Existing Integration Methods Research Articles

A New Framework for Integrating DNN-Based Geographic Simulation Models within GISystems

As a crucial spatial decision support tool, Geographic Information Systems (GISystems) are widely used in fields such as digital watersheds, resource management, environmental assessment, and regional governance, with their core strength lying in the integration of geographic simulation models from various disciplines, enabling the analysis of complex geographical phenomena and the resolution of comprehensive spatial problems. With the rapid advancement of artificial intelligence, deep neural network-based geographic simulation models (DNN-GSMs) have increasingly replaced traditional models, offering significant advantages in simulation accuracy and inference speed, and have become indispensable components in GISystems. However, existing integration methods do not adequately account for the specific characteristics of DNN-GSMs, such as their formats and input/output data types. To address this gap, we propose a novel tight integration framework for DNN-GSMs, comprising four key interfaces: the data representation interface, the model representation interface, the data conversion interface, and the model application interface. These interfaces are designed to describe spatial data, the simulation model, the adaptation between spatial data and the model, and the model’s application process within the GISystem, respectively. To validate the proposed method, we construct a spatial morphology simulation model based on CNN-LSTM, integrate it into a GISystem using the proposed interfaces, and conduct a series of predictive experiments on island morphology evolution. The results demonstrate the effectiveness of the proposed integration framework for DNN-GSMs.

Vul-LMGNNs: Fusing language models and online-distilled graph neural networks for code vulnerability detection

Code Language Models (codeLMs) and Graph Neural Networks (GNNs) are widely used in code vulnerability detection. However, a critical yet often overlooked issue is that GNNs primarily rely on aggregating information from adjacent nodes, limiting structural information transfer to single-layer updates. In code graphs, nodes and relationships typically require cross-layer information propagation to fully capture complex program logic and potential vulnerability patterns. Furthermore, while some studies utilize codeLMs to supplement GNNs with code semantic information, existing integration methods have not fully explored the potential of their collaborative effects.To address these challenges, we introduce Vul-LMGNNs that integrates pre-trained CodeLMs with GNNs, leveraging knowledge distillation to facilitate cross-layer propagation of both code semantic knowledge and structural information. Specifically, Vul-LMGNNs utilizes Code Property Graphs (CPGs) to incorporate code syntax, control flow, and data dependencies, while employing gated GNNs to extract structural information in the CPG. To achieve cross-layer information transmission, we implement an online knowledge distillation (KD) program that enables a single student GNN to acquire structural information extracted from a simultaneously trained counterpart through an alternating training procedure. Additionally, we leverage pre-trained CodeLMs to extract semantic features from code sequences. Finally, we propose an ”implicit-explicit” joint training framework to better leverage the strengths of both CodeLMs and GNNs. In the implicit phase, we utilize CodeLMs to initialize the node embeddings of each student GNN. Through online knowledge distillation, we facilitate the propagation of both code semantics and structural information across layers. In the explicit phase, we perform linear interpolation between the CodeLM and the distilled GNN to learn a late fusion model. The proposed method, evaluated across four real-world vulnerability datasets, demonstrated superior performance compared to 17 state-of-the-art approaches. Our source code can be accessed via GitHub: https://github.com/Vul-LMGNN/vul-LMGGNN.

A novel heating system of solar-assisted enhanced jet enthalpy air source heat pump for cold regions

Solar energy and air source heat pumps are common heat sources for clean heating, but at lower ambient temperatures, the performance of solar energy and air source heat pumps will decline significantly, which can be ameliorated by integrating the two heat sources. However, the range of solar radiation and ambient temperature variation is very wide, and the existing integration methods cannot guarantee that they are optimal over a wide range. Therefore, this paper proposes a novel integration method of solar energy and jet enthalpy air source heat pump suitable for cold regions, where solar energy is used to heat the refrigerant in the jet branch to increase the flow rate of the jet branch, which enhances the effect of jet enthalpy. Under varying outdoor temperatures and solar radiations, the performances of the new integrated system and the traditional integrated systems are compared by simulation method. The results show that within the scope of this study (outdoor temperatures ranging from −30 °C to 10 °C, and radiations from 200 to 800 W/m2), the heat output and coefficient of performance of the new system are higher than those of the existing integrated systems. At an outdoor temperature of −30 °C and a radiation of 500 W/m2, the new system increases heat output by 45.8 % and coefficient of performance by 22.5 % compared with the existing two systems. This research can promote the application of solar energy and air source heat pump heating in cold regions.

A DECOUPLED MAGNETIC INTEGRATION TECHNIQUE FOR DUAL-FREQUENCY TOPOLOGIES

The dual-frequency converter comprises a high-frequency buck unit and a low-frequency buck unit, which has the advantages of high efficiency and fast dynamic response. But also, the increase of magnetic parts leads to the power density of the converter being greatly reduced; for this problem, this paper proposes a three-section winding method to decouple the two inductors in the converter integrated into a magnetic core, given the integrated magnetic parts structure and establish its gyrator-capacitor model to derive the decoupling conditions. The flux density distribution of the magnetic components is analyzed by finite element simulation software and compared with the existing integration method; this decoupling integration method can make the flux distribution more uniform and improve the core utilization. The results show that the volume and weight are reduced by 31.2 % and 25.3 % compared with those of the separated magnetic parts, and the efficiency is also consistent with that of the separated magnetic parts, which verifies the correctness and feasibility of the theoretical analysis.

A coherent integration method of an active sonar for maneuvering turning target detection.

The detection probability of underwater weak targets using active sonar is low, and inter-pulse coherent integration can improve the signal-to-noise ratio of echoes. When a target executes a maneuvering turn, complex range and Doppler frequency migrations occur during the coherent integration time that decrease the coherent integration gain. Most existing integration methods simplify the target motion to a finite-order polynomial model but fail to integrate a maneuvering turning target (MTT) due to model mismatch. Hence, this study proposes an underwater MTT integration method based on the modified Radon-Fourier transform. The proposed method constructs a theoretically accurate motion model for the MTT and a phase compensation function to compensate for the Doppler frequency migration. Furthermore, it yields a well-focused integration peak in the range-velocity and offset angle-turn rate dimensions and accurately estimates the target motion parameters. Moreover, the proposed method is suitable for targets with radial and oblique uniform motions. The effectiveness of the proposed method is demonstrated through simulations and a lake test. The proposed method demonstrates good integration performance, with an integration gain approximately 4-7 dB higher than that of traditional methods when using 30 integration pulses.

Efficient numerical simulation of fractional-order Van der Pol impulsive system

This paper presents an innovative and efficient method for solving the fractional-order Van der Pol impulsive system. In particular, the proposed scheme utilizes finite difference techniques for approximating fractional integrals, and its efficacy is compared to existing integration methods presented in the literature. Moreover, the proposed approach is applied to fractional impulsive systems, specifically the Fractional Van der Pol system with impulse behavior. The results demonstrate the effectiveness of the impulsive treatment effects for the system under consideration. In general, this study offers an insightful contribution to the field of fractional calculus, while providing a practical and efficient solution for solving impulsive systems.

GINGER: an integrated method for high-accuracy prediction of gene structure in higher eukaryotes at the gene and exon level.

The prediction of gene structure within the genome sequence is the starting point of genome analysis, and its accuracy has a significant impact on the quality of subsequent analyses. Gene structure prediction is roughly divided into RNA-Seq-based methods, ab initio-based methods, homology-based methods, and the integration of individual prediction methods. Integrated methods are mainstream in recent genome projects because they improve prediction accuracy by combining or taking the best individual prediction findings; however, adequate prediction accuracy for eukaryotic species has not yet been achieved. Therefore, we developed an integrated tool, GINGER, that solves various issues related to gene structure prediction in higher eukaryotes. By handling artefacts in alignments of RNA and protein sequences, reconstructing gene structures via dynamic programming with appropriately weighted and scored exon/intron/intergenic regions, and applying different prediction processes and filtering criteria to multi-exon and single-exon genes, we achieved a significant improvement in accuracy compared to the existing integration methods. The feature of GINGER is its high prediction accuracy at the gene and exon levels, which is pronounced for species with more complex gene architectures. GINGER is implemented using Nextflow, which allows for the efficient and effective use of computing resources.

Encoder–decoder-based image transformation approach for integrating multiple spatial forecasts

As the damage caused by heavy rainfall worsens, there is a growing demand for improved forecasts. One practical approach to address this issue is the linear integration of multiple existing forecasts, which allows for visualizing the contribution of each forecast at different locations. However, current methods such as arithmetic and Bayesian averages utilize a single weight shared across the entire space, making it difficult to account for local variations in importance. Additionally, while U-Net-based spatial forecasts have been proposed, they are limited to short-term predictions and do not facilitate the visualization of individual forecast contributions due to their non-linear processes. To overcome these challenges, we propose a new integration framework based on U-Net image transformation. This framework generates weight images that dynamically integrate forecasts based on both time and location. To effectively handle large and imbalanced precipitation data, we introduce novel extensions to the U-Net model. These extensions address heavily imbalanced precipitation data and enable position and time-dependent integration. Experimental results using real precipitation forecast data in Japan demonstrate that our proposed method outperforms existing integration methods.

Demand planning for the digital supply chain: How to integrate human judgment and predictive analytics

AbstractOur research examines how to integrate human judgment and statistical algorithms for demand planning in an increasingly data‐driven and automated environment. We use a laboratory experiment combined with a field study to compare existing integration methods with a novel approach: Human‐Guided Learning. This new method allows the algorithm to use human judgment to train a model using an iterative linear weighting of human judgment and model predictions. Human‐Guided Learning is more accurate vis‐à‐vis the established integration methods of Judgmental Adjustment, Quantitative Correction of Human Judgment, Forecast Combination, and Judgment as a Model Input. Human‐Guided Learning performs similarly to Integrative Judgment Learning, but under certain circumstances, Human‐Guided Learning can be more accurate. Our studies demonstrate that the benefit of human judgment for demand planning processes depends on the integration method.

Variational Direct Modeling: A Framework Towards Integration of Parametric Modeling and Direct Modeling in CAD

Feature-based parametric modeling is the de facto standard in CAD. Boundary representation-based direct modeling is another CAD paradigm developed recently. They have complementary advantages and limitations, thereby offering huge potential for improvement towards an integrated CAD modeling scheme. Most existing integration methods are developed by industry and typically treat direct edits as pseudo-features, where little can be said about seamless integration. This paper presents an alternative method for seamless parametric/direct integration, which allows parametric and direct edits to work in a unified way. The fundamental issues and challenges of parametric/direct integration are first explained. A framework is then proposed to handle those information inconsistencies, based on a detection-then-resolution strategy. Algorithms that can systematically detect and resolve all possible types of information inconsistencies are also given to implement the framework. With them, model validity can be maintained during the whole model editing process, and then the discrepancy between direct edits and parametric edits can be resolved. The effectiveness of the proposed approach has been shown with a series of case studies and comparisons, based on a preliminary prototype.

Circular Economy Trends – Potential Role of Emerging Technologies

The circular economy and digital transformation are two of the major trends over the last decade. Integrative methodological advances such as life cycle assessments, material flow analysis, and input-output tables are some of the current trends in circular economy case studies and scenarios. However, more efficient processes are required, and methods need to be adapted to the unique attributes of circular economy systems. This paper presents a descriptive analysis of current technological trends and topics in the circular economy. A scoping review and an automated content analysis were conducted in over 6000 abstracts available in Springer journals. Strategies such as applying data-driven design in the field of circular economy and using innovative information and communication technologies (ICTs) offer new possibilities for optimizing existing integration methods. New circular economy approaches, and systems could emerge based on taking advantage of technologies such as: artificial intelligence (AI), Internet of Things (IoT), Advanced Data Analytics, etc. Also, emerging topics in circular economy technologies focus on energy, policy, models, and global systems. The paper concludes with an outline of emerging technologies and identifies several future research directions.

Synchronous Integration Method of System and Simulation Models for Mechatronic Systems Based on SysML

System architecture could be described clearly and unambiguously by a descriptive system model during the system-level design of mechatronic products. However, for system verification, it is necessary to integrate the system modeling tools and simulation tools for analysis. Most of the existing integration methods focus on the integration in the physical architecture phase of the system model. A model integration framework that can integrate simulation into the logical architecture phase of the entire model for synchronous model verification is proposed in this paper. In the integration process, a Model-to-Text method is employed for information extraction and model transformation from the SysML model to the simulation model. The system model can be transformed by reusing and updating the generated simulation model through the framework for model verification. A framework with the Simulink model as the transformation target is constructed as an example in this paper, and the capability of this framework is verified by a design process of a quadruped robot. The result shows that this framework can realize the system design paradigm of integrating system verification for the entire logical architecture, which can improve the design efficiency and provide a reference for other system design and integration work.

Deep knowledge integration of heterogeneous features for domain adaptive SAR target recognition

How to integrate various heterogeneous features for better recognition performance is increasingly critical for automatic target recognition. Existing integration methods present the following drawbacks: (1) most feature integration methods ignore the information, both common and discriminate knowledge, among different types of features; (2) most decision integration methods ignore the fact that different knowledge contributes differently; (3) the feature weights of integration model learned in the source domain cannot perform well in the target domain. To tackle these problems, we propose a deep Knowledge Integration framework by combining heterogeneous features for Domain Adaptive synthetic aperture radar (SAR) target recognition (KIDA). In the training phase, we implement deep knowledge integration at both feature and decision levels. At the feature level, to exploit the common and discriminative knowledge, multiple heterogeneous features are projected from the feature space into a unified label space by exploring the shared and specific structures simultaneously. The shared structure integrates common information in different features, while the specific structure reserves discriminative information of each type of feature. At the decision level, to reveal the relative importance of different knowledge, a decision integration strategy with feature weights is adopted in the label space. In the online testing phase, to improve the generalization of the model in dynamical environments, we employ online learning with sequential target domain knowledge to update the feature weights, thus achieving domain adaptation. Extensive experiments on different datasets validate the effectiveness and advantages of the proposed KIDA, especially in noisy environments.

Steady-State Stand-Alone Power Flow Solvers for Integrated Transmission-Distribution Networks: A Comparison Study and Numerical Assessment

This paper compares and assesses several numerical methods that solve the steady-state power flow problem on integrated transmission-distribution networks. The integrated network model consists of a balanced transmission and an unbalanced distribution network. It is important to analyze these integrated electrical power systems due to the changes related to the energy transition. We classified the existing integration methods as unified and splitting methods. These methods can be applied to homogeneous (complete three-phase) and hybrid (single-phase/three-phase) network models, which results in four approaches in total. These approaches were compared on their accuracy and numerical performance—CPU time and number of iterations—to demonstrate their applicability on large-scale electricity networks. Furthermore, their sensitivity towards the amount of distributed generation and the addition of multiple distribution feeders was investigated. The methods were assessed by running power flow simulations using the Newton–Raphson method on several integrated power systems up to 25,000 unknowns. The assessment showed that unified methods applied to hybrid networks performed the best on these test cases. The splitting methods are advantageous when complete network data sharing between system operators is not allowed. The use of high-performance techniques for larger test cases containing multiple distribution networks will make the difference in speed less significant.

Revisiting integration in the material point method

The material point method (MPM) recently demonstrated its efficacy at simulating many materials and the coupling between them on a massive scale. However, in scenarios containing debris, MPM manifests more dissipation and numerical viscosity than traditional Lagrangian methods. We have two observations from carefully revisiting existing integration methods used in MPM. First, nearby particles would end up with smoothed velocities without recovering momentum for each particle during the particle-grid-particle transfers. Second, most existing integrators assume continuity in the entire domain and advect particles by directly interpolating the positions from deformed nodal positions, which would trap the particles and make them harder to separate. We propose an integration scheme that corrects particle positions at each time step. We demonstrate our method's effectiveness with several large-scale simulations involving brittle materials. Our approach effectively reduces diffusion and unphysical viscosity compared to traditional integrators.

A truly self-starting implicit family of integration algorithms with dissipation control for nonlinear dynamics

In this paper, a novel implicit family of composite two sub-step algorithms with controllable dissipations is developed to effectively solve nonlinear structural dynamic problems. The primary superiority of the present method over other existing integration methods lies that it is truly self-starting and so the computation of initial acceleration vector is avoided, but the second-order accurate acceleration responses can be provided. Besides, the present method also achieves other desired numerical characteristics, such as the second-order accuracy of three primary variables, unconditional stability and no overshoots. Particularly, the novel method achieves adjustable numerical dissipations in the low and high frequency by controlling its two algorithmic parameters ( $$ \gamma $$ and $$ \rho _{\infty }$$ ). The classical dissipative parameter $$ \rho _{\infty }$$ determines numerical dissipations in the high-frequency while $$ \gamma $$ adjusts numerical dissipations in the low-frequency. Linear and nonlinear numerical examples are given to show the superiority of the novel method over existing integration methods with respect to accuracy and overshoot.

A Selective Review of Multi-Level Omics Data Integration Using Variable Selection.

High-throughput technologies have been used to generate a large amount of omics data. In the past, single-level analysis has been extensively conducted where the omics measurements at different levels, including mRNA, microRNA, CNV and DNA methylation, are analyzed separately. As the molecular complexity of disease etiology exists at all different levels, integrative analysis offers an effective way to borrow strength across multi-level omics data and can be more powerful than single level analysis. In this article, we focus on reviewing existing multi-omics integration studies by paying special attention to variable selection methods. We first summarize published reviews on integrating multi-level omics data. Next, after a brief overview on variable selection methods, we review existing supervised, semi-supervised and unsupervised integrative analyses within parallel and hierarchical integration studies, respectively. The strength and limitations of the methods are discussed in detail. No existing integration method can dominate the rest. The computation aspects are also investigated. The review concludes with possible limitations and future directions for multi-level omics data integration.

The integration of weighted gene association networks based on information entropy.

Constructing genome scale weighted gene association networks (WGAN) from multiple data sources is one of research hot spots in systems biology. In this paper, we employ information entropy to describe the uncertain degree of gene-gene links and propose a strategy for data integration of weighted networks. We use this method to integrate four existing human weighted gene association networks and construct a much larger WGAN, which includes richer biology information while still keeps high functional relevance between linked gene pairs. The new WGAN shows satisfactory performance in disease gene prediction, which suggests the reliability of our integration strategy. Compared with existing integration methods, our method takes the advantage of the inherent characteristics of the component networks and pays less attention to the biology background of the data. It can make full use of existing biological networks with low computational effort.

A State-of-the-Art Review on the Integration of Building Information Modeling (BIM) and Geographic Information System (GIS)

The integration of Building Information Modeling (BIM) and Geographic Information System (GIS) has been identified as a promising but challenging topic to transform information towards the generation of knowledge and intelligence. Achievement of integrating these two concepts and enabling technologies will have a significant impact on solving problems in the civil, building and infrastructure sectors. However, since GIS and BIM were originally developed for different purposes, numerous challenges are being encountered for the integration. To better understand these two different domains, this paper reviews the development and dissimilarities of GIS and BIM, the existing integration methods, and investigates their potential in various applications. This study shows that the integration methods are developed for various reasons and aim to solve different problems. The parameters influencing the choice can be summarized and named as “EEEF” criteria: effectiveness, extensibility, effort, and flexibility. Compared with other methods, semantic web technologies provide a promising and generalized integration solution. However, the biggest challenges of this method are the large efforts required at early stage and the isolated development of ontologies within one particular domain. The isolation problem also applies to other methods. Therefore, openness is the key of the success of BIM and GIS integration.

Efficient numerical integration of arbitrarily broken cells using the moment fitting approach

AbstractThe finite cell method is based on a fictitious domain approach, providing a simple and fast mesh generation of structures with complex geometries. However, this simplification leads to intersected cells where the standard Gauss quadrature does not perform well. To perform the numerical integration of these cells, we use the moment fitting approach that generates an individual quadrature rule for every broken cell. In this paper, we will perform a non‐linear optimization approach to find the optimal position and number of the integration points. The findings show that the proposed method leads to efficient quadrature rules that require less integration points than other existing integration methods. (© 2016 Wiley‐VCH Verlag GmbH & Co. KGaA, Weinheim)