Space Model Research Articles

Topological consistency with low imperceptibility for graph adversarial attacks

Recent research shows that graph neural networks (GNNs) are easy to receive disruptions due to the lack of robustness, the phenomenon that poses a serious security threat. Currently, most efforts to attack GNNs mainly use gradient information to guide the attacks. However, the unreliability of gradient information, and the perceptibility of adversarial examples pose challenges that impede further progress in researching on graph adversarial attacks. From the unreliability of gradient information, we propose a Graph Distance Topological Consistency (GDTC). The scheme introduces graph connectivity, geodesic distance, cosine similarity, and Minkowski distance to construct the similarity matrices of the input space and embedding space of the surrogate model. The difference between the two similarities matrices is constrained during the training process of the surrogate model so that the surrogate model fully learns the topology of the original graph. From adversarial examples perceptibility, we propose attack loss with homogeneity restriction. Experiments show that GDTC can study the topological information of the original graph, enhance the reliability of gradient information, and significantly boost attack performance.

Structure-based out-of-distribution (OOD) materials property prediction: a benchmark study

In real-world materials research, machine learning (ML) models are usually expected to predict and discover novel exceptional materials that deviate from the known materials. It is thus a pressing question to provide an objective evaluation of ML model performances in property prediction of out-of-distribution (OOD) materials that are different from the training set. Traditional performance evaluation of materials property prediction models through the random splitting of the dataset frequently results in artificially high-performance assessments due to the inherent redundancy of typical material datasets. Here we present a comprehensive benchmark study of structure-based graph neural networks (GNNs) for extrapolative OOD materials property prediction. We formulate five different categories of OOD ML problems for three benchmark datasets from the MatBench study. Our extensive experiments show that current state-of-the-art GNN algorithms significantly underperform for the OOD property prediction tasks on average compared to their baselines in the MatBench study, demonstrating a crucial generalization gap in realistic material prediction tasks. We further examine the latent physical spaces of these GNN models and identify the sources of CGCNN, ALIGNN, and DeeperGATGNN’s significantly more robust OOD performance than those of the current best models in the MatBench study (coGN and coNGN) as a case study for the perovskites dataset, and provide insights to improve their performance.

PT -symmetry-enabled stable modes in a multicore fiber

Open systems with balanced gain and loss, described by parity-time (PT-symmetric) Hamiltonians have been deeply explored over the past decade. Most explorations are limited to finite discrete models (in real or reciprocal spaces) or continuum problems in one dimension. As a result, these models do not leverage the complexity and variability of two-dimensional continuum problems on a compact support. Here, we investigate eigenvalues of the Schrödinger equation on a disk with zero boundary condition, in the presence of constant, PT-symmetric, gain-loss potential that is confined to two mirror-symmetric disks. We find a rich variety of exceptional points, re-entrant PT-symmetric phases, and a nonmonotonic dependence of the PT-symmetry breaking threshold on the system parameters. By comparing results of two model variations, we show that this simple model of a multicore fiber supports propagating modes in the presence of gain and loss. Published by the American Physical Society 2024

Research on Zoning and Carbon Sink Enhancement Strategies for Ecological Spaces in Counties with Different Landform Types

Ecological carbon sinks, pivotal in mitigating carbon emissions, are indispensable for climate change mitigation. Counties, as the fundamental units of ecological space management, directly impact the achievement of regional dual carbon targets through their levels of carbon sink. However, existing research has overlooked the intricate relationship between terrain features and ecological spaces, leading to a lack of specific guidance on enhancing the carbon sink for counties with diverse landform characteristics. This study focused on Jingbian County (Loess Plateau), Fuping County (Guanzhong Plain), and Chenggu County (Qinba Mountains), each characterized by distinct landform characteristics. This study proposes a comprehensive identification model for ecological space within the context of dual carbon targets. Utilizing this model as a basis, the land use structure, carbon sink potential, and ecological spatial patterns of different counties were systematically analyzed. The results indicated substantial disparities in land use structure, carbon sink capabilities, and ecological space distributions among counties with different landform types. Specifically, Jingbian County was predominantly covered by grassland, exhibiting a moderate overall carbon sink capacity, with baseline ecological spaces playing a significant role. Conversely, Fuping County, dominated by cultivated land and construction land, exhibited the lowest carbon sink capacity, with non-ecological spaces accounting for a staggering 85.93%. Chenggu County, on the other hand, was characterized by the dominance of forestland, with nearly all its carbon sink originating from forestland, and core ecological spaces occupying a leading position. Tailored optimization strategies are recommended based on varying terrain features: Jingbian County should prioritize ecosystem restoration and conservation, while Fuping County should concentrate on optimizing land use structure and promoting urban greening. Reinforcing the carbon sink capacity of existing ecosystems is crucial for Chenggu County. This study broadens the perspective on ecological space optimization and provides scientific guidance and pragmatic insights tailored to regional disparities, which are instrumental in assisting various regions to achieve their dual carbon targets.

Linear control strategy of the dilution rate for stability in the anaerobic digestion process

Anaerobic digestion is a biochemical process of decomposition of organic matter, mediated by anaerobic bacteria, in an oxygen-free environment, from which biogas is produced. Biogas is used to generate thermal and electrical energy, where environmental conditions such as pH, temperature, concentration of volatile fatty acids, have a great influence on the efficiency and performance of the anaerobic digestion process, and can cause the process to be ruined. This document presents a control strategy that seeks to regulate the conditions of the anaerobic digestion process to keep it within the stability zone, using a linearization technique of the model in the state space for the design of controllers. In this work, a compensator is designed, for which the approximate linearization technique of the AM2 nonlinear model is proposed, to obtain the linearized model from which a classic lag-lead control strategy is proposed, using the dilution rate as a control variable. As a result, an improvement in system performance is obtained by reducing the maximum overshoot and reducing the settling time, evaluated at the operating point, which is validated through simulation and analysis of the response provided by the nonlinear model at the point of operation.

A Bayesian non-stationary heteroskedastic time series model for multivariate critical care data.

We propose a multivariate GARCH model for non-stationary health time series by modifying the observation-level variance of the standard state space model. The proposed model provides an intuitive and novel way of dealing with heteroskedastic data using the conditional nature of state-space models. We follow the Bayesian paradigm to perform the inference procedure. In particular, we use Markov chain Monte Carlo methods to obtain samples from the resultant posterior distribution. We use the forward filtering backward sampling algorithm to efficiently obtain samples from the posterior distribution of the latent state. The proposed model also handles missing data in a fully Bayesian fashion. We validate our model on synthetic data and analyze a data set obtained from an intensive care unit in a Montreal hospital and the MIMIC dataset. We further show that our proposed models offer better performance, in terms of WAIC than standard state space models. The proposed model provides a new way to model multivariate heteroskedastic non-stationary time series data. Model comparison can then be easily performed using the WAIC.

Wind and Wave-Induced Vibration Reduction Control for Floating Offshore Wind Turbine Using Delayed Signals

Active vibration control is a critical issue of the wind turbine in the field of marine energy. First, based on a three-degree-of-freedom wind turbine, a state space model subject to wind and wave loads is obtained. Then, a delayed state feedback control scheme is illustrated to reduce the vibration of platform pitch angle and tower top foreaft displacement, where the control channel includes time-delay state signals. The designed controller’s existence conditions are investigated. The simulation results show that the delayed feedback H∞ controller can significantly suppress wind- and wave-induced vibration of the wind turbine. Furthermore, it presents potential advantages over the delay-free feedback H∞ controller and the classic linear quadratic regulator in two aspects: vibration control performance and control cost.

Sediment Cover Modulates Landscape Erosion Patterns and Channel Steepness in Layered Rocks: Insights From the SPACE Model

AbstractErosional perturbations from changes in climate or tectonics are recorded in the profiles of bedrock rivers, but these signals can be challenging to unravel in settings with non‐uniform lithology. In layered rocks, the surface lithology at a given location varies through time as erosion exposes different layers of rock. Recent modeling studies have used the Stream Power Model (SPM) to highlight complex variations in erosion rates that arise in bedrock rivers incising through layered rocks. However, these studies do not capture the effects of coarse sediment cover on channel evolution. We use the “Stream Power with Alluvium Conservation and Entrainment” (SPACE) model to explore how sediment cover influences landscape evolution and modulates the topographic expression of erodibility contrasts in horizontally layered rocks. We simulate river evolution through alternating layers of hard and soft rock over million‐year timescales with a constant and uniform uplift rate. Compared to the SPM, model runs with sediment cover have systematically higher channel steepness values in soft rock layers and lower channel steepness values in hard rock layers. As more sediment accumulates, the contrast in steepness between the two rock types decreases. Effective bedrock erodibilities back‐calculated assuming the SPM are strongly influenced by sediment cover. We also find that sediment cover can significantly increase total relief and timescales of adjustment toward landscape‐averaged steady‐state topography and erosion rates.

Uranus and Neptune as methane planets: Producing icy giants from refractory planetesimals

Uranus and Neptune are commonly considered ice giants, and it is often assumed that, in addition to a solar mix of hydrogen and helium, they contain roughly twice as much water as rock. This classical picture has led to successful models of their internal structure and has been understood to be compatible with the composition of the solar nebula during their formation (Reynolds and Summers, 1965; Podolak and Cameron, 1974; Podolak and Reynolds, 1984; Podolak et al., 1995; Nettelmann et al., 2013). However, the dominance of water has been recently questioned (Teanby et al., 2020; Helled and Fortney, 2020; Podolak et al., 2022). Planetesimals in the outer solar system are composed mainly of refractory materials, leading to an inconsistency between the icy composition of Uranus and Neptune and the ice-poor planetesimals they accreted during formation (Podolak et al., 2022). Here we elaborate on this problem, and propose a new potential solution. We show that chemical reactions between planetesimals dominated by organic-rich refractory materials and the hydrogen in gaseous atmospheres of protoplanets can form large amounts of methane ’ice’. Uranus and Neptune could thus be compatible with having accreted refractory-dominated planetesimals, while still remaining icy. Using random statistical computer models for a wide parameter space, we show that the resulting methane-rich internal composition could be a natural solution, giving a good match to the size, mass and moment of inertia of Uranus and Neptune, whereas rock-rich models appear to only work if a rocky interior is heavily mixed with hydrogen. Our model predicts a lower than solar hydrogen to helium ratio, which can be tested. We conclude that Uranus, Neptune and similar exoplanets could be methane-rich, and discuss why Jupiter and Saturn cannot.

Theme park greening VR design based on entertainment robots and genetic algorithm optimization: Digital entertainment design experience

Theme parks should provide attractive green design and digital interactive experiences to attract tourists and improve their satisfaction. Therefore, this study aims to achieve a digital interactive experience in theme park greening design through the use of entertainment robots and genetic algorithm optimization. The study analyzed the characteristics of experiential theme parks and introduced landscape 3D modeling techniques for creating interactive 3D models of green spaces in theme parks. By using technologies such as sensors and cameras, entertainment robots can perceive the presence and behavior of tourists. Entertainment robots can interact with tourists in real-time, customize them according to their interests and needs, and provide unique and enjoyable experiences for tourists. By collecting and analyzing feedback and behavioral data from tourists, the advantages and improvement points of digital interactive experiences can be identified. Through comparative experimental analysis with traditional experience methods, it was found that digital interactive experience can significantly improve tourist participation and satisfaction. Through user data analysis of interactive experience, this design can effectively improve the greening design of theme parks, enhance tourist participation and entertainment experience.

Solution and numerical simulation of thermal-structural characteristics model of wire deflagration based on partition coupling time domain propulsion method

The dynamic analysis of transmission lines under the thermal shock load of mountain fire is a complex nonlinear coupling problem of thermal structure. Taking the transmission line as the research object, based on the temperature rise model of mountain fire deflagration space, considering the temperature characteristics of the wire, the mathematical model of the thermal structure coupling characteristics of the wire deflagration is established. In this paper, the partition coupling time domain propulsion method is used to solve the problem. Through MATLAB programming and ANSYS secondary development, the transient characteristics and dynamic response of the coupling effect of the heated structure of the LGJ-400/35 wire with a span of 100 m under different positions and numbers of deflagration points are simulated. It is concluded that the temperature rise rate is positively correlated with the displacement rate and negatively correlated with the stress reduction rate. When the temperature rise rate increases by 0.1 °C/s, the displacement rate increases by 0.0004 m/s. The global deflagration is the most unfavorable mountain fire condition, and the temperature rise rate and displacement rate of the wire under this condition are about 1.3 and 1.9 times that of each single deflagration point condition. The wire temperature is symmetrically distributed with the detonation point symmetrically arranged relative to the mid-span.

Three-way decision-based Takagi–Sugeno-Kang fuzzy classifier for partially labeled data

The salient features of Takagi–Sugeno-Kang (TSK) fuzzy classifiers are their superior nonlinear fitting capability and better interpretability, rendering them widely applicable in the domains of data mining and machine learning. However, using TSK fuzzy classifiers for partially labeled data has received little attention. Based on the three-way decision (TWD) theory, this study proposed a TSK fuzzy model to learn from partially labeled data. First, an adaptive sample weight strategy is introduced for the fuzzy partition of the antecedent part and the cross-entropy loss of the consequent part, respectively. Subsequently, a prototype-based firing strength loss mechanism is proposed to constrain the optimization of the model in the fuzzy representation space. It employs a TWD strategy based on the entropy criterion to classify samples as useful, useless, and uncertain, where the proposed model enhances the performance by iteratively utilizing a certain number of useful pseudo-labeled samples. Finally, the validity of the model is theoretically analyzed based on the principle of noise learning. The experimental results on UCI datasets demonstrate that the proposed model exhibits superior performance compared with classical semi-supervised methods, even attaining a performance comparable to that of the model trained on all data with ground-truth labels.

Ultra-fast semantic map perception model for autonomous driving

Autonomous driving relies on real-time perception of environmental semantic maps to make decisions, accurate and real-time perception of environmental semantics is crucial for safe navigation, as well as for efficient construction of high-definition maps in the backend. Our research found that when the output space of the semantic map perception model is an orthogonal projection subspace of the input perception space, it significantly improves the accuracy and real-time performance of map semantic segmentation while reducing learning pressure on deep network models. Based on this, we propose a simple yet efficient semantic map segmentation network that aims to achieve extremely fast speed and precision. Furthermore, we implement a semantic map fusion process using a Bayesian framework that promotes global semantic information accumulation in local semantics over time and space. Our experimental results demonstrate the superiority of our method, achieving state-of-the-art performance in network speed and mIoU metrics. At the same spatial resolution of semantic maps, our method reaches over 350+ frames per second, which is at least 15 times faster than the previous state-of-the-art method. Code released at https://github.com/xzx-ai/MDSegMap.

Ultra-fast semantic map perception model for autonomous driving

Autonomous driving relies on real-time perception of environmental semantic maps to make decisions, accurate and real-time perception of environmental semantics is crucial for safe navigation, as well as for efficient construction of high-definition maps in the backend. Our research found that when the output space of the semantic map perception model is an orthogonal projection subspace of the input perception space, it significantly improves the accuracy and real-time performance of map semantic segmentation while reducing learning pressure on deep network models. Based on this, we propose a simple yet efficient semantic map segmentation network that aims to achieve extremely fast speed and precision. Furthermore, we implement a semantic map fusion process using a Bayesian framework that promotes global semantic information accumulation in local semantics over time and space. Our experimental results demonstrate the superiority of our method, achieving state-of-the-art performance in network speed and mIoU metrics. At the same spatial resolution of semantic maps, our method reaches over 350+ frames per second, which is at least 15 times faster than the previous state-of-the-art method. Code released at https://github.com/xzx-ai/MDSegMap.

LPV interpolation modeling and modal-based pole placement control for ball screw drive with dynamic variations

This paper presents a linear parameter varying (LPV) interpolation modeling method and modal-based pole placement (PP) control strategy for the ball screw drive (BSD) with varying dynamics. The BSD is modeled as a global LPV model with position-load dependence by selecting position and load as scheduling variables. The global LPV model is obtained from local subspace closed-loop identification and LPV interpolation modeling. A modal-based global LPV model is obtained through the similarity transformation. Based on this model, a modal-based LPV PP control strategy is proposed to achieve various modal control. Specifically, a state feedback control structure with an LPV state observer is designed to realize online state estimation and real-time state feedback control of modal state variables which cannot be measured directly. The steady-state error is minimized by introducing an error state space (SS) model with the integral effects. Moreover, the stability of the closed-loop system is analyzed according to the controllable decomposition and principle of separation. It is experimentally demonstrated that the proposed modal-based LPV PP control strategy can effectively achieve precise tracking and outstanding robustness meantime.

HTQ: Exploring the High-Dimensional Trade-Off of mixed-precision quantization

Mixed-precision quantization, where more sensitive layers are kept at higher precision, can achieve the trade-off between accuracy and complexity of neural networks. However, the search space for mixed-precision grows exponentially with the number of layers, making the brute force approach infeasible on deep networks. To reduce this exponential search space, recent efforts use Pareto frontier or integer linear programming to select the bit-precision of each layer. Unfortunately, we find that these prior works rely on a single constraint. In practice, model complexity includes space complexity and time complexity, and the two are weakly correlated, thus using simply one as a constraint leads to sub-optimal results. Besides this, they require manually set constraints, making them only pseudo-automatic. To address the above issues, we propose High-dimensional Trade-off Quantization (HTQ), which automatically determines the bit-precision in the high-dimensional space of model accuracy, space complexity, and time complexity without any manual intervention. Specifically, we use the saliency criterion based on connection sensitivity to indicate the accuracy perturbation after quantization, which performs similarly to Hessian information but can be calculated quickly (more than 1000× speedup). The bit-precision is then automatically selected according to the three-dimensional (3D) Pareto frontier of the total perturbation, model size, and bit operations (BOPs) without manual constraints. Moreover, HTQ allows for the joint optimization of weights and activations, and thus the bit-precisions of both can be computed concurrently. Compared to state-of-the-art methods, HTQ achieves higher accuracy and lower space/time complexity on various model architectures for image classification and object detection tasks. Code is available at: https://github.com/zkkli/HTQ.

Design and implementation of a collaborative modeling architecture based on MDE (Multi-DDS Edge)

Abstract Virtual space is an integral part of the Metaverse. Typical virtual spaces are modeled step by step using artificial means. With the development needs of Metaverse modeling and simulation technology and the need for simulation fidelity and content richness in virtual space construction, efficient and correct modeling methods are required. It has become a bottleneck restricting the efficiency of virtual space modeling. This paper designs a multi-person collaborative virtual space modeling method, which improves the modeling efficiency, communication security, and interaction stability of the virtual space collaborative modeling method. Compared with existing collaborative modeling methods, this paper proposes a virtual space collaborative modeling method based on multi-edge DDS technology. By constructing MDE (multi-DDS edges) and MDE on user migration, collaborative construction in multiple virtual spaces is completed. Model data communication extends collaborative modeling technology from the virtual space creation stage to the virtual space simulation and deduction stage, and designs and verifies collaborative content, collaborative architecture, and collaborative communication. Experiments show that the architecture synergy efficiency is 145.6%, the architecture synergy efficiency is 212.5%, and the architecture synergy efficiency is 105.4%. As the number of people in collaborative modeling increases, the operational complexity and demonstration take a logarithmic form, meeting the needs of the future development of the metaverse.

A novel model-based Cauchy-Schwarz divergence condition indicator for gears monitoring during fluctuating speed conditions

Gear monitoring and fault diagnosis are vital for preventing accidents and minimizing economic losses in transportation and industrial systems. Traditional methods use vibration sensors and a two-stage analysis approach: preprocessing data to remove noise and extract relevant components, and generating a condition indicator to detect behavioral anomalies in the gears over time.Time synchronous averaging is a notable tool for monitoring gears at constant speeds. Such a tool filters sensor signals and extracts rotation-related components by using statistical measurements as condition indicators. However, it has limitations in scenarios with time-varying sampling rates and fluctuating speeds, where statistical measures may not fully capture changes in system parameters.This article proposes a novel methodology for monitoring gears in multivariate rotordynamical systems under fluctuating speed conditions. The method integrates time synchronous averaging, system identification algorithms, and statistical tools. It generates a time-synchronous average signal considering speed fluctuations, computes a state–space model of gear behavior in healthy states, extracts residual data from a data-driven model, and generates a condition indicator based on the Cauchy–Schwarz divergence.The proposed methodology was evaluated using experimental data from three rotor dynamical setups under different operational conditions. Validation showed its effectiveness, especially under high-load conditions with significant speed fluctuations.

Color Image Hybrid Noise Filtering Algorithm based on Deep Convolution Neural Network

To solve the problems of the classical color image hybrid noise filtering method, a deep convolutional neural network improved by evolutionary strategy and jump connection is proposed and applied to the filtering noise reduction of color images. First, the color information of the image is described quantitatively by digital means. The common method is to build color space model. According to the characteristics of color and the needs of human vision, mathematical algorithms are used to convert images into machine recognizable data. The distance between pixels is measured according to the difference of pixels in the color image determined above. Then, the probability density function and noise probability density function of Gaussian noise are calculated to determine the hybrid noise feature points of color image. The filtering algorithm structure designed this time is as follows: A color image hybrid noise filter is used to map the noise points in the mapped image to the feature space, and linear regression is performed on the noise point data. Relaxation variables are introduced in the network to improve the denoising ability. The experimental results show that the Peak Signal to Noise Ratio and structural similarity index values of the filtering algorithm designed in this study are higher than the two methods in the literature. The color image hybrid noise filtering model designed in this study has good filtering performance, good image cleanliness, and high filtering efficiency.

Spatio-temporal monitoring of plant water status using optical remote sensing data and in situ measurements

Water content as an important physiological status variable in plants, is closely linked to transpiration, photosynthesis, water stress, and biomass productivity. Obtaining plant water information that provides sufficient spatial and temporal resolution for such applications remains a challenge. In this study, the data distribution space of fractional vegetation cover (FVC) and shortwave infrared transformed reflectance (STR) with linear and nonlinear edges was used to monitor plant water content. Four indicators of plant water status, equivalent water thickness (EWT), leaf water content (LWC), relative water content (RWC), and leaf water potential (LWP), were measured in a 10-ha corn field at the Karkaj Agricultural Research Station of Tabriz University, Tabriz, Iran. Furthermore, the Sentinel-2 Level 2 prototype processor (SL2P) was utilized to estimate EWT and compare the results with FVC-STR space. The FVC-STR space with nonlinear edges (FSNLE) provided better estimation accuracy of plant water status than the FVC-STR space with linear edges (FSLE). The root mean square errors of the EWT, LWC, RWC and LWP estimates for FSLE were 0.00306 g/cm2, 4.03%, 6.56%, and 0.38 bar, respectively, while those for FSNLE were 0.00303 g/cm2, 3.75%, 5.57%, and 0.37 bar, respectively. In addition, the R2 value for FSNLE was higher than that for FSLE (0.43–0.70 vs. 0.39–0.64). The RMSE and R2 for SL2P were 0.0041 g/cm2 and 0.401, respectively. Among the four measured indicators, the highest and lowest estimation accuracies in both FSLE and FSNLE were obtained with EWT and RWC, respectively. It can be concluded that FVC-STR space model based on Sentinel-2 imagery data provide acceptable accuracy for estimating plant water content. The FVC-STR space with nonlinear edges provided a better estimation accuracy for plant water indicators than the FVC-STR space with linear edges.