Dynamic Adaptive Strategy Research Articles

A stage-related online incremental transfer learning-based remaining useful life prediction method of bearings

Remaining useful life (RUL) prediction of bearings holds significant research value in bearing health management, particularly in the rapidly advancing field of intelligent manufacturing. How to realize the accurate online RUL prediction still faces many challenges, including the diverse degradation trends in bearings, the scarcity of labeled data, and the inconsistencies in the data marginal distributions. This paper proposes a Stage-related Online Incremental Adversarial Domain Adaption (SR-OIADA) transfer learning algorithm for bearing RUL prediction. It employs a dynamic domain adaptation strategy, which integrates incremental learning and transfer learning, facilitating the online acquisition of degradation knowledge aligned with degradation stages. A novel online degradation stage division algorithm is proposed to adaptively detect the health status of online monitored bearings, thereby enhancing the effectiveness of online learning. By incrementally training the deep models in a prediction-to-transfer manner, the model undergoes incremental updates at checkpoints, taking full advantage of new online samples. Experiments are conducted on real rolling bearing datasets to validate the proposed methods, demonstrating excellent predictive performance and highlighting the superiority compare with the non-transfer and offline methods.

Deep Learning-Based ADL Assessment and Personalized Care Planning Optimization in Adult Day Health Centers

This paper presents an in-depth learning process for Activities of Daily Living (ADL) and self-care planning in Adult Day Health Centers. The framework integrates multi-modal sensor data fusion with deep learning architectures to provide continuous monitoring and automated evaluation of older people's legal status. The system uses a hierarchical system combined with convolutional neural networks (CNNs) and long short-term memory (LSTM) networks, enhanced by monitoring systems for physical-spatial learning. Many data streams from wearable devices, environmental sensors, and medical monitoring devices are becoming pre-processed and de-processed. The framework incorporates a dynamic care plan adaptation strategy utilizing reinforcement learning techniques for intervention optimization. Experimental validation conducted across three Adult Day Health Centers with 150 participants over six months demonstrated superior performance compared to traditional assessment methods. The system achieved 92.8% accuracy in ADL recognition tasks, with a 35% reduction in assessment time and a 62% decrease in false alarm rates. Clinical validation through 25 detailed case studies revealed early detection of health deterioration, averaging 3.2 days ahead of conventional methods. The proposed framework significantly enhances the efficiency and accuracy of elderly care delivery while reducing healthcare provider workload by 40%. This research contributes to advancing intelligent healthcare by creating a solution for ADL measurement and self-correction.

Adaptive Dual-Layer Resource Allocation for Maximizing Spectral Efficiency in 5G Using Hybrid NOMA-RSMA Techniques

The unprecedented growth in data demands for 5G communication systems necessitates advanced techniques to maximize spectral efficiency while ensuring user fairness and low latency. This study proposes Adaptive Dual-Layer Resource Allocation (ADLRA), a novel hybrid technique combining Non-Orthogonal Multiple Access (NOMA) and Rate-Splitting Multiple Access (RSMA). The ADLRA framework introduces dynamic user pairing, hierarchical beamforming, and adaptive power and rate allocation strategies to optimize resource utilization. Key features include dynamic user pairing, leveraging machine learning algorithms for efficient group formation based on channel conditions, and hierarchical beamforming, which prioritizes high-priority users in the NOMA layer while effectively managing shared resources in the RSMA layer. Interference mitigation is achieved through spatial filtering and multi-user diversity techniques, ensuring minimal intra-cell and inter-cell interference. Simulation results demonstrate significant performance gains Spectral efficiency improved by 32%, compared to traditional NOMA. Latency reduced by 18%, ensuring seamless communication for ultra-reliable low-latency applications. Achieved a 94% fairness index, reflecting equitable resource allocation among users. Enhanced throughput, with an average gain of 28%, compared to RSMA-only systems. These results highlight the potential of ADLRA to meet the stringent requirements of next-generation 5G systems, offering a scalable and efficient solution for diverse communication scenarios. The proposed method sets a foundation for future hybrid access strategies in wireless communication networks.

A robust adaptive error state Kalman filter for MEMS IMU attitude estimation under dynamic acceleration

Accurate estimation of attitude (pitch, roll) is a prerequisite for ensuring safe navigation during vehicle control execution. In dynamic environments, the presence of acceleration often degrades the accuracy and robustness of attitude estimation using traditional algorithms with Micro-Electro-Mechanical Systems (MEMS) Inertial Measurement Unit (IMU). This paper proposes a robust adaptive error state Kalman filter (RAESKF) algorithm for attitude estimation of MEMS IMU under dynamic acceleration conditions. The RAESKF algorithm decomposes the true attitude Direction Cosine Matrix (DCM) into nominal and error components. An error state Kalman filter is utilized to fuse real-time measurements from MEMS gyroscope and accelerometer, estimating the error component, which is then used to correct the nominal attitude DCM. Furthermore, within the error state Kalman filtering framework, a dynamic acceleration robust adaptive adjustment strategy is implemented. This strategy relies on real-time data monitoring, threshold-based decision-making, and switching mechanisms. It adaptively selects between dynamic acceleration model compensation and online adjustment of the measurement noise covariance matrix. The primary objective of this strategy is to mitigate the impact of disturbance induced by dynamic acceleration on attitude estimation, thereby enhancing the accuracy and robustness. Laboratory rotation experiments have demonstrated that the algorithm improved pitch and roll measurement accuracy by at least 8.81% and 11.69%, respectively, under rotational conditions. In dynamic acceleration conditions, pitch and roll measurement accuracy improved by at least 42.76% and 67.21%, respectively.

Integrating reinforcement learning and supervisory control theory for optimal directed control of discrete-event systems

Directed control is crucial for implementing controllers on programmable logic controllers. A directed controller is one that actively selects at most one controllable event (control command) to execute at any instant. This study investigates the numerical optimization problem of directed control for discrete-event systems, such as traffic systems and robotic systems. By integrating supervisory control theory and reinforcement learning, an algorithmic procedure is designed to synthesize an optimal directed controller, which minimizes the total cost associated with event execution while ensuring the safety and liveness of the controlled system. Two improved Q-learning algorithms incorporating dynamic parameter adaptation strategies are developed to enhance the global search ability of basic Q-learning. To demonstrate applicability and effectiveness, we apply the proposed method to control a guideway system and a multi-train traffic system, respectively. The experimental results indicate the superiority of the proposed method over the comparison methods.

IMOABC: An efficient multi-objective filter–wrapper hybrid approach for high-dimensional feature selection

With the development of data science, the challenge of high-dimensional data has become increasingly prevalent. High-dimensional data contains a significant amount of redundant information, which can adversely affect the performance and effectiveness of machine learning algorithms. Therefore, it is necessary to select the most relevant features from the raw data and perform feature selection on high-dimensional data. In this paper, a novel filter–wrapper feature selection method based on an improved multi-objective artificial bee colony algorithm (IMOABC) is proposed to address the feature selection problem in high-dimensional data. This method simultaneously considers three objectives: feature error rate, feature subset ratio, and distance, effectively improving the efficiency of obtaining the optimal feature subset on high-dimensional data. Additionally, a novel Fisher Score-based initialization strategy is introduced, significantly enhancing the quality of solutions. Furthermore, a new dynamic adaptive strategy is designed, effectively improving the algorithm’s convergence speed and enhancing its global search capability. Comparative experimental results on microarray cancer datasets demonstrate that the proposed method significantly improves classification accuracy and effectively reduces the size of the feature subset when compared to various traditional and state-of-the-art multi-objective feature selection algorithms. IMOABC improves the classification accuracy by 2.27% on average compared to various multi-objective feature selection methods, while reducing the number of selected features by 88.76% on average. Meanwhile, IMOABC shows an average improvement of 4.24% in classification accuracy across all datasets, with an average reduction of 76.73% in the number of selected features compared to various traditional methods.

Improved ACO algorithm fused with improved Q-Learning algorithm for Bessel curve global path planning of search and rescue robots

Addressing issues with traditional ant colony and reinforcement learning algorithms, such as low search efficiency and the tendency to produce insufficiently smooth paths that easily fall into local optima, this paper designs an improved ant colony optimization algorithm fusion with improved Q-Learning (IAC-IQL) algorithm for Bessel curve global path planning of search and rescue (SAR) robots. First, the heuristic function model in the ant colony algorithm is improved, the elite ant search strategy and the adaptive pheromone volatility factor strategy are introduced, and the initial path is searched in realize the motion environment with the help of the improved ant colony algorithm, and the initialized pheromone matrix is constructed. Second, the improved ant colony algorithm and Q-Learning (QL) algorithm are fused by utilizing the similarity between the pheromone matrix in the improved ant colony algorithm and the Q-matrix in the QL algorithm. A heuristic learning evaluation model is designed to dynamically adjust the learning factor and provide guidance for the search path. Additionally, a dynamic adaptive greedy strategy is introduced to balance the exploration and exploitation of the robot in the environment. Finally, the paths are smoothed using third-order Bessel curves to eliminate the problem of excessive steering angles. Through three sets of comparative simulation experiments conducted in Pycharm platform, the effectiveness, superiority, and practicality of the IAC-IQL algorithm were verified. The experimental results demonstrated that the IAC-IQL algorithm integrates the strong search capability of ant colony algorithm and the self-learning characteristics of QL algorithm. SAR robots equipped with the improved IAC-IQL algorithm exhibit significantly enhanced iterative search efficiency in grid simulation environment and image sampling simulation environment. The global path optimization indicators demonstrate high efficiency, and the paths are smoother.

A Power Optimization Approach for Large-scale RM-TB Dual Logic Circuits Based on an Adaptive Multi-Task Intelligent Algorithm

Logic synthesis is a crucial step in integrated circuit design, and power optimization is an indispensable part of this process. However, power optimization for large-scale Mixed Polarity Reed-Muller (MPRM) logic circuits is an NP-hard problem. In this article, we divide Boolean circuits into small-scale circuits based on the idea of divide and conquer using the proposed Dynamic Adaptive Grouping Strategy (DAGS) and the proposed circuit decomposition model (CDM). Each small-scale Boolean circuit is transformed into an MPRM logic circuit by a polarity transformation algorithm. Based on the gate-level integration, we integrate small-scale circuits into an MPRM and Boolean Dual Logic (RBDL) circuit. Furthermore, the power optimization problem of RBDL circuits is a multi-task, multi-extremal, high-dimensional combinatorial optimization problem, for which we propose an Adaptive Multi-task Intelligent Algorithm (AMIA), which includes global task optimization, population reproduction, valuable knowledge transfer (VKT), and local exploration to search for the lowest power for RBDL circuits. Moreover, based on the proposed Fast Power Decomposition Algorithm (FPDA), we proposed a Power Optimization Approach (POA) for an RBDL circuit with the lowest power using the AMIA. Experimental results based on Microelectronics Center of North Carolina (MCNC) Benchmark test circuits demonstrate the effectiveness and superiority of the POA compared to state-of-the-art POAes.

Optimal Rule-Interposing Reinforcement Learning-Based Energy Management of Series—Parallel-Connected Hybrid Electric Vehicles

P2–P3 series–parallel hybrid electric vehicles exhibit complex configurations with multiple power sources and operational modes, presenting a difficulty in developing efficient energy management strategies. This paper takes a P2–P3 series–parallel hybrid power system-KunTye 2DHT system as the research object and proposes a deep reinforcement learning framework based on pre-optimized energy management to improve the energy consumption performance of the hybrid electric vehicles. Firstly, a control-oriented model is established based on its system configuration and characteristics. Then, the optimal distribution of the motor energy under different operating modes is pre-optimized, which aims to reduce the energy management task’s dimensionality by equating two motors as an equivalent motor. Subsequently, based on real-time traffic information under connected conditions, deep reinforcement learning is utilized to optimize the optimal operating modes of the hybrid system and the optimal distribution between the engine and equivalent motors. Combining the pre-optimized results, the optimal energy distribution between the engine and the two motors in the system is achieved. Finally, performance comparisons are made between the predictive control and the traditional Dynamic Programming and Adaptive Equivalent Consumption Minimization Strategy, revealing the proposed optimization algorithm’s promising potential in reducing fuel consumption.

Optimization of node deployment in underwater internet of things using novel adaptive long short‐term memory‐based egret swarm optimization algorithm

SummaryOptimizing node deployment in the underwater Internet of Things (UIoT) poses significant challenges due to the complex and dynamic nature of underwater environments. This research introduces the adaptive long short‐term memory‐based egret swarm optimization algorithm (ALSTM‐ESOA), a novel approach designed to enhance network coverage and performance efficiently. Unlike traditional methods, ALSTM‐ESOA incorporates cognitive learning capabilities from long short‐term memory (LSTM) and dynamic adaptation strategies inspired by the hunting behaviors of egrets. The algorithm's effectiveness was tested through extensive simulations in MATLAB, demonstrating notable improvements over existing models: network throughput increased by up to 55.56%, deployment time decreased by 88.89%, and energy efficiency improved significantly. These enhancements are critical for robust, real‐time data collection and monitoring in underwater settings, providing substantial benefits for marine research and resource management. The findings suggest that ALSTM‐ESOA significantly outperforms conventional algorithms, offering a promising new tool for the advancement of UIoT applications. After being implemented in MATLAB, the suggested ALSTM‐ESOA model for the node deployment optimization in UIoT is examined. The proposed ALSTM‐ESOA in terms of network throughput is 55.56%, 38.89%, 36.11%, and 11.11% better than CNN, LSTM, ARO‐RTP, and IGOR‐TSA, respectively. Similarly, the proposed ALSTM‐ESOA with respect to deployment time is 88.89%, 81.82%, 75%, and 50% better than CNN, LSTM, ARO‐RTP, and IGOR‐TSA, respectively. For the purpose of exploring marine resources, monitoring underwater environments, and conducting marine scientific investigation, the research's findings are extremely valuable.

Assessment and strategies for water supply security risks in the estuarine city Shanghai under normalized extreme climate conditions

Facing the escalation of extreme climate events, estuarine delta cities like Shanghai grapple with significant water supply challenges. This study employs system dynamics and dynamic adaptive policy pathways to assess Shanghai's water supply security risks and responsive strategies in 2022–2050, amidst extreme climate conditions. Utilizing data from 2000 to 2021, we constructed a system dynamics model to predict future water demand under various development modes. Focusing on the unusual 2022 drought in the Yangtze River Basin, we simulated 15 scenarios, including economy, population, water efficiency, and reservoir levels, to identify the extent and timing of potential water supply risks, then proposed pertinent dynamic adaptive strategies to address them. Our findings suggest that the 2022 drought significantly reduced Shanghai's water supply capacity, leading to a notable deficit. Under scenarios of accelerated economic growth, water supply security risks are heightened, with projections indicating a reduction of days of supply available to merely 33–67 days, and escalating water shortage amount to 592–896 million m3 by 2050. Short-to-medium-term recommendations include optimizing both local and transit water resources, strengthening emergency water reserves, enhancing water use efficiency, and maintaining stable reservoir water levels. For the long term, expanding water storage infrastructure and promoting integrated water resource management within the Yangtze River Delta is key to establishing a resilient and diversified water supply system, effectively mitigating future water security risks. This study provides a scientific basis and reference for the sustainable management of water resources in estuarine cities confronting normalized extreme climate conditions. It offers valuable insights for policymakers and actionable suggestions for urban planners.

Novel heuristic algorithm incorporating dynamic penalty and adaptive evolution strategy for heat integration network design

Designing efficient heat-exchanger networks is essential for the effective use of energy in the process industries, fostering sustainable and low-carbon development. Traditional heuristic algorithms often fail to balance global and local search processes adequately when solving heat exchanger network synthesis. This frequently leads to early structural fixation and limits the exploration of integer variable search spaces. This study presents an innovative structural enhancement approach, incorporating a dynamic penalty for utility operating costs and an adaptive evolution strategy. The developed method increases the number of heat exchangers by dynamically penalizing utility operating costs and allows evolutionary parameters to be adaptively adjusted based on historical solutions. This approach diminishes the dependence on manual parameter tuning while enhancing flow matching diversity. Moreover, novel metrics for assessing structural diversity, based on node-based nonstructural model with stream splitting are established to facilitate the transition between various optimization paths, thereby improving the algorithm's robustness throughout different stages of the optimization process. Finally, computational analysis on five industrial-scale cases yields more optimal solutions with varied structural configurations compared to previous studies, demonstrating the method's feasibility and effectiveness.

Intergenerationally tied relocation and care circulation: Motivations, struggles, and future arrangements

AbstractObjectiveThis study investigates the processes and experiences of intergenerationally tied relocation in which older parents from mainland China join their already migrated children in Hong Kong to provide domestic and childcare support.BackgroundStudies on tied migration have mainly focused on couple‐based moves within a human capital framework. With the increasing trend of grandparenting migrants, studies have expanded to address intergenerationally tied relocation.MethodWe conducted thematic analysis of in‐depth interview data of 34 mainland Chinese respondents, including 15 parent–child pairs and four adult children.ResultsThe decision to relocate is influenced by the established history of intergenerational supportive exchange relationships. The initial motivation for tied relocation is to benefit the adult children. Older parents' postrelocation experiences can affect their later decision to leave or remain, with key factors including financial dependence, emotional stress, and difficulty in balancing the duty of care between migrant children and distant family members. In planning for future eldercare, the goal is to meet the needs of both generations. Options include non‐coresidential care, reliance on the parents' hometown sibling network, or living in nearby mainland cities closer to Hong Kong.ConclusionThe study highlights the complex interplay of family relationships and responsibilities in the phenomenon of intergenerationally tied relocation. The findings unveil the dynamic roles and adaptive strategies families utilize to facilitate the transfer of care and support across generations and geographies.ImplicationsPolicy and services should be developed to assist aging immigrant families in their adaptation and integration, and formulate strategies for meeting future care needs.

Dynamic Difficulty Adaptation Based on Stress Detection for a Virtual Reality Video Game: A Pilot Study

Virtual reality (VR) is continuing to grow as more affordable technological devices become available. Video games are one of the most profitable applications, while rehabilitation has the most significant social impact. Both applications require a proper user evaluation to provide personalized experiences that avoid boring or stressful situations. Despite the successful applications, there are several opportunities to improve the field of human–machine interactions, one of the most popular ones being the use of affect detection to create personalized experiences. In that sense, this study presents the implementation of two dynamic difficulty adaptation strategies. The person’s affective state is estimated through a machine learning classification model, which later serves to adapt the difficulty of the video game online. The results show that it is possible to maintain the user at a given difficulty level, which is analogous to achieving the well-known flow state. Among the two implemented strategies, no statistical differences were found in the workload induced by the users. However, more physical demands and a higher frustration were induced by one of the strategies, validated with the recorded muscular activity. The results obtained contribute to the state of the art of DDA strategies in virtual reality driven by affective data.

3WAUS: A novel three-way adaptive uncertainty-suppressing model for facial expression recognition

Facial Expression Recognition (FER) remains a challenging task due to the uncertainty caused by the noisy labels, interrater disagreement, and the inter-class similarity of facial expressions. However, the majority of existing FER methods are deterministic, resulting in limited generalization performance in dealing with uncertainty. By adding a non-deterministic option, three-way decision (3WD) enhances the ability to handle uncertainty, which is widely used in different domains to reduce decision risk. Therefore, we introduce 3WD model into FER, and propose a novel three-way adaptive uncertainty-suppressing model (3WAUS). The model divides the sample space into three regions: high-importance, medium-importance, and low-importance, with different action strategies to effectively suppress the uncertainty in FER. Specifically, 3WAUS utilizes a weighted learning strategy to reinforce learning in the high-importance (low-uncertainty) region, thus acquiring high-quality sample features. For medium-importance samples, a dynamic adaptive re-labeling strategy is proposed to modify noisy labels, enabling the model to learn more accurate representations and avoid overfitting noise. Additionally, 3WAUS also attenuates the learning for low-importance (high-uncertainty) samples, effectively reducing the misguidance of noise on the model. Finally, extensive experiments on three public datasets and synthetic noisy datasets indicate that the proposed 3WAUS outperforms state-of-the-art methods in dealing with uncertainty.

DATFNets-dynamic adaptive assigned transformer network for fire detection

Fires cause severe damage to the ecological environment and threaten human life and property. Although the traditional convolutional neural network method effectively detects large-area fires, it cannot capture small fires in complex areas through a limited receptive field. At the same time, fires can change at any time due to the influence of wind direction, which challenges fire prevention and control personnel. To solve these problems, a novel dynamic adaptive distribution transformer detection framework is proposed to help firefighters and researchers develop optimal fire management strategies. On the one hand, this framework embeds a context aggregation layer with a masking strategy in the feature extractor to improve the representation of low-level and salient features. The masking strategy can reduce irrelevant information and improve network generalization. On the other hand, designed a dynamic adaptive direction conversion function and sample allocation strategy to fully use adaptive point representation while achieving accurate positioning and classification of fires and screening out representative fire samples in complex backgrounds. In addition, to prevent the network from being limited to the local optimum and discrete points in the sample from causing severe interference to the overall performance, designed a weighted loss function with spatial constraints to optimize the network and penalize the discrete points in the sample. The mAP in the three baseline data sets of FireDets, WildFurgFires, and FireAndSmokes are 0.871, 0.909, and 0.955, respectively. The experimental results are significantly better than other detection methods, which proves that the proposed method has good robustness and detection performance.

DDAGCN: an unsupervised cross-domain identification method for tie rod bolt loosening in a rod-fastening rotor system under different working conditions

The cross-domain diagnosis of tie rod bolt loosening is essential for guaranteeing the healthy operation of rod-fastening rotor (RFR) systems. The unsupervised domain adaptation (UDA) method effectively alleviates the impact of domain discrepancy and has been applied for cross-domain diagnosis. Traditional UDA methods mainly focus on the marginal and conditional distributions with fixed weights to adapt the domain distribution discrepancy. However, the fixed distribution combination cannot satisfy the requirement of feature domain alignment under different working conditions, and the relative importance of the two distributions cannot be evaluated quantitatively. This paper proposes an improved dynamic distribution adaptive graph convolutional network (DDAGCN) for the cross-domain diagnosis of tie rod bolt loosening under different working conditions. This method can quantitatively evaluate the relative significance of each distribution in representing the distribution discrepancy. First, it combines the convolutional neural network and the graph convolutional network to extract the features in the graph structure by using the connection relationship between nodes, and realizes the full extraction of neighbourhood information of nodes. Then, the dynamic distribution adaptive alignment strategy is introduced to construct the dynamic linear combination of marginal and conditional distributions, so as to measure the distribution discrepancy between domains. Meanwhile, the domain adversarial module is combined to further reduce the domain gap and finally realize feature alignment. The extracted domain invariant features can effectively enhance the generalization ability and fault identification ability of the model. The case of the public bearing dataset verifies that the effectiveness and generalization ability of the proposed method for cross-domain fault diagnosis under different working conditions is superior to other compared methods. In addition, the identification ability of the proposed method for the degree of tie rod bolt loosening is verified by the self-made bolt loosening dataset of the RFR system.

Speech adaptation in conversation: Effects of segment cue-weighting strategy for non-native speakers

Research has investigated how speakers make phonetic adaptations by adjusting their speech sounds to either converge to or diverge from their interlocutor’s. Less is known about the dynamic adaptive strategies non-native speakers use to improve intelligibility when interacting with native speakers, particularly how their strategies change over the course of spontaneous conversations. The current study examines phonetic adaptations in English words contrasting tense and lax vowels (e.g., sheep-ship) in unscripted conversations between non-native Japanese English and native English speakers during an engaging computer game task. Japanese speakers have previously been found to rely on temporal cues (vowel length) for tensity distinctions due to their L1 cue-weighting strategy rather than spectral cues (vowel quality) that are predominantly used in English. Acoustic analyses are conducted to examine changes in non-native vowel productions before, during and after the conversation task. We predict that, in an attempt to overcome miscommunication, non-native speakers may initially lengthen the tense vowels to distinguish them from their lax counterparts. As the conversation progresses, non-native speakers may adapt to a more native-like cue-weighting pattern by shifting to spectral distinctions. Results are discussed in terms of adaptations to cue-weighting strategies for intelligibility gains by interlocutors of different linguistic backgrounds.

Optimization of Ampacity in High-Voltage Underground Cables with Thermal Backfill Using Dynamic PSO and Adaptive Strategies

This article addresses challenges in the design of underground high-voltage transmission lines, focusing on thermal management and cable ampacity determination. It introduces an innovative proposal that adjusts the dimensions of the backfill to enhance ampacity, contrasting with the conventional approach of increasing the core cable’s cross-sectional area. The methodology employs a particle swarm optimization (PSO) technique with adaptive penalization and restart strategies, implemented in MATLAB for parameter autoadaptation. The article emphasizes more efficient solutions than traditional PSO, showcasing improved convergence and precise results (success probability of 66.1%). While traditional PSO is 81% faster, the proposed PSO stands out for its accuracy. The inclusion of thermal backfill results in an 18.45% increase in cable ampacity, considering variations in soil thermal resistivity, backfill properties, and ambient temperature. Additionally, a sensitivity analysis was conducted, revealing conservative values that support the proposal’s robustness. This approach emerges as a crucial tool for underground installation, contributing to continuous ampacity improvement and highlighting its impact on decision making in energy systems.

An optimized sparse deep belief network with momentum factor for fault diagnosis of radar transceivers

Transceiver is a crucial component of radar system that allows for the regulation of signal phase and amplitude as well as the amplification of both transmitted and received signals. Its operational efficiency has a significant impact on the whole dependability of the radar system. To ensure the safe and reliable operation of the radar system, an optimized sparse deep belief network with momentum factor is developed to diagnose potential faults of radar transceivers. Firstly, a momentum term is added into the parameter update to enhance the anti-oscillation ability of model parameters in training, while a sparse regular term is integrated into the deep belief network to prevent the model from overfitting. Secondly, to automatically configure the model hyper-parameters, a hybrid sine cosine algorithm (HSCA) with dynamic inertia weight and adaptive strategies is proposed. Thus, an effective diagnostic model named HSCA-MS-DBN is formed by combining sparse deep belief network with momentum factor and HSCA. The efficiency of the proposed HSCA-MS-DBN model is confirmed using an actual-world radar transceiver dataset, and the findings from experiments reveal that this model surpasses multiple prominent intelligent models.