Large language model-assisted Evolutionary Search (LES) has emerged as a promising approach for Automated Algorithm Discovery (AAD). While many evolutionary search strategies have been developed for classic optimization problems, LES operates in abstract language spaces, presenting unique challenges for applying these strategies effectively. To address this, we propose a general LES framework that incorporates feature-assisted niche construction within abstract search spaces, enabling the seamless integration of niche-based search strategies from evolutionary computation. Building on this framework, we introduce PartEvo, an LES method that combines niche collaborative search and advanced prompting strategies to improve algorithm discovery efficiency. Experiments on both synthetic and real-world optimization problems show that PartEvo outperforms human-designed baselines and surpasses prior LES methods, such as Eoh and Funsearch. In particular, on resource scheduling tasks, PartEvo generates meta-heuristics with low design costs, achieving up to 90.1\% performance improvement over widely-used baseline algorithms, highlighting its potential for real-world applications.

ABSTRACT As a core industrial equipment, the stable operation of centrifugal compressors is crucial to production. Current research on its fault diagnosis mostly focuses on structured monitoring data, with insufficient mining of unstructured operation and maintenance experience data. To address this, this paper constructs an intelligent diagnosis model integrating ontology knowledge reasoning, knowledge graph modeling, and Bayesian Network (BN), realizing cross‐modal fault accurate localization and root cause analysis through the deep integration of “knowledge + probability.” Firstly, an ontology knowledge model is established based on fault information mined from unstructured data (such as fault reports and maintenance records), enabling standardized expression and semantic association of fault knowledge. The model is then imported into the Neo4j database, and specific fault information files are exported through Python queries to serve as the basic data for BN reasoning. Next, a probabilistic reasoning model between fault components and symptoms is built based on BN. Combining expert experience and historical data, the node conditional probabilities are determined to describe the uncertainty of fault propagation. Finally, a fusion method of ontology and BN is designed: ontology reasoning is used to optimize the BN structure, and intelligent diagnostic reasoning is realized through dynamic updating of posterior probabilities. Experiments using fault reports of centrifugal compressors from a certain enterprise show that the proposed fusion model can improve the interpretability and dynamic reasoning ability of fault diagnosis. Case verification demonstrates that the fault recognition accuracy of this method reaches 85%, indicating good performance. Therefore, this research provides a feasible solution for utilizing unstructured operation and maintenance data, enhancing the practicality of intelligent diagnosis for complex industrial equipment, which can shorten fault downtime, reduce maintenance costs, and thus has practical application value.

A collaborative system that includes mobile devices (MDs), edge nodes (ENs), and the cloud is needed where ENs at the network edge can run offloaded tasks of MDs with limited resources and energy for timely processing for latency-sensitive applications. Unlike existing studies, we formulate a total cost minimization problem for the system for applications, which can be divided into several interdependent subtasks. Each subtask can be executed in MDs, ENs, and the cloud. This work formulates a mixed-integer nonlinear program to minimize the total system cost. To address it, a novel meta-heuristic optimization algorithm calledGeneticSimulated-annealing-basedParticle swarm optimization withAuto-Encoder (GSPAE) is proposed, which innovatively combines feature extraction of deep learning and global search of meta-heuristic optimization. Genetic operations provide diverse solutions, the Metropolis acceptance of annealing offers a robust global search, and autoencoders (AEs) extract distribution characteristics of particles toward high-quality regions for fast convergence. Thus, GSPAE optimizes the associations between ENs and MDs and the scheduling of subtasks among MDs, ENs, and the cloud. Experiments with large-scale Google cluster datasets show that compared to state-of-the-art benchmark methods, GSPAE reduces the total cost by at least 17% while strictly meeting limits of application latency, available energy, computing, and communication resources of ENs and MDs.

In recent years, cloud computing has witnessed widespread applications across numerous organizations. Predicting workload and computing resource data can facilitate proactive service operation management, leading to substantial improvements in quality of service and cost efficiency. However, these data often exhibit non-linearity, high volatility, and interdependencies across different categories, presenting challenges for accurate forecasting. Consequently, there is a critical need to develop a method that thoroughly and comprehensively analyzes all available data to forecast future trends effectively. This work proposes a novel integrated data-enhanced prediction model named SVAPI for achieving high-accuracy workload prediction in cloud computing systems. SVAPI employs the Savitzky-Golay filter, Variational mode decomposition, and the mode selection based on Amplitude-aware Permutation entropy for feature processing, whose features are subsequently utilized by Informer for multivariate joint analysis of the enhanced data, achieving high-precision prediction. Ablation and comparative experiments with advanced prediction models are conducted on the Google cluster trace and other typical datasets. Realistic data-driven results indicate that SVAPI improves the prediction accuracy by 37.7% compared to the original Informer, with each module contributing to the performance enhancement. Furthermore, compared with Autoformer, SVAPI enhances the prediction accuracy of workload, CPU, and memory by 65.6%, 66.9%, and 70.8%, respectively, demonstrating that SVAPI owns strong abilities in noise filtering, feature processing, and multivariate joint analysis for achieving higher prediction accuracy.

In recent years, distributed computing has wit-nessed widespread applications across numerous organizations. Predicting workload and computing resource data can facilitate proactive service operation management, leading to substantial improvements in quality of service and cost efficiency. However, these data often exhibit non-linearity, high volatility, and inter-dependencies across different categories, presenting challenges for accurate forecasting. Consequently, there is a critical need to develop a method that thoroughly and comprehensively analyzes all available data to forecast future trends effectively. This work proposes a novel integrated data-enhanced prediction model named SVI for achieving high-accuracy workload prediction in distributed computing systems. SVI employs the Savitzky-Golay filter and variational mode decomposition for feature processing, whose features are subsequently utilized by Informer for multivariate joint analysis of the enhanced data, achieving high-precision prediction. Ablation and comparative experiments with advanced prediction models are conducted on the Google cluster trace and other typical datasets. Realistic data-driven results indicate that SVI improves the prediction accuracy by 35.4% compared to the original Informer, with each module contributing to the performance enhancement. Furthermore, compared with Autoformer, SVI enhances the prediction accuracy of workload, CPU, and memory by 62.5%, 65.6%, and 69.1 %, respectively.

Current Industrial Internet supports the sharing of information on heterogeneous resources and elements in a process of industrial production. It enables intelligent production processes and supports cost-effective scheduling. However, collaborative manufacturing and scheduling planning for enterprises with multiple plants cause several major challenges because of a large number of decision variables and constraints of manufacturing abilities of plants, resources of production, etc. Existing methods cannot comprehensively optimize the cost of multiple products in different plants, and fail to consider machine-level optimization of tasks of manufacturing. We propose a comprehensive machine-level architecture for enterprises with multiple plants. Based on this architecture, we formulate a limited non-linear integer optimization problem to decrease the total cost of transportation, production, and sales. In it, several real-life complicated nonlinear constraints are jointly considered, and they include constraints of storage space, replacement times, pairing production, substitution, and order fulfillment rates. To solve this optimization problem, we design a hybrid meta-heuristic optimization algorithm named Genetic Simulated annealing-based Particle Swarm Optimizer with Auto-Encoders (GSPAE). Extensive experiments with real-life data show that GSPAE decreases the total cost by 25% than other state-of-the-art methods.

Cloud-edge hybrid systems are known to support delay-sensitive applications of contemporary industrial Internet of Things (IoT). While edge nodes (ENs) provide IoT users with real-time computing/network services in a pay-as-you-go manner, their resources incur cost. Thus, their profit maximization remains a core objective. With the rapid development of 5G network technologies, an enormous number of mobile devices (MDs) have been connected to ENs. As a result, how to maximize the profit of ENs has become increasingly more challenging since it involves massive heterogeneous decision variables about task allocation among MDs, ENs, and a cloud data center (CDC), as well as associations of MDs to proper ENs dynamically. To tackle such a challenge, this work adopts a divide-and-conquer strategy that models applications as multiple subtasks, each of which can be independently completed in MDs, ENs, and a CDC. A joint optimization problem is formulated on task offloading, task partitioning, and associations of users to ENs to maximize the profit of ENs. To solve this high-dimensional mixed-integer nonlinear program, a novel deep-learning algorithm is developed and named as a Genetic Simulated-annealing-based Particle-swarm-optimizer with Stacked Autoencoders (GSPSA). Real-life data-based experimental results demonstrate that GSPSA offers higher profit of ENs while strictly meeting latency needs of user tasks than state-of-the-art algorithms.

Cloud-edge hybrid systems can support delay-sensitive applications of industrial Internet of Things. Edge nodes (ENs) as service providers, provide users computing/network services in a pay-as-you-go manner, and they also suffer from the high cost brought by providing computing resources. Thus, the problem of profit maximization is highly important to ENs. However, with the development of 5G network technologies, a large number of mobile devices (MDs) are connected to ENs, making the above-mentioned problem a high-dimensional challenge, which is highly difficult to solve. This work formulates a joint optimization problem of task offloading, task partitioning, and associations of large-scale users to ENs to maximize the profit of ENs. This work focuses on applications that can be split into multiple subtasks, each of which can be completed in MDs, ENs and a cloud data center. Specifically, a mixed integer nonlinear program is formulated to maximize ENs’ profit. Then, a novel hybrid algorithm named Genetic Simulated-annealing-based Particle swarm optimizer with a Stacked Autoencoder (GSPSA) is designed to solve it. Real-life data-based experimental results demonstrate that compared with other peer algorithms, GSPSA increases the profit of ENs while strictly meeting latency needs of users’ tasks. The dimension of the problem that can be solved is increased by more than 50% with GSPSA.

As a new paradigm, industrial Internet provides information sharing of various elements and resources in a whole industrial production process. It makes industrial production processes intelligent and provides low-cost and efficient scheduling. Manufacturing planning for multi-plant enterprises in industrial Internet brings many big challenges due to numerous optimization variables and limits of manufacturing capacities of plants, production resources, etc. Current studies fail to jointly consider the cost of different products in multiple heterogeneous plants, and ignore machine-level scheduling of manufacturing tasks. This work designs an improved framework for multi-plant enterprises, based on which a constrained non-linear integer program for reducing the total cost including production cost and transportation one is formulated. It jointly considers many complex nonlinear constraints, e.g., limits of replacement times, storage space, substitution and pairing production. It investigates machine-level task scheduling where different machines have heterogeneous manufacturing capacities. To solve it, this work proposes an algorithm named Genetic Simulated annealing-based Particle Swarm Optimization (GSPSO). Realistic data-based experiments demonstrate GSPSO reduces the cost of a multi-plant system by at least 23% than its typical peers.

Mobile Devices (MDs) support various delay/computation-intensive applications. Yet they only have limited battery energy and computing resources, thereby failing to totally run all applications. A mobile edge computing (MEC) paradigm has been proposed, and its servers are often deployed in both macro base stations (MBSs) and small base stations (SBSs). Thus, it is highly challenging to associate resource-limited MDs to them with high performance, and realize partial computation offloading among them for minimizing total energy consumption of an MEC system. This work formulates total energy consumption minimization as a constrained mixed integer non-linear program. To solve it, this work designs an improved meta-heuristic optimization algorithm called Particle swarm optimization based on Genetic Learning (PGL), which integrates strong local search capacity of a particle swarm optimizer, and genetic operations of a genetic algorithm. PGL jointly optimizes task offloading among MDs, SBSs and MBS, users’ connection to SBSs, MDs’ CPU speeds and transmission power, SBSs and MBS, and bandwidth allocation of available channels. Simulations with real-world data collected from Google cluster trace demonstrate that PGL significantly outperforms other existing methods in total energy consumption.

Data caching at mobile clients is an important technique for improving the performance of wireless data dissemination systems. However, variable data sizes, data updates, limited client resources, and frequent client disconnections make cache management a challenge. We propose a gain-based cache replacement policy, Min-SAUD, for wireless data dissemination when cache consistency must be enforced before a cached item is used. Min-SAUD considers several factors that affect cache performance, namely, access probability, update frequency, data size, retrieval delay, and cache validation cost. The paper employs stretch as the major performance metric since it accounts for the data service time and, thus, is fair when items have different sizes. We prove that Min-SAUD achieves optimal stretch under some standard assumptions. Moreover, a series of simulation experiments have been conducted to thoroughly evaluate the performance of Min-SAUD under various system configurations. The simulation results show that, in most cases, the Min-SAUD replacement policy substantially outperforms two existing policies, namely, LRU and SAIU.

A good cache management method for mobile wireless environments has to handle problems associated with limited client resources and frequent client disconnections, in addition to standard problems found in wired environments, such as variable data sizes and data updates. In this paper we propose a gain-based cache replacement policy, Min-SAUD, for wireless data dissemination when cache consistency must be enforced before a cached item is used. Min-SAUD considers several factors that affect cache performance, namely access probability, update frequency, data size, retrieval delay, and cache validation cost. Min-SAUD is optimal in terms of the stretch performance measure. Preliminary experimental results show that in most cases the Min-SAUD replacement policy substantially outperforms two existing policies, namely LRU and SAIU.

Abstract not available.

Studies power conservation techniques for multi-attribute queries on wireless data broadcast channels. Indexing data on broadcast channels can improve the client filtering capability, while clustering and scheduling can reduce both the access time and the tune-in time. Thus, indexing techniques should be coupled with clustering and scheduling methods to reduce the battery power consumption of mobile computers. In this study, three indexing schemes for multi-attribute queries, namely the index tree, signature and hybrid index, are discussed. We develop cost models for these three indexing schemes and evaluate their performance based on multi-attribute queries on wireless data broadcast channels.

This paper investigates power conserving indexing techniques for data disseminated on a broadcast channel. A hybrid indexing method combining strengths of the signature and the index tree techniques is presented. Different from previous studies, our research takes into consideration two important data organization factors, namely, clustering and scheduling. Cost models for index, signature and hybrid methods are derived by taking into account various data organizations accommodating these two factors. Based on our analytical comparisons, the signature and the hybrid indexing techniques are the best choices for power conserving indexing of various data organizations on wireless broadcast channels.

Various techniques have been developed to improve the performance of wireless information services. Techniques such as information broadcasting, caching of frequently accessed data, and point-to-point channels for pull-based data requests are often used to reduce data access time. To efficiently utilize information broadcast, indexing and scheduling schemes are employed for the organization of data broadcast. Most of the studies in the literature focused either on individual technique or a combination of them with some restrictive assumptions. There is no study considering these techniques working together in an integrated manner. In this paper, we propose a dynamic data delivery model for wireless communication environments. An important feature of our model is that data are disseminated through various storage mediums according to the dynamically collected data access patterns. Various results are presented in a set of simulation studies, which give some of the intuitions behind the design of a wireless data delivery system. 1

We discuss the wireless channel allocation problem for data dissemination in mobile computing systems. Methods for accessing data through broadcast and on-demand channels are described. We provide analytical models and cost formulae for the exclusive broadcast channels and the exclusive on-demand channels and propose a dynamic channel allocation algorithm for optimizing system performance. Our performance evaluation shows that dynamic channel allocation significantly improves system performance and the channel allocation algorithm gives us the optimal solution for various system parameter settings.

Caching of frequently accessed data items can reduce the bandwidth requirement in a mobile wireless computing environment. Periodically broadcast of invalidation reports is an efficient cache invalidation strategy. However, this strategy is severely affected by the disconnection and mobility of the clients. In this paper, we present two adaptive cache invalidation report methods, in which the server broadcasts different invalidation reports according to the update and query rates/patterns and client disconnection time while spending little uplink cost. Simulation results show that the adaptive invalidation methods are efficient in improving mobile caching and reducing the uplink and downlink costs without degrading the system throughput.

We discuss several channel allocation methods for data dissemination in mobile computing systems. We suggest that the broadcast and on-demand channels have different access performance under different system parameters and that a mobile cell should use a combination of both to obtain optimal access time for a given workload and system parameters. We study the data access efficiency of three channel configurations: all channels are used as on-demand channels (exclusive on-demand); all channels are used for broadcast (exclusive broadcast); and some channels are on-demand channels and some are broadcast channels (hybrid). Simulations on obtaining the optimal channel allocation for lightly-loaded, medium-loaded, and heavy-loaded conditions is conducted and the result shows that an optimal channel allocation significantly improves the system performance.