Automatic Design Method Research Articles

Optimization of underground mining production layouts considering geological uncertainty using deep reinforcement learning

Mineral extraction plays a key role in the global raw materials supply chain, however the exhaustion of shallow deposits and typical scarcity of sampled data during exploration activities creates challenges in mine planning and design, where decision-making is highly sensitive to uncertainty in geology and mineral grade prediction. Geostatistical techniques are commonly used to generate a set of equiprobable simulated numerical models to capture these uncertainties, however incorporating these simulated models within a mine planning and design framework remains a major challenge. The purpose of this paper is to propose a novel approach to decision-making in underground mine design that can use information from an ensemble of numerical realizations of a mineral resource to improve the financial performance of the asset. A deep reinforcement learning (DRL) framework, using the proximal policy optimization (PPO) algorithm, is developed for the design of underground mining production level layouts. A case study is presented using a gold mineral resource characterized by an ensemble of 100 numerical realizations to verify the advantages of the proposed method, considering a baseline consisting of an industry standard automated design method. The DRL approach achieved an 8.3% higher expected profit, a 3.4% more gold mined than the baseline, and has the added functionality of considering uncertainty in mineral grades.

Automatic Design of Robot Swarms under Concurrent Design Criteria: A Study Based on Iterated F‐Race

Automatic design is an appealing approach to realizing robot swarms. In this approach, a designer specifies a mission that the swarm must perform, and an optimization algorithm searches for the control software that enables the robots to perform the given mission. Traditionally, research in automatic design has focused on missions specified by a single design criterion, adopting methods based on single‐objective optimization algorithms. In this study, we investigate whether existing methods can be adapted to address missions specified by concurrent design criteria. We focus on the bi‐criteria case. We conduct experiments with a swarm of e‐puck robots that must perform sequences of two missions: each mission in the sequence is an independent design criterion that the automatic method must handle during the optimization process. We consider modular and neuroevolutionary methods that aggregate concurrent criteria via the weighted sum, hypervolume, or ‐norm. We compare their performance with that of Mandarina, an original automatic modular design method. Mandarina integrates Iterated F‐race as an optimization algorithm to conduct the design process without aggregating the design criteria. Results from realistic simulations and demonstrations with physical robots show that the best results are obtained with modular methods and when the design criteria are not aggregated.

Automatic Optimal Design Method for Minimum Total Resistance Hull Based on Enhanced FFD Method

Hull optimization plays a crucial role in enhancing ship performance and efficiency. This study aimed to improve ship performance, particularly by reducing total resistance, through automated optimization methods. To this end, an integrated, fully automated optimization program was developed based on Python, incorporating Enhanced Free-Form Deformation (FFD) technology, scripted CFD numerical evaluation, and the Particle Swarm Optimization (PSO) algorithm. This program allowed precise control of hull form, improving efficiency while reducing costs. The KCS hull, known for its excellent resistance performance, was chosen as the optimization target, with the goal of minimizing total resistance by adjusting the bulbous bow line plan. The research results indicated that the optimized scheme exhibited lower resistance characteristics compared to the original design while satisfying design constraints. This study not only provides a new optimization strategy for ship design but also lays a foundation for future hull optimization research.

Automatic optimization design of laser triangulation ranging sensors using an improved genetic algorithm

Optical system parameter design is of great importance to ensure the accuracy of asymmetry systems such as laser triangulation ranging systems. However, the system parameter determination often depends on the experience and manual attempts of designers, which is not only time-consuming but also inevitable to introduce human errors. Therefore, in this paper an automatic optimization design method based on nonlinear programming genetic algorithm with elitism strategy (E-NPGA) is proposed, to accurately and fast determine the optimal system parameters of laser triangulation ranging systems assisting in improving the measurement accuracy. Firstly, an optimization model of system parameters is developed under the Scheimpflug rule establishing the constraints for various measurement resolutions and ranges. An image size constraint is constructed for the first time to improve and evaluate the parameter optimization. Secondly, the E-NPGA is proposed with nonlinear optimization and elitism strategy, which can determine the optimal system parameters in 15 iterations avoiding local extremum. In design examples, using the E-NPGA determined system parameters ZEMAX simulation and experimental results of the parameters depended image spot size show a slight relative difference below 0.6%. Moreover, the experiment results demonstrate the sensor system designed by using the E-NPGA enables a distance measurement with submicron absolute error and 10−4 relative uncertainty. The automatic optimization method proposed in this paper is compared with the conventional GA method and PSO method, and it is validated that the convergence accuracy of the proposed method is much higher than the conventional ones.

Applications of reinforcement learning, machine learning, and virtual screening in SARS-CoV-2-related proteins

Similarly, to all coronaviruses, SARS-CoV-2 uses the S glycoprotein to enter host cells, which contains two functional domains: S1 and S2 receptor binding domain (RBD). Angiotensin-converting enzyme 2 (ACE2) is recognizable by the S proteins on the surface of the SARS-CoV-2 virus. The SARS-CoV-2 virus causes SARS, but some mutations in the RBD of the S protein markedly enhance their binding affinity to ACE2. Searching for new compounds in COVID-19 is an important initial step in drug discovery and materials design. Still, the problem is that this search requires trial-and-error experiments, which are costly and time-consuming. In the automatic molecular design method based on deep reinforcement learning, it is possible to design molecules with optimized physical properties by combining a newly devised coarse-grained representation of molecules with deep reinforcement learning. Also, structured-based virtual screening uses protein 3D structure information to evaluate the binding affinity between proteins and compounds based on physicochemical interactions such as van der Waals forces, Coulomb forces, and hydrogen bonds, and select drug candidate compounds. In addition, AlphaFold can predict 3D protein structures, given the amino acid sequence, and the protein building blocks. Ensemble docking, in which multiple protein structures are generated using the molecular dynamics method and docking calculations are performed for each, is often performed independently of docking calculations. In the future, the AlphaFold algorithm can be used to predict various protein structures related to COVID-19.

Swarm Robotics Navigation Task: A Comparative Study of Reinforcement Learning and Particle Swarm Optimization Methodologies

Automatic design methods focus on generating the collective behavior of swarm robotic systems. These methods enable multiple robots to coordinate and execute complex tasks in their environments autonomously. This research paper investigated two prominent methodologies: particle swarm optimization (PSO) and reinforcement learning (RL). A new comparative study was conducted to analyze the performance of a group of mobile robots through extensive experimentation. The objective was to produce navigational collective behavior through unknown environments. These environments differ in complexity ranging from obstacle-free environments to cluttered ones. The core metrics of the comparison include the time efficiency of individual robots and the overall swarm, flexibility in pathfinding, and the ability to generalize solutions for new environments. The obtained results from the Webots simulator with Python controller suggested that RL excels in environments closely aligned with its training conditions. RL achieved a faster completion time and demonstrated superior coordination among individual robots. However, its performance dips when facing untrained scenarios necessitating computationally expensive retraining or structural complexities to enhance adaptability. Conversely, PSO showed commendable consistency in performance. Despite its slower pace, it exhibited robustness in various challenging settings without reconfiguration.

A bionic soft actuator for gripper based on pneumatic metastructure and design method for it

In order to achieve fast and convenient design of fully flexible actuators suitable for special gripping modes, a soft actuator for gripper based on pneumatic metastructure is proposed inspired by the pulvinus of Mimosa pudica. The hyperelastic material is used to fabricate the soft actuators. Finite element models are built to analyze and predict the deformation of both the single unit cell and entire metastructures. Finally, the automatic design method is presented. The actuators designed corresponding to the modes are fabricated and tested. The experimental results show that the gripper achieves a higher grasping stability in special gripping modes.

Architecture Optimization for Hybrid Deep Residual Networks in Liver Tumor Segmentation Using a GA

Nowadays, liver tumor segmentation has been widely applied for vital medical objectives such as illness diagnosis, treatment, and evaluation of liver function. In this work, we aim to improve the performance of hybrid convolution neural network (CNN) models for liver tumor segmentation. A new automatic design method is introduced to build a hybrid CNN model with an optimal architecture using a combination of the original U_Net and the Res-UNet. In the proposed design method, the Res-UNet model is divided into different blocks, and then many different candidate architectures for the hybrid CNN model can be created using gradually these blocks to replace their corresponding blocks of the original U-Net. The replacement is executed if the candidate architecture using a residual block gives better performance than the corresponding one that uses the original U-Net block. We use a genetic algorithm (GA) to get the optimal architecture among the candidate architectures that gives the best performance for the hybrid CNN model. During the optimization process, in addition to the optimal architecture, other crucial configurations such as the learning algorithm, learning rate, and batch size are also determined. We refer to the hybrid CNN model that has the best architecture as GA_Res_UNet. Comparison has been made with other CNN models that are commonly used in image segmentation of liver tumors. We use the 3D-IRCADb01 liver tumor dataset that contains 2800 magnetic resonance (MR) images of 20 patients. Each image has a liver tumor with its corresponding mask. All MR images are resized to 256 × 256 in DICOM format. The findings indicate that GA_Res_UNet outperformed the other compared models over the considered dataset. It achieves 98.5% of dice coefficient and 99.8% of accuracy. Moreover, two other liver tumor segmentation datasets (LiTS17 and CHAOS) are utilized in another comparison to show the superior performance of our proposed model. The reported results show that the proposed model outperformed the other compared state-of-the-art models.

Multi-shelf graphic equalizer

A graphic equalizer (GEQ) is a standard tool in audio production and effect design. Adjustable gain control frequencies are fixed along the logarithmic frequency axis, and an automatic design method matches the magnitude response to them whenever target gains are changed. Most commonly, the GEQ comprises a set of peak filters centered an octave apart, possibly with a shelving filter at the bottom and top of the frequency range. While accurate designs were proposed, the dynamic range is typically limited to 24 dB. In this paper, we propose two innovations. First, we introduce a GEQ based on shelving filters only, which can cover an extensive dynamic range of over 60 dB. Secondly, we introduce an order-switching technique that combines shelf filters of different order. We demonstrate the performance and advantages of the proposed filter with design examples. The proposed shelf-filter-based GEQ offers a wider dynamic range and a smoother magnitude response than traditional peak-filter-based GEQ designs.

Automated method for designing extractive dividing wall columns used to separate minimum azeotropic mixtures with a heavy entrainer

Despite that the extractive divided wall column (E-DWC), which is one of the best examples of process intensification, has already found an industrial application, its design remains a new area of research. None of the few studies that have been published in the literature present a method for designing an E-DWC. Therefore, the first and fully automated design method, for the separation of minimum boiling azeotropic mixtures with a heavy entrainer (class (1.0–1a)-m1) using an extractive dividing wall column, is developed in this work. The developed method is based on the values of the phase equilibrium constant along the sides of composition simplex, with the help of the composition profile properties in the infinitely sharp splits mode and difference point equations. The efficiency of this new design method was tested on the separation of two ternary azeotropic mixtures that belong to class 1.0–1a-m1. The results show that the column trajectories, along the E-DWC, obtained by the rigorous simulation are in very good agreement with those calculated by the new design method. These results prove the reliability and accuracy of the new method.

Multi-objective optimisation research of ship form in waves based on Kernel principal component analysis

ABSTRACT In the process of multi-objective optimisation design of ship forms, subjective weighting methods rely on experts’ experience and have a certain impact on the scientificity of results. The objective weighting method can effectively solve this drawback and reduce the interference of human factors. Therefore, this paper proposes a multi-objective optimisation design programme for ship forms based on KPCA method, and constructs a multi-objective optimisation design framework. In this paper, the KPCA method is used to calculate the contribution rate of each indicator for objective weighting, give the comprehensive evaluation score of the optimal solution set and make decisions, solve the technical drawbacks of subjective weighting, and obtain a satisfactory optimal hull form. The results show that the automatic multi-objective optimisation design method proposed in this paper has certain engineering application value, and can provide theoretical basis and technical support for green design and manufacturing of ships.

A novel design model of flow channel paths for additive manufacturing

PurposeThis paper aims to propose a method for automatic design of additive manufacturing (AM) flow channel paths driven by path length and pressure loss. The research focuses on the automatic design of channel paths, intending to achieve the shortest flow channel length or minimum pressure loss and improve the design efficiency of AM parts.Design/methodology/approachThe initial layout of the flow channels is redesigned to consider the channels print supports. Boundary conditions and constraints are defined according to the redesigned channels layout, and the equation consisting of channel length and pressure loss is used as the objective function. Then the path planning simulation is performed based on particle swarm algorithm. The proposed method describes the path of flow channels using spline cures. The spline curve is controlled by particle (one particle represents a path), and the particle is randomly generated within the design space. After the path planning simulation is completed, the generated paths are used to create 3D parts.FindingsCase study 1 demonstrates the automatic design of hydraulic spool valve. Compared to conventional spool valve, the pressure loss was reduced by 86% and the mass was reduced by 83%. The design results of case study 2 indicate that this approach is able to find the shortest channel path with lower computational cost.Originality/valueThe automatic design method of flow channel paths driven by path length and pressure loss presented in this paper provides a novel solution for the creation of AM flow components.

SWAN: A multihead autoregressive attention model for solar wind speed forecasting

High speed streams of solar wind can impact Earth’s magnetic environment. Current state of the art geomagnetic effect forecasting methods can predict perturbations up to two hours in advance by using solar wind measurements made in the L1 Lagrange point. By forecasting these solar wind parameters, it should be possible to extend the time range for which said methods can be used.We present Solar Wind Attention Network (SWAN), an attention-based autoregressive model for the forecasting of daily average bulk speed values. We apply a novel automatic design method to create its architecture by developing a pipeline of hyperparameter optimization techniques as a proof of concept. Usage of this pipeline leads to a robust, reliable architecture which we empirically verify the convergence of in a variety of problem configurations regarding solar wind speed forecasting.An improvement in terms of root mean squared error (RMSE) is found when compared to WindNet and convolutional neural networks (CNNs) expanding it, even when SWAN does not incorporate images of the solar disk into the input. SWAN’s RMSE of 68.79 ± 1.05 km s−1 with a lead time of three days compares to RMSE values in the order of 80.28 ± 3.05, 76.30 ± 1.87 and 71.93 ± 0.40 km s−1. From these results, we discuss the possibility of incorporating both autoregressive and image information as input to enhance predictive capability of newly developed models. Furthermore, we analyze the resulting model’s behavior to characterize its main error sources, and find its performance to be at its best in scenarios not involving newly formed coronal holes or coronal mass ejections. It is found that this behavior matches well with a priori expectations derived from the underlying physics. Finally, we briefly discuss the need for an in-depth study of different evaluation metrics to provide a more thorough understanding of modeling results on the field.

Automated intelligent design of modified uni-traveling carrier photodectors

This paper introduces an automatic intelligent design method for the modified uni-traveling carrier photodetector (MUTC-PD). The conventional photodetector design process often relies on the numerical solution of complex nonlinear partial differential equations to simulate and optimize device performance, which is not only computationally intensive but also inefficient. To overcome this challenge, we apply the charge control principle to calculate the photodetector bandwidth, which improves the computational speed by a factor of approximately 1800 compared to the numerical solution of nonlinear partial differential equations. To further optimize the structure of the photodetector, we incorporate the Velocity Varying Climbing Particle Swarm Optimization (VVCPSO) algorithm. This is an improved algorithm based on the traditional particle swarm algorithm, which is able to quickly find the optimal solution in a complex parameter space. By applying the VVCPSO algorithm, we successfully fine-tuned the photodetector structure and obtained structural parameters with optimal performance. Our thorough verification process confirms that the proposed method is consistent with the results of ATLAS simulation software. Automated design has resulted in a high-performance MUTC-PD with a responsivity of 0.52A/W and a bandwidth of 60 GHz (@-3 V) at a mesa diameter of 16µm. Compared to the pre-optimized device, the bandwidth is increased to three times the original. By reducing the mesa diameter to 4µm, the bandwidth can be further increased to 82 GHz (@-3 V). The proposed method's calculation speed is fast enough, enabling extensive parameter studies to optimize device performance.

DARTS-based morphological neural network design and application in bearing fault diagnosis

Morphological filters are widely used in data pre-processing, whereas the existing morphological filters are not optimal in terms of performance, since they are designed by combining the basic morphological operators together artificially. Hence, we proposed an automatic morphological neural network design method based on differentiable architecture search (DARTS), which is a neural architecture search (NAS) method. Firstly, the mechanism is revealed that the morphological dilation with flat structural elements behaves the same as Maxpooling without down-sampling. Secondly, morphological dilation and erosion neural network layers were defined. Thirdly, using DARTS and the search space with morphological neural network layers, end-to-end morphological neural network was implemented. Finally, the morphological neural network was successfully applied in bearing fault diagnosis. Experiments on the three datasets show that the proposed method improved the feature extraction capabilities of neural networks, then achieved end-to-end bearing fault diagnosis, and the diagnosis accuracy was improved by more than 2.0%.

Automatic Knowledge Graph matching via Self-adaptive Designed Genetic Programming

Knowledge Graph (KG) provides a structured representation of domain knowledge by formally defining entities and their relationships. However, distinct communities tend to employ different terminologies and granularity levels to describe the same entity, leading to the KG heterogeneity issue that hampers their communications. KG matching can identify semantically similar entities in two KGs, which is an effective solution to this problem. Similarity Measures (SMs) are the foundation of the KG matching technique, and due to the complexity of entity heterogeneity, it is necessary to construct a high-level SM by selecting and combining the basic SMs. However, the large number of SMs and their intricate relationships make SM construction an open challenge. Inspired by the success of Evolutionary Algorithms (EA) in addressing the entity matching problem, this work further proposes a novel Self-adaptive Designed Genetic Programming (SDGP) to automatically construct the SM for KG matching. To overcome the drawbacks of the classic EA-based matching methods, a new individual representation and a novel fitness function are proposed to enable SDGP automatically explore the SM selection and combination. Then, a new Adaptive Automatic Design (AAD) method is introduced to adaptively trade off SDGP’s exploration and exploitation, which can determine the timing of AAD and efficiently determine the suitable breeding operators and control parameters for SDGP. The experiment uses the Ontology Alignment Evaluation Initiative’s Knowledge Graph (KG) data set to test the performance of SDGP. The experimental results show that SDGP can effectively determine high-quality KG alignments, which significantly outperform state-of-the-art KG matching methods.

Low-cost architecture performance evaluation strategy based on pixel difference degree contrast measurement

The time and effort required to manually design deep neural architectures is extremely high, which has led to the development of neural architecture search technology as an automatic architecture design method. However, the neural architecture search convergence process is slow and expensive, and the process requires training a large number of candidate architectures to get the final result. If the final accuracy of an architecture can be predicted from its initial state, this problem can be greatly alleviated. Therefore, this paper proposes a low-cost architecture performance evaluation strategy based on pixel difference degree contrast measurement, which takes 1) the difference matrix value between the feature map generated in the untrained architecture and the original image, and 2) the predicted accuracy of the neural network as evaluation indices. A new multi-index weight comprehensive measurement strategy was introduced to comprehensively score the multi-index, the real architecture performance can be approximately represented by score, which greatly reduces the cost of architecture evaluation. The experimental show that the proposed scoring strategy is highly correlated with real architecture accuracy. In the practical engineering application research, this strategy can search a high-performance architecture with an accuracy of 96.2% within 343.3 s, which proves that the proposed strategy can significantly improve the search efficiency in practical applications, reduce the subjectivity of artificial architecture design, and promote the application of practical time-consuming projects.

Automatic Design of Serial Linkage Using Virtual Screw Joint

Here, an automatic design method for a serial link mechanism is proposed. This method outputs all kinematic parameters of joints, position, orientation, number, and type of joint (revolute or prismatic). Several studies have been conducted on optimizing only the positions and directions of joints for a desired path. However, automatically determining the numbers and types of joints requires an excessive calculation time owing to the complicity of kinematics. To handle heavy computation, a virtual screw joint (VSJ) is introduced based on a screw axis and the product of exponentials formula. Screw joints have the advantage of including both rotation and translation. First, an additional joint is optimized as a VSJ. Then, adopting its position and orientation and selecting a revolute or prismatic joint facilitate an efficient design process. To demonstrate the effectiveness of this study, two task motions addressed in a related work are adopted as target paths. Consequently, the proposed method automatically generates serial linkages that contain both revolute and prismatic joints and can follow along desired paths.

Dispersion curve engineering for automated topology design of a unit cell in spoof surface plasmon polaritons

In this paper, an automatic design method is proposed for unit cell in spoof surface plasmon polaritons (SSPP) with an almost arbitrary dispersion curve. In this method, the pixel configuration is considered for the unit cell and, by using the binary particle swarm optimization method, the proper topology of the unit cell is explored so as to reach the target dispersion curve. Unlike the traditional method of controlling the dispersion curve, which is performed based on changing the geometric parameters of the predetermined unit cell, in this method, there is no need to know the shape of the unit cell, and the dispersion curve of the modes of SSPP unit cell can be controlled independently with more freedom. Two unit cell samples are designed in order to show the efficiency of the procedure. In the first sample, the dispersion curve is designed to have the lowest asymptotic frequency; in the second sample, the dispersion curve of the second mode is controlled independently from the first mode and is changed arbitrarily. SSPP transmission lines which are related to the unit cells of the two samples are designed, and it is demonstrated that measurement and simulation results are greatly in line with each other.

Deep Learning for Polarization Optical System Automated Design

Aiming at the problem that traditional design methods make it difficult to control the polarization aberration distribution of optical systems quickly and accurately, this study proposes an automatic optimization design method for polarization optical systems based on deep learning. The unsupervised training model based on ray tracing and polarized ray tracing was constructed by learning the reference lens structural feature data from the optical lens library, and the generalization ability of the deep neural network model was improved to achieve the automatic optimization design of the polarized optical system. The design results show that the optical system structure optimized by the network model is close to the reference lens in the full field of view and the full spectrum and that the optical system structure can be designed for different focal length requirements. The success rate of 1 million groups of initial structures designed is better than 96.403%, and the polarization effect of the optical system is effectively controlled. The proposed deep learning approach to optical design provides a new solution for future complex optical systems and also provides an effective way to improve the design accuracy of special optical systems such as polarization optical systems.