circular cylinder for a series of time instants is presented. The CNN was trained using a signed distance function (SDF), a flow region channel (FRC) and the t-1 sample as inputs, and the ground-truth CFD data as the output. The results showed that the CNN was able to predict multiple time instants with low error rates for turbulent flows with variable input velocities to the domain.

When training a feedforward stochastic gradient descendent trained neural network, there is a possibility of not learning a batch of patterns correctly that causes the network to fail in the predictions in the areas adjacent to those patterns. This problem has usually been resolved by directly adding more complexity to the network, normally by increasing the number of learning layers, which means it will be heavier to run on the workstation. In this paper, the properties and the effect of the patterns on the network are analysed and two main reasons why the patterns are not learned correctly are distinguished: the disappearance of the Jacobian gradient on the processing layers of the network and the opposite direction of the gradient of those patterns. A simplified experiment has been carried out on a simple neural network and the errors appearing during and after training have been monitored. Taking into account the data obtained, the initial hypothesis of causes seems to be correct. Finally, some corrections to the network are proposed with the aim of solving those training issues and to be able to offer a sufficiently correct prediction, in order to increase the complexity of the network as little as possible.

There is presently a need for more robust navigation algorithms for autonomous industrial vehicles. These have reasonably guaranteed the adequate reliability of the navigation. In the current work, the stability of a modified algorithm for collision-free guiding of this type of vehicle is ensured. A lateral control and a longitudinal control are implemented. To demonstrate their viability, a stability analysis employing the Lyapunov method is carried out. In addition, this mathematical analysis enables the constants of the designed algorithm to be determined. In conjunction with the navigation algorithm, the present work satisfactorily solves the localization problem, also known as simultaneous localization and mapping (SLAM). Simultaneously, a convolutional neural network is managed, which is used to calculate the trajectory to be followed by the AGV, by implementing the artificial vision. The use of neural networks for image processing is considered to constitute the most robust and flexible method for realising a navigation algorithm. In this way, the autonomous vehicle is provided with considerable autonomy. It can be regarded that the designed algorithm is adequate, being able to trace any type of path.

The stochastic gradient descendant algorithm is one of the most popular neural network training algorithms. Many authors have contributed to modifying or adapting its shape and parametrizations in order to improve its performance. In this paper, the authors propose two modifications on this algorithm that can result in a better performance without increasing significantly the computational and time resources needed. The first one is a dynamic learning ratio depending on the network layer where it is applied, and the second one is a dynamic drop-out that decreases through the epochs of training. These techniques have been tested against different benchmark function to see their effect on the learning process. The obtained results show that the application of these techniques improves the performance of the learning of the neural network, especially when they are used together.