Differential Strategies Research Articles

LSMOF-AD: Three-Stage Optimization Approach with Adaptive Differential for Large-Scale Container Scheduling

Container technology has gained a widespread application in cloud computing environments due to its low resource overhead and high flexibility. However, as the number of containers grows, it becomes increasingly challenging to achieve the rapid and coordinated optimization of multiple objectives for container scheduling, while maintaining system stability and security. This paper aims to overcome these challenges and provides the optimal allocation for a large number of containers. First, a large-scale multi-objective container scheduling optimization model is constructed, which involves the task completion time, resource cost, and load balancing. Second, a novel optimization algorithm called LSMOF-AD (large-scale multi-objective optimization framework with muti-stage and adaptive differential strategies) is proposed to effectively handle large-scale container scheduling problems. The experimental results show that the proposed algorithm has a better performance in multiple benchmark problems compared to other advanced algorithms and can effectively reduce the task processing delay, while achieving a high resource utilization and load balancing compared to other scheduling strategies.

Conditional Mixture Path Guiding for Differentiable Rendering

The efficiency of inverse optimization in physically based differentiable rendering heavily depends on the variance of Monte Carlo estimation. Despite recent advancements emphasizing the necessity of tailored differential sampling strategies, the general approaches remain unexplored. In this paper, we investigate the interplay between local sampling decisions and the estimation of light path derivatives. Considering that modern differentiable rendering algorithms share the same path for estimating differential radiance and ordinary radiance, we demonstrate that conventional guiding approaches, conditioned solely on the last vertex, cannot attain this density. Instead, a mixture of different sampling distributions is required, where the weights are conditioned on all the previously sampled vertices in the path. To embody our theory, we implement a conditional mixture path guiding that explicitly computes optimal weights on the fly. Furthermore, we show how to perform positivization to eliminate sign variance and extend to scenes with millions of parameters. To the best of our knowledge, this is the first generic framework for applying path guiding to differentiable rendering. Extensive experiments demonstrate that our method achieves nearly one order of magnitude improvements over state-of-the-art methods in terms of variance reduction in gradient estimation and errors of inverse optimization. The implementation of our proposed method is available at https://github.com/mollnn/conditional-mixture.

Nondestructive and accurate prediction of IGBT junction temperature in wind power converters: based on IDMOA-ELM prediction method

ABSTRACT This study has made significant advancements in predicting junction temperature of IGBTs in wind power converters. An improved dwarf mongoose optimisation algorithm (IDMOA) with superior performance was developed. The IDMOA improves the initial solution quality through Hammersley initialisation, balances global search and local optimisation with nonlinear dynamic decreasing weight factors, and enhances population diversity via differential evolution strategy to avoid local optima. The IDMOA optimizes random parameters of extreme learning machine (ELM), overcoming the limitations of traditional manual parameter setting and constructing a high-precision IDMOA optimised ELM (IDMOA-ELM) model. Simulation results indicate that the IDMOA-ELM excels in predicting the IGBT junction temperature, achieving a low RMSE value of 0.1034 ℃. This significantly surpasses existing technologies and provides robust support for temperature management in wind power converters. Furthermore, to verify the performance of the DMOA-ELM in practical application scenarios, IGBT accelerated ageing experiments were conducted, yielding a comprehensive validation dataset. The high accuracy and effectiveness of the IDMOA-ELM were confirmed through comparative analysis with experimental data, enhancing the credibility of the research. This study achieved non-destructive and high-precision prediction of IGBT junction temperature in wind power converters and significantly improving the reliability of wind power converters.

Comparative analysis of time-frequency transform-based differential protection strategy for distributed generation integrated microgrid

This study presents a differential protection strategy for distributed generation-integrated power systems. It explores and compares two enhanced traditional signal processing methods: the Hilbert-Huang Transform (HHT) and the S-transform. The HHT has been improved by incorporating Variational Mode Decomposition (VMD), while the S-transform has been augmented with a Fourier component to create the Fast Fourier Discrete Orthonormal Stockwell Transform (FFDOST). VMD and FFDOST are applied to three-phase current signals from microgrid buses to calculate the Spectral Energy (SE). Fault detection is based on Differential Energy (DE), which measures the difference in SE between two specific faulty buses, with a tripping signal activated when DE surpasses a predetermined threshold. The effectiveness of this approach is rigorously tested and validated on a typical DG-integrated microgrid that includes synchronous generators and wind farms. A comprehensive set of case studies conducted on an IEC microgrid model examines various fault types (symmetrical and asymmetrical) across different operational modes (islanded and grid-connected) and configurations (radial and looped) under varying noise conditions. The performance of these two modified differential strategies is critically compared to determine the more effective method. Detailed sample-wise analysis and graphical presentations enrich the documentation of results. The VMD-based method outperforms the FFDOST approach, detecting faults 3 to 15 samples sooner and achieving quicker fault detection by approximately 0.05 milliseconds. Comparative analyses suggest that VMD is a faster, more secure, and highly reliable method for fault detection and mitigation, enhancing microgrids' operational security and reliability.

Mechanisms of biodiversity loss under nitrogen enrichment: unveiling a shift from light competition to cation toxicity.

The primary mechanisms contributing to nitrogen (N) addition induced grassland biodiversity loss, namely light competition and soil cation toxicity, are often examined separately in various studies. However, their relative significance in governing biodiversity loss along N addition gradient remains unclear. We conducted a 4-yr field experiment with five N addition rates (0, 2, 10, 20, and 50 g N m-2 yr-1) and performed a meta-analysis using global data from 239 observations in N-fertilized grassland ecosystems. Results from our field experiment and meta-analysis indicate that both light competition and soil cation (e.g. Mn2+ and Al3+) toxicity contribute to plant diversity loss under N enrichment. The relative importance of these mechanisms varied with N enrichment intensity. Light competition played a more significant role in influencing species richness under low N addition (≤ 10 g m-2 yr-1), while cation toxicity became increasingly dominant in reducing biodiversity under high N addition (>10 g m-2 yr-1). Therefore, a transition from light competition to cation toxicity occurs with increasing N availability. These findings imply that the biodiversity loss along the N gradient is regulated by distinct mechanisms, necessitating the adoption of differential management strategies to mitigate diversity loss under varying intensities of N enrichment.

A lift-off suppression algorithm to improve the accuracy of ACFM in crack detection of ferromagnetic materials

Ferromagnetic pressurized components are widely used in petroleum, aerospace and nuclear industries, and their safety and stability are the key to the reliable operation of pressure equipment. In the process of crack detection of ferromagnetic materials by ACFM technology, the changes in lift-off distance caused by coating corrosion and irregularity of the surface will greatly affect the extraction of useful detection signals and reduce the detection accuracy. To overcome the drawback, a lift-off suppression algorithm for crack detection of ACFM is proposed. First, a differential strategy is proposed to discover the time-domain lift-off point of intersection (LOI) from the raw detection signals of ACFM for ferromagnetic materials. Then, the corresponding Bx and Bz signals are demodulated based on the amplitude of the LOI. The simulation and experimental results of the surface cracking of ferromagnetic materials by ACFM show that the novel Bx signals obtained according to the proposed method vary in a small range for different lift-off distances. The lift-off invariant property of the LOI enables the method to improve the interference of the lift-off effect on the conventional Bx signals and to enhance the detection accuracy of crack depth of ferromagnetic materials. The novel Bz signals perform consistently with the conventional Bz signals, and are virtually unaffected by the changes in lift-off distance.

Self-immunity study of quadrotor UAV based on Modelica system modeling and disturbance feed-forward compensation

For addressing the challenges of decreased attitude and trajectory tracking accuracy and a delayed response in the flight control operations of quadcopter unmanned aerial vehicles (UAVs) under the uncertainties of model parameters and external disturbances, this study leverages the advantages of the non-causal declarative modeling language Modelica in system modeling and simulation. In addition, it incorporates the nonlinear Active Disturbance Rejection Control (ADRC) framework for disturbance observation, estimation, and compensation. A state observer is designed to mitigate the impact of external disturbances and model uncertainties through feed-forward compensation, and stability analysis is conducted. Numerical simulations for hover resistance demonstrate that, compared to the cascade proportional integral differential (PID) control strategy, PID-NLADRC reduces the maximum deviation induced by wind disturbances by ∼50% and shortens the disturbance influence time by around 40%. Simulations for different trajectories, such as planar or spatial, smooth or abrupt changes, indicate that under the PID-NLADRC control strategy, the real-time spatial distance deviation mean is reduced by 69.5%, and the peak time is shortened by 75.7%. Validation through multi-objective applications and physical experiments highlights the advantages of PID-NLADRC in terms of positioning accuracy, rapid tracking, and disturbance suppression, aligning well with the fast, precise, and robust flight control requirements of quadcopter UAVs.

Incidence of pregnancy related pulmonary embolism in Spain 2016-2021: an observational population-based retrospective study

ObjectiveThis study aimed to estimate the overall and annual age-standardized incidence of pregnancy-related pulmonary embolism (PE) in Spain from 2016 to 2021, explore the distribution of PE events during pregnancy and the postpartum period, identify potential risk factors, and estimate mortality rates during hospital admission. MethodsIn a retrospective, observational, population-based study, data from the Spanish National Hospital Discharge Database were analyzed to identify women with hospital episodes of pregnancy-related-PE. The primary outcome was the overall and annual age-standardized incidence of pregnancy-related-PE, with secondary aims including the distribution of events during pregnancy and postpartum and the calculation of age-standardized mortality rates during admission. ResultsAmong 2,178,805 births from 2016 to 2021, 522 women were diagnosed with pregnancy-related PE, yielding an overall age-standardized incidence of 2.83 cases per 10,000 births. A non-significant increasing trend was observed from 2.43 to 4.18 cases per 10,000 births (p = 0.06). Comorbidities were low, with a notable association between PE and SARS-CoV-2 infection during the last two years. The mortality rate among women with pregnancy-related PE was 2.8%, with a higher incidence of PE reported during the postpartum period. ConclusionThe incidence of pregnancy-related-PE in Spain exhibits a non-significant increasing trend, with a significant risk of mortality. The association with SARS-CoV-2 infection underscores the importance of vigilant monitoring and management of pregnant women, particularly during pandemics. This study contributes specific data on the incidence and characteristics of pregnancy-related-PE in Spain, emphasizing the need to consider PE in the differential diagnosis and management strategies for pregnant and postpartum women.

Inversion strategies for Q estimation in viscoacoustic full-waveform inversion

Estimation of an attenuation parameter, represented by the quality factor Q, holds paramount importance in seismic exploration. One of the main challenges in Q estimation through viscoacoustic full-waveform inversion (FWI) is effectively decoupling Q from velocity. In this study, our objective is to enhance Q inversion by addressing critical aspects, such as gradient preconditioning, workflow, and misfit design. By developing a new preconditioner that approximates the diagonal of the Hessian, we facilitate automatic parameter tuning across different classes, ensuring comparable magnitudes of preconditioned gradients for velocity and Q. Moreover, our investigations confirm the efficacy of the two-stage hierarchical strategy in mitigating velocity- Q trade-offs, enabling more accurate Q estimation by first focusing on velocity reconstruction before jointly estimating velocity and Q. The analysis and numerical examples also highlight the importance of broadband data and long-offset acquisition for a reliable Q estimation. In addition, leveraging amplitude information can improve Q estimation to some extent, but careful consideration of frequency band and noise effects is necessary. We explore two misfit functions that capture amplitude variation with frequency in the time-frequency domain, noting their sensitivity to noise. To address this, we develop a differential strategy that can effectively mitigate the effects of low-frequency noise. This comprehensive study on enhancing Q estimation in viscoacoustic FWI offers valuable insights for multiparameter inversion in realistic scenarios.

Synchronizing real-time and high-precision LDoS defense of learning model-based in AIoT with programmable data plane, SDN

The availability of SD-AIoT is currently under complicated and serious cyber threats, especially Low-rate Denial-of-Service attacks. However, traditional defense schemes for such attacks with characteristics of high concealability and periodicity suffer from serious challenges with high detection difficulty, low accuracy of detection models, and inefficiency of mitigation approaches. In this paper, one novel cooperative defense scheme against hybrid LDoS attacks is proposed, which consists of a timely-response hardware-based Renyi Entropy edge checkpoint intent detection algorithm, the high-precision detection mechanism based on a hybrid deep learning model, and a Markov-chain-based differential rate-limiting mitigation strategy. The detection algorithm deployed at the edge checkpoint activates a hybrid CNN-RF-based deep learning model after filtering the intent information of the flows to detect which are malicious LDoS flows with high accuracy, where the multi-stage detection scheme not only extracts and learns the hidden features of the flow data, but also has better representation capabilities. Enhanced dynamic threshold-based whitelisting automatically adapts to the real-time state of the network environment to improve mitigation flexibility. Markov chain-based differential rate-limiting mitigation strategy reduces the packet loss error rate to mitigate network attacks promptly and ensures the continuation of network services. The results of several comparative experiments show that the proposed scheme detects LDoS attacks more accurately and mitigates them more effectively than traditional schemes.

Pink Slips (for Some): Campus Employment, Social Class, and COVID-19

Although undergraduates from all class backgrounds work while attending college, little is known about how students approach finding work and the benefits they reap from different on-campus roles. Drawing on interviews with 110 undergraduates at Harvard University, we show that in the absence of clear institutional expectations surrounding on-campus work opportunities, students draw on class-based strategies to determine which jobs are “right for them.” Upper-income students pursued “life of the mind” jobs that permitted them access to institutional resources and networks. Alternatively, lower-income students pursued more transactional “work for pay” positions that yielded fewer institutional benefits and connections. The consequences of these differential strategies were amplified during COVID-19 campus closures as work-for-pay positions were eliminated while life of the mind continued remotely. Through documenting heterogeneity in work experiences, we reveal a class-segregated labor market on campus and extend previous analyses of how university practices exacerbate class differences and reproduce inequality.

Optimization Design of PSS and SVC Coordination Controller Based on the Neighborhood Rough Set and Improved Whale Optimization Algorithm

Aimed at reducing the redundancy of parameters for the power system stabilizer (PSS) and static var compensator (SVC), this paper proposes a method for coordinated control and optimization based on the neighborhood rough set and improved whale optimization algorithm (NRS-IWOA). The neighborhood rough set (NRS) is first utilized to simplify the redundant parameters of the controller to improve efficiency. Then, the methods of the Sobol sequence initialization population, nonlinear convergence factor, adaptive weight strategy, and random differential mutation strategy are introduced to improve the traditional whale optimization algorithm (WOA) algorithm. Finally, the improved whale optimization algorithm (IWOA) is utilized to optimize the remaining controller parameters. The simulation results show that the optimization parameters were reduced from 12 and 18 to 3 and 4 in the single-machine infinity bus system and dual-machine power system, and the optimization time was reduced by 74.5% and 42.8%, respectively. In addition, the proposed NRS-IWOA method exhibits more significant advantages in optimizing parameters and improving stability than other algorithms.

A two-stage clonal selection algorithm for local feature selection on high-dimensional data

Various evolutionary computation algorithms have shown their excellent performance for high-dimensional feature selection (FS). However, most of current FS methods choose a global feature subset and ignore the correlations between feature subsets and sample subspaces, which limits the performance of methods in the sample space with different probability distributions. To solve the issue, we propose a two-stage clonal selection algorithm for filter-based local feature selection (TSCSA-LFS) that integrates symmetric uncertainty and a discrete clonal selection algorithm with three contributions. First, unlike conventional feature selection methods, TSCSA-LFS introduces local sample behaviors and assigns subsets of features for different sample regions. Second, an improved discrete clonal selection algorithm is developed for searching relevant features, which contains mutual information-based individual initialization, a differential evolution-based mutation strategy pool and a local search technique. Third, a two-part antibody representation is employed for automatical adjustment of the weight-related parameter. Our method shows the promising experimental results compared with well-known global FS and clonal selection-based local FS methods on fourteen high-dimensional datasets. For instance, our method can obviously outperform local FS, filter-based and hybrid methods on at least nine datasets

Spotting mistakes: Reappraisal of Spotted Drum Stellifer punctatissimus (Meek & Hildebrand, 1925) (Teleostei: Sciaenidae) reveals species misidentification trends and suggests latitudinal sexual dimorphism

A major part of the described species is understudied, falling into the Linnean shortfall. This is a major concern for cryptic species, which require integrative approaches to better evaluate their diversity. We conducted morphological analyses using specimens of Stellifer punctatissimus, S. gomezi, and S. menezesi to reassess their taxonomical identity. We evaluated the allometric and sexual components of the morphology of the Stellifer punctatissimus species complex, and tested and discussed species hypotheses. The combined evidence of our work and previous studies agrees with the current morphological hypothesis of three species, as opposed to the two-lineage molecular hypothesis. However, as cryptic species, they overlap in most their traits, especially females and juveniles. Previously unaccounted variation of allometric and sexually dimorphic characters in this species complex revealed a confounding effect that might explain past and current taxonomic errors. Taxonomical practice of using body depth as a diagnostic character had led to juveniles and females being, respectively, described as a different species or wrongfully identified. Hence, taxonomical studies demand better assessment of allometric and sexual dimorphism components. Herein, we present new characters in a key to the Atlantic species of Stellifer, which disclose size and sexual variation unnoticed in previous studies. The contrasting growth patterns among these species could imply distinct habitat use. As a result, it could be suggested that such species are under different threats, which highlights the need of differential management and conservation strategies.

Adaptive differential steering strategy for distributed driving unmanned ground vehicle with variable configurations based on modified localized modelling sliding mode control

When performing complex tasks such as position transfer and material transportation, the distributed driving unmanned platform with variable configurations needs to address the challenge of multi-wheel cooperative torque distribution control to achieve high-performance differential steering and enhance vehicle dynamics. The configuration change will impact the dynamic performance of the unmanned platform, posing a challenge to the performance of the existing control strategy based on mathematical model development. In order to address the aforementioned issues, this paper analyzes the impact of changes in vehicle configuration on steering gain and proposes a hierarchical adaptive differential steering strategy based on variable vehicle configurations. Firstly, the response characteristics of the yaw angle relative to the active yaw moment under the influence of changes in wheelbase and tread are analyzed. Based on this analysis, two structural modes, maneuverable and balanced, are selected. Secondly, a localized-modelling sliding mode control method with an extended state observer is proposed to estimate the desired yaw moment in the upper controller, considering the motor's execution delay. Then, the lower controller optimizes the torque of each wheel in real-time using the whale optimization algorithm. It aims to optimize tire energy dissipation and tire load rate while ensuring driving stability and achieving differential steering. Finally, through co-simulation and experiments on a scaled prototype, the reliability of the dynamics theory and the superiority of the control algorithm are validated. This optimization has led to significant improvements in the tire dissipation energy index and tire load rate index.

Monoclonal gammopathy-associated peripheral neuropathies: Uncovering pearls and challenges.

Monoclonal gammopathy-related peripheral neuropathies encompass a spectrum of clinical presentations in which the monoclonal protein directly damages the tissues, including the peripheral nervous system. Given the prevalence of both peripheral neuropathy and monoclonal gammopathy in the general population, these conditions may overlap in clinical practice, posing a challenge for clinicians in determining causality. Therefore, a comprehensive understanding of primary clinical syndromes and their neurophysiological patterns is of great importance for accurate differential diagnoses and effective treatment strategies. In this article, we examine the main forms of monoclonal gammopathies that affect the peripheral nerve. We explore the clinical and electrophysiological aspects and their correlation with each syndrome's corresponding monoclonal protein type. This knowledge is essential for healthcare professionals to diagnose better and manage patients presenting with monoclonal gammopathy-related peripheral nervous system involvement.

Differential Game Strategies for Social Networks With Self-Interested Individuals

Differential Game Strategies for Social Networks With Self-Interested Individuals

Differential releasing hydrogel loaded with oncolytic viruses and anti-CAFs drug to enhance oncology therapeutic efficacy

Interstitial hypertension and extracellular matrix (ECM) barriers imposed by cancer-associated fibroblasts (CAFs) at the tumor site significantly impede the retention of intratumorally administered oncolytic viruses (OVs) as well as their efficacy in infecting and eradicating tumor cells. Herein, a stable, controllable, and easily prepared hydrogel was developed for employing a differential release strategy to deliver OVs. The oncolytic herpes simplex virus-2 (oH2) particles were loaded within sodium alginate (ALG), together with the small molecule drug PT-100 targeting CAFs. The rapid release of PT-100 functions as an anti-CAFs agent, reducing ECM, and alleviating interstitial pressure at the tumor site. Consequently, the delayed release of oH2 could more effectively invade and eradicate tumor cells while also facilitating enhanced infiltration of immune cells into the tumor microenvironment, thereby establishing an immunologically favorable milieu against tumors. This approach holds significant potential for achieving highly efficient oncolytic virus therapy with minimal toxicity, particularly in tumors rich in stromal components.

Optimal Voltage Recovery Learning Control for Microgrids with N-Distributed Generations via Hybrid Iteration Algorithm

Considering that the nonlinearity and uncertainty of the microgrid model complicate the derivation and design of the optimal controller, an adaptive dynamic programming (ADP) algorithm is designed to solve the model-free non-zero-sum game. By combining the advantages of policy iteration and value iteration, an optimal learning control scheme based on hybrid iteration is constructed to provide stringent real power sharing for the nonlinear and coupled microgrid systems with N-distributed generations. First, using non-zero-sum differential game strategy, a novel distributed secondary voltage recovery consensus optimal control protocol is built using a hybrid iteration method to realize the voltage recovery of microgrids. Then, the data of the system state and input are gathered along a dynamic system trajectory and a data-driven optimal controller learns the game solution without microgrid physics information, enhancing convenience and efficiency in practical applications. Furthermore, the convergence analysis is given in detail, and it is proved that the control protocol can converge to the optimal solution so as to ensure the stability of the voltage recovery of the microgrid system. Convergence analysis proves the convergence of the the protocol to the optimal solution, ensuring voltage recovery stability. Simulation results validate the feasibility and effectiveness of the proposed scheme.

Neuroevolution reinforcement learning for multi-echelon inventory optimization with delivery options and uncertain discount

The advanced information technology has enabled supply chain to make centralized optimal decision, allowing to make a global optimal solution. However, dealing with uncertainty is important in inventory management. Besides demand and supply uncertainties, supplier discounts also often arise unexpectedly. Further, suppliers or third-parties typically offer various delivery options in which trade-off occurs between cost and lead time. Thus, this study introduces new problem namely Multi-Echelon Inventory Optimization with Delivery Options and Uncertain Discount (MEIO-DO-UD). As a solution, Neuroevolution Reinforcement Learning (NERL) framework is developed for minimizing total system cost. The environment is modeled via System Dynamics (SD) and actor is presented by integration of Artificial Neural Network and Evolutionary Algorithm (EA), creating an effective decision-making model under dynamic uncertainty. The experimental study has been conducted where two different supply chain networks are given namely serial and divergence. Three EA algorithms are compared namely Differential Evolution (DE), Memetic Algorithm (MA), and Evolution Strategy (ES). Furthermore, NERL is also compared with the EA-optimized classical continuous review model namely (s,Q). The result shows that regardless what EA type is used, the proposed NERL always outperforms EA-optimized (s,Q) model. The more complex the problem, the further improvement can be made i.e. cost reduction up to 58%, followed by the fill rate improvement. The result also shows that NERL can avoid overfitting. Managerial implications are highlighted where NERL provides the more stable inventory level among all supply chain partners and bull-whip effect can be damped.