Self-learning Strategy Research Articles

Temporal online self-learning stochastic configuration networks: A study on strip deviation prediction

Stochastic configuration networks (SCNs), with the rapid learning abilities and universal approximation properties, have garnered growing interest in the industrial domain in recent years. In the hot rolling process, strip deviation affects product quality and production safety. In this paper, a novel model named temporal online self-learning stochastic configuration networks (TOSL-SCNs) designed for early prediction of strip deviation in the hot rolling process is introduced. To address the challenges of time series prediction, conventional SCNs are extended to a two-dimensional format, enabling effective processing of time series data. Given the dynamic nature of industrial environments, the model incorporates an online self-learning capability to adapt to nonstationary data. Additionally, to ensure the stability of the model’s predictive performance, a data buffer strategy is introduced to dynamically refresh the training dataset. This paper demonstrates the capabilities of TOSL-SCNs in time series modeling, online self-learning, and data buffer strategy from multiple perspectives. The case study on strip deviation prediction highlights the necessity of these capabilities in industrial applications.

A Self-Learning Hyper-Heuristic Algorithm Based on a Genetic Algorithm: A Case Study on Prefabricated Modular Cabin Unit Logistics Scheduling in a Cruise Ship Manufacturer.

Hyper-heuristic algorithms are known for their flexibility and efficiency, making them suitable for solving engineering optimization problems with complex constraints. This paper introduces a self-learning hyper-heuristic algorithm based on a genetic algorithm (GA-SLHH) designed to tackle the logistics scheduling problem of prefabricated modular cabin units (PMCUs) in cruise ships. This problem can be regarded as a multi-objective fuzzy logistics collaborative scheduling problem. Hyper-heuristic algorithms effectively avoid the extensive evaluation and repair of infeasible solutions during the iterative process, which is a common issue in meta-heuristic algorithms. The GA-SLHH employs a genetic algorithm combined with a self-learning strategy as its high-level strategy (HLS), optimizing low-level heuristics (LLHs) while uncovering potential relationships between adjacent decision-making stages. LLHs utilize classic scheduling rules as solution support. Multiple sets of numerical experiments demonstrate that the GA-SLHH exhibits a stronger comprehensive optimization ability and stability when solving this problem. Finally, the validity of the GA-SLHH in addressing real-world decision-making issues in cruise ship manufacturing companies is validated through practical enterprise cases. The results of a practical enterprise case show that the scheme solved using the proposed GA-SLHH can reduce the transportation time by up to 37%.

A domain feature decoupling network for rotating machinery fault diagnosis under unseen operating conditions

Operating conditions reflect the mission evolution of rotating machinery in specific application scenarios. The monitoring data under different operating conditions exhibit disparate statistic distributions in terms of amplitude and frequency properties. In the modern operation and maintenance, domain adaptation approaches, which align the feature distributions of source and target domains, are commonly employed to address cross-domain diagnosis challenges. However, the availability of the target domain is serendipitous rather than guaranteed. To meet the engineering requirements, a domain feature decoupling network (DFDN) is proposed to achieve the fault diagnosis of rotating machinery under unseen operating conditions. The core concept of DFDN is to decouple the extracted features into domain-related features and fault-related features. These features are then adaptively integrated into domain-invariant features using a significance fusion method, thereby enabling the feature learning and knowledge transfer from seen source domains to unseen target domains. Furthermore, a self-learning joint optimization strategy is developed for DFDN to facilitate the efficient acquisition of generalized diagnosis knowledge. Two bearing datasets from different test rigs are considered to assess the domain generalization performance of DFDN. The experimental results demonstrate the superiority and potential of this method in coping with zero-shot fault diagnosis under unseen operating conditions.

Design of sliding mode controller with improved reaching law through self-learning strategy to mitigate the torque ripple in BLDC motor for electric vehicles

Brushless Direct Current (BLDC) motors and their evolution in electric vehicles (EVs) garner attention for their streamlined operation and control methods, enhancing driving range while reducing energy consumption. The primary objective is integrating sliding mode control to enhance BLDC motor capabilities, efficiently regulating speed and torque. The proposed sliding mode (SM) controller mitigates torque ripple and improves dynamic performance amidst fluctuating loads. Speed stabilization and torque control involve comparing reference and predicted speeds, with error feedback integrated into the SM controller. Integrating real-time rules and numerical data with an ANN (Artificial Neural Network)-Fuzzy self-learning algorithm empowers the SM controller to anticipate and mitigate error rates in the BLDC motor, efficaciously modulating motor torque with minimal ripple. Extensive MATLAB/Simulink model-in-loop (MIL) and hardware-in-loop (HIL) simulations demonstrate SM controller superiority over conventional PID, Fuzzy, and PID-Fuzzy controllers, with a rise time of 0.01 s, steady-state error of 0.001 %, settling time of 0.02 s, and peak overshoot of 0.066 %. Moreover, the experiment results show an impressive 29.41 % reduction in torque ripple and the development of efficiency maps to attain the maximum efficiency of 96.12 %. This initiative substantiates that the proposed SM controller to enhance BLDC motor efficiency and EV driving range across real-time conditions.

An adaptive semi-supervised self-learning method for online state of health estimation of lithium-ion batteries

Accurate and precise online estimation of the state of health (SOH) is crucial when managing lithium-ion batteries. Most existing SOH estimation methods rely on supervised learning algorithms utilizing large amounts of labeled data. However, lithium-ion batteries are typically operated under dynamic conditions, including significant amounts of unlabeled charging or discharging data in online application scenarios. To fully utilize these data, we propose an adaptive semi-supervised self-learning teacher-student model (AS3LTSM) method for online SOH estimation. First, four physically interpretable health indicators (PIHIs) are extracted from the voltage and current data. The Pearson correlation coefficient (PCC) is used to assess significant associations between PIHIs and the SOH. Regressive and autoregressive long short-term memory (LSTM) models are selected as the teacher and student networks. Knowledge is transferred from the teacher to the student through pseudolabels, which guide the updating and evolution of the student network. Furthermore, a self-learning strategy and a retraining process for improving the long-term estimation accuracy are proposed. Two public datasets are used for comparison and ablation experiments. Experimental analysis validates the improved effectiveness and performance of the proposed method, with the RMSE and MAPE of the three experimental groups all within 1.3 % and 1.29 %, respectively.

Graph Attentive Dual Ensemble learning for Unsupervised Domain Adaptation on point clouds

Due to the annotation difficulty of point clouds, Unsupervised Domain Adaptation (UDA) is a promising direction to address unlabeled point cloud classification and segmentation. Recent works show that adding a self-supervised learning branch for target domain training consistently boosts UDA point cloud tasks. However, most of these works simply resort to geometric deformation, which ignores semantic information and is hard to bridge the domain gap. In this paper, we propose a novel self-learning strategy for UDA on point clouds, termed as Graph Attentive Dual Ensemble learning (GRADE), which delivers semantic information directly. Specifically, with a pre-training process on the source domain, GRADE further builds dual collaborative training branches on the target domain, where each of them constructs a temporal average teacher model and distills its pseudo labels to the other branch. To achieve faithful labels from each teacher model, we improve the popular DGCNN architecture by introducing a dynamic graph attentive module to mine the relation between local neighborhood points. We conduct extensive experiments on several UDA point cloud benchmarks, and the results demonstrate that our GRADE method outperforms the state-of-the-art methods on both classification and segmentation tasks with clear margins.

A novel reinforcement learning based Heap-based optimizer

The Heap-Based Optimization (HBO), a novel meta-heuristic algorithm, constructs a hierarchical heap resembling a company structure to facilitate interaction among search agents. This enables the search agent to learn various levels of information from the population to obtain improved solutions, but it has drawbacks such as limited search capability and slow convergence speed when tackling complex optimization problems. To address these challenges, we propose an enhanced heap optimizer based on reinforcement learning(RLHBO), which demonstrates robust stability, rapid convergence, and high solution accuracy. Firstly, fixed operators are replaced with a reinforcement learning dynamic control search strategy in RLHBO to regulate the search range of agents. This prevents agents from being restricted in their search when encountering local or global optima, thus enhancing their search capability. Additionally, RLHBO introduces a dimensional self-learning strategy and a convergence strategy based on search direction similarity to address issues such as invalid and discrete solutions, accelerating convergence. Furthermore, an initialization strategy based on quasi-oppositional learning is employed to increase the diversity of the algorithm-generated population, surpassing the limitations of random initialization and further boosting convergence speed. Extensive testing on various complex functions from CEC2017 and CEC2022 demonstrates that RLHBO, as presented in this paper, outperforms HBO and numerous other advanced MAs in terms of search capability and convergence speed.

Physics-Based Self-Learning Spiking Neural Network enhanced time-integration scheme for computing viscoplastic structural finite element response

The present study introduces a new physics-based self-learning spiking neural framework to compute geometrically and physically nonlinear structural response. While the so-called traditional or second-generation deep neural networks are used in many applications in the class of physics-informed neural networks, we propose a hybrid model that consists of third-generation Leaky-Integrated and Fire (LIF) neurons, Recurrent Leaky-Integrated and Fire (RLIF) neurons and second-generation dense transformation. The third-generation neurons are inspired by the human brain’s energy-efficient functioning which introduces inherent temporal and sparse behavior leading to a more sustainable AI approach. However, the sparse nature of the spiking neurons poses a challenge to tackle nonlinear regression tasks. Thus, in the present study, we use an autoencoding strategy that converts the real-valued signals to its spiking representation and enables the spiking neurons to learn nonlinear material response. The proposed hybrid network is firstly pretrained with combined data-driven and physics-based loss functions and then deployed in the plastic corrector step of the implicit integration which is used in the Finite Element solver and is validated on a series of Boundary Value Problems (BVPs) consisting of plate elements. A self-learning strategy is introduced for the hybrid network to train itself during FE simulation using the proposed physics-based loss function. Two major advantages were observed: an overall computational gain in the excess of 30% and the self-learning/online training ability of the model that bolsters its convergence behavior. Finally, the proposed third-generation layers are deployed on the Xylo-Av2 neuromorphic chip and its energy performance is compared with the second-generation Recurrent Neural Networks (RNNs).

A Multi-strategy Enhanced Arithmetic Optimization Algorithm and Its Application in Path Planning of Mobile Robots

A multi-strategy enhanced arithmetic optimization algorithm called MSEAOA is proposed to address the issues of low population diversity, imbalanced exploration and exploitation capabilities, and low accuracy of optimal solution in the Arithmetic Optimization Algorithm. Firstly, using the good point set strategy for population initialization to improve population diversity and thus accelerate convergence speed. Secondly, we integrate the exploration and exploition capabilities of differential self-learning strategy, best example learning strategy, and second-order differential perturbation strategy balancing algorithm. Finally, the introduction of somersault foraging strategy improves the accuracy of the optimal solution. We select 14 classical benchmark test functions and the CEC2019 function test set to test the optimization ability of MSEAOA, and apply MSEAOA to the path planning problem of mobile robots. MSEAOA is compared with other meta-heuristic optimization algorithms, and the experimental results are statistically analyzed by the Wilcoxon rank-sum test. The simulation experimental results show that MSEAOA performs the best among 14 benchmark functions, but for 10 CEC2019 functions, MSEAOA has the best optimization performance among 5 of them (50%). In the path optimization problem of mobile robots, the path obtained by MSEAOA is also the best among all algorithms, its path shortening rate exceeds 8.8% in 83% of environments. The results indicate that MSEAOA is a reliable algorithm suitable for function optimization and practical optimization problems.

Diseño de estrategia de aprendizaje para la licenciatura de diseño industrial bajo un enfoque de la educación 4.0

This study emphasizes how the implementation of a learning strategy based on an education 4.0 approach can influence the assimilation of knowledge in students of the industrial design degree (ID). The research project was conducted at a higher education institution in Querétaro, Mexico. A case study methodology was employed to compare a self-learning strategy based on audiovisual media with the traditional model, which is still used in most classes. The study compared the data from two experiment interventions, in which a design project was presented, thus following the theoretical-practical approach of the ID program. The groups segmentation was done through a process of randomization of the participants. The purpose of the experiment was to check the effectiveness of the traditional teaching method compared to one based on the implementation of audiovisual media, which presented the content of the project, through a series of videos and the Google classroom platform. The results showed that the traditional method obtained a better score compared to the method that uses audiovisual media, however, from the feedback of teachers and students, it was concluded that the creation of a hybrid method may be the most efficient alternative to achieve better learning.

A Colorectal Coordinate-Driven Method for Colorectum and Colorectal Cancer Segmentation in Conventional CT Scans.

Automated colorectal cancer (CRC) segmentation in medical imaging is the key to achieving automation of CRC detection, staging, and treatment response monitoring. Compared with magnetic resonance imaging (MRI) and computed tomography colonography (CTC), conventional computed tomography (CT) has enormous potential because of its broad implementation, superiority for the hollow viscera (colon), and convenience without needing bowel preparation. However, the segmentation of CRC in conventional CT is more challenging due to the difficulties presenting with the unprepared bowel, such as distinguishing the colorectum from other structures with similar appearance and distinguishing the CRC from the contents of the colorectum. To tackle these challenges, we introduce DeepCRC-SL, the first automated segmentation algorithm for CRC and colorectum in conventional contrast-enhanced CT scans. We propose a topology-aware deep learning-based approach, which builds a novel 1-D colorectal coordinate system and encodes each voxel of the colorectum with a relative position along the coordinate system. We then induce an auxiliary regression task to predict the colorectal coordinate value of each voxel, aiming to integrate global topology into the segmentation network and thus improve the colorectum's continuity. Self-attention layers are utilized to capture global contexts for the coordinate regression task and enhance the ability to differentiate CRC and colorectum tissues. Moreover, a coordinate-driven self-learning (SL) strategy is introduced to leverage a large amount of unlabeled data to improve segmentation performance. We validate the proposed approach on a dataset including 227 labeled and 585 unlabeled CRC cases by fivefold cross-validation. Experimental results demonstrate that our method outperforms some recent related segmentation methods and achieves the segmentation accuracy in DSC for CRC of 0.669 and colorectum of 0.892, reaching to the performance (at 0.639 and 0.890, respectively) of a medical resident with two years of specialized CRC imaging fellowship.

A self-learning knowledge-based MOEA/D for distributed heterogeneous assembly permutation flowshop scheduling with batch delivery

With the requirements of just-in-time and make-to-order, it is necessary to consider the batch delivery in distributed heterogeneous assembly flow shop scheduling as well as the minimization of inventory and tardiness costs. Hence, this work proposes a new mixed integer linear programming model and a self-learning knowledge-based multi-objective evolutionary algorithm based on decomposition. In this algorithm, 4 problem-specific properties are extracted and embedded in the designed solution representation, forward-reverse objective calculation and local search operators. A new hybrid initialization with six knowledge-based NEH heuristics and a random generation is developed to generate better initial solutions. A self-learning selection strategy is developed to coordinate the above knowledge-based local search operators. The experimental study, based on 900 small and 540 large instances, demonstrates that the designed components such as hybrid initialization and self-learning selection strategy are effective, and the proposed algorithm is superior to 6 existing algorithms in the literature in solving this new problem.

Daily load prediction study on SCASL and BiGRU models

In the short-term power load prediction, the environmental uncertainty will have an impact on the prediction accuracy and increase the complexity of the prediction. To deal with this problem, the paper proposes a bidirectional gated recurrent unit (BiGRU) network model for and sine cosine optimization algorithm with self-learning strategy and Levy flight (SCASL). With real power load data as a data set, high correlation parameters are selected through the Pearson correlation coefficient analysis as input, using the SCASL algorithm and nonlinear weight factor. Finally, the key time points in capturing the historical information of the load changes are established according to the optimization parameters, and BiGRU model is used to learn to predict its accuracy. The experimental results show that the MAPE (Mean Absolute Percentage Error) and RMSE (Root Mean Square Error) of BiGRU model are better than the comparison model gated recurrent unit (GRU) network model. The final results show that the prediction accuracy of the power load system has been significantly improved.

Unlocking operational excellence: A deep dive into a communication-driven multi-strategy state transition algorithm for industrial process optimization

The optimization of industrial processes is crucial for enhancing operational safety, productivity, and energy efficiency. However, the increasing complexity of industrial processes poses challenges to the applicability of intelligent optimization algorithms, such as the state transition algorithm. The main challenges include limited individual diversity, the intricate task of escaping local optima, and the underutilization of solutions beyond predefined boundaries. To overcome these challenges, this study proposes a novel intelligent stochastic optimization algorithm called the communication-driven multi-strategy state transition algorithm (CDMS-STA). CDMS-STA incorporates three self-designed strategies: communication, self-learning, and cross-boundary solution processing. Specifically, the communication strategy enhances the diversity of candidate solutions by promoting information exchange and sharing among individuals. The self-learning strategy facilitates adaptive adjustments based on historical information. The cross-boundary solution processing strategy maximizes the utilization of solutions beyond defined boundaries. To evaluate the effectiveness of the proposed algorithm, extensive experiments were conducted on a numerical sample case and two chemical process cases. The experimental results demonstrate that CDMS-STA outperforms several state-of-the-art optimization methods to find the optimal solution, which also validates the potential of the proposed algorithm for practical application in industrial process optimization.

Self-learning sliding mode control based on adaptive dynamic programming for nonholonomic mobile robots

This paper proposes a self-learning sliding mode control (SlSMC) strategy with stability guarantee for the trajectory tracking of nonholonomic mobile robots (NMRs) under matched uncertainties, which improves the control performance of NMRs by optimizing the reaching law and the sliding mode surface of SMC as well as retaining the finite-time convergence and the robustness to uncertainties. In the presence of adverse factors such as skidding, slipping and environmental noise, the kinematic model of NMRs is reconstructed and an integral terminal sliding mode controller is designed for the trajectory tracking of NMRs. Then, based on the sliding mode controller, the proposed control strategy formulates the optimization of the SMC’s reaching law and the sliding mode surface under stability constraints as two asynchronous optimal control problems with control constraints. Meanwhile, an online continuous-time receding-horizon optimization mechanism based on an actor-critic algorithm is proposed to solve the optimal problems asynchronously and improve online learning efficiency. The stability and the convergence of the proposed strategy are validated both in theory and simulations. Furthermore, extensive contrastive simulation results illustrate that the proposed receding horizon learning-based control strategy outperforms three recent methods in control performance. Finally, experiments of the proposed self-learning SMC strategy are carried out based on a real intelligent vehicle, and the experimental results also verify that the proposed method can meet the actual control needs of NMRs.

Impact of self-learning based high-frequency traders on the stock market

In this paper we investigate the role of self-learning agents in multi-agent models of financial markets. We develop an agent-based simulation model of a financial market and, in addition to the agents with fixed strategies used in previous research, we introduce an agent with a self-learning strategy. To model the behavior of such an agent, we use deep reinforcement learning algorithms, namely deep deterministic policy gradient (DDPG). Next, we conduct a comparative analysis of the results of the constructed model with outcomes of previously proposed models, as well as with the characteristics of real market. To conduct comparative analysis, we use stylized facts of asset returns that allow us to evaluate and compare the characteristics of the markets. Our results show that a model with a self-learning agent gives a better approximation of the real market than a model with classic agents. In particular, unlike the model with classical agents, the model with a self-learning agent turns out to be not so heavy-tailed. Thus, we demonstrate that for a complete understanding of market processes simulation models should take into account self-learning agents that have a significant presence at modern financial markets.Finally, we present the python package11https://github.com/Kiri28/Agent-Based_Market_simulator., which was developed by us as part of the research implementation. This package allows to simulate the financial market, as well as create your own agents and evaluate their impact on the market.

Function approximation reinforcement learning of energy management with the fuzzy REINFORCE for fuel cell hybrid electric vehicles

In the paper, a novel self-learning energy management strategy (EMS) is proposed for fuel cell hybrid electric vehicles (FCHEV) to achieve the hydrogen saving and maintain the battery operation. In the EMS, it is proposed to approximate the EMS policy function with fuzzy inference system (FIS) and learn the policy parameters through policy gradient reinforcement learning (PGRL). Thus, a so-called Fuzzy REINFORCE algorithm is first proposed and studied for EMS problem in the paper. Fuzzy REINFORCE is a model-free method that the EMS agent can learn itself through interactions with environment, which makes it independent of model accuracy, prior knowledge, and expert experience. Meanwhile, to stabilize the training process, a fuzzy baseline function is adopted to approximate the value function based on FIS without affecting the policy gradient direction. Moreover, the drawbacks of traditional reinforcement learning such as high computation burden, long convergence time, can also be overcome. The effectiveness of the proposed methods were verified by Hardware-in-Loop experiments. The adaptability of the proposed method to the changes of driving conditions and system states is also verified.

A dynamic self-learning grid-responsive strategy for battery sharing economy—multi-objective optimisation and posteriori multi-criteria decision making

District building-vehicle interactive energy network with systematic energy interactions can improve techno-economic performances and enhance cleaner power penetration, whereas the battery cycling aging will be accelerated due to an increase in frequency of charging/discharging cycles. In this study, a dynamic self-learning grid-responsive strategy was proposed to improve techno-economic performances, energy flexibility, and the relative capacity (RC) of electrochemical batteries, through dynamically adjustable off-peak grid charging power. A collaborative framework on an energy district with building-integrated photovoltaics and electric vehicles has been formulated, through synergic energy interactions and integrations. Cycling aging of both static batteries in buildings and mobile vehicle batteries has been considered. In respect to intrinsic contradiction characteristics of each objective, trade-off solutions have been proposed through multi-objective optimisation. Furthermore, regarding the multiple optimal solutions along the Pareto front, a weighted Eulerian distance-based methodology was adopted to identify the ‘best of best’ solution and then assist multi-criteria decision-making. Research results show that, with the proposed dynamic self-learning grid-responsive strategy, the grid can provide a more flexible grid-to-battery charging power with a higher RC of batteries, compared to the strategy with dynamic renewable-demand signals. Through the multi-objective optimisation, the import cost can be reduced from 213.2 to 203.8 HK$/m2·a, and the RC can be improved from 0.800 to 0.999. This study also demonstrates a weighted Eulerian distance-based methodology to identify the ‘best of best’ solution to multi-criteria decision-makers. Research results are useful to promote district interactive energy networks in smart cities, with multivariant optimal design, Pareto optimal solutions, and weighted Eulerian distance-based ‘best of best’ solution.

Dynamic job shop scheduling based on deep reinforcement learning for multi-agent manufacturing systems

• The manufacturing equipment in the workshop is constructed as an equipment agent with the support of edge computing node, and a multi-agent manufacturing system with the ability of online scheduling and scheduling strategy optimization is presented. • A self-organizing negotiation mechanism based on contract network protocol (CNP) is proposed to guide the cooperation and competition among multiple agents , and to provide support for intelligent decision-making. • The decision-making module of equipment agent called artificial intelligence (AI) scheduler is established based on the multi-layer perceptron, and enables the manufacturing equipment to intelligently generate an optimal production strategy according to the perceived workshop state. • An adaptive learning strategy of AI scheduler based on proximal policy optimization (PPO) algorithm is proposed to improve its decision-making performance under the disturbance of orders and resources. • Experimental results show that the proposed methodology achieves better performance in terms of production strategy learning and disturbance handling compared with other optimization-based approaches and dispatching rules . Personalized orders bring challenges to the production paradigm, and there is an urgent need for the dynamic responsiveness and self-adjustment ability of the workshop. Traditional dispatching rules and heuristic algorithms solve the production planning and control problems by making schedules. However, the previous methods cannot work well in a changeable workshop environment when encountering a large number of stochastic disturbances of orders and resources. Recently, the potential of artificial intelligence (AI) algorithms in solving the dynamic scheduling problem has attracted researchers' attention. Therefore, this paper presents a multi-agent manufacturing system based on deep reinforcement learning (DRL), which integrates the self-organization mechanism and self-learning strategy. Firstly, the manufacturing equipment in the workshop is constructed as an equipment agent with the support of edge computing node, and an improved contract network protocol (CNP) is applied to guide the cooperation and competition among multiple agents, so as to complete personalized orders efficiently. Secondly, a multi-layer perceptron is employed to establish the decision-making module called AI scheduler inside the equipment agent. According to the perceived workshop state information, AI scheduler intelligently generates an optimal production strategy to perform task allocation. Then, based on the collected sample trajectories of scheduling process, AI scheduler is periodically trained and updated through the proximal policy optimization (PPO) algorithm to improve its decision-making performance. Finally, in the multi-agent manufacturing system testbed, dynamic events such as stochastic job insertions and unpredictable machine failures are considered in the verification experiments. The experimental results show that the proposed method is capable of obtaining the scheduling solutions that meet various performance metrics, as well as dealing with resource or task disturbances efficiently and autonomously.

A Self-Learning Discrete Jaya Algorithm for Multiobjective Energy-Efficient Distributed No-Idle Flow-Shop Scheduling Problem in Heterogeneous Factory System.

In this study, a self-learning discrete Jaya algorithm (SD-Jaya) is proposed to address the energy-efficient distributed no-idle flow-shop scheduling problem (FSP) in a heterogeneous factory system (HFS-EEDNIFSP) with the criteria of minimizing the total tardiness (TTD), total energy consumption (TEC), and factory load balancing (FLB). First, the mixed-integer programming model of HFS-EEDNIFSP is presented. An evaluation criterion of FLB combining the energy consumption and the completion time is introduced. Second, a self-learning operators selection strategy, in which the success rate of each operator is summarized as knowledge, is designed for guiding the selection of operators. Third, the energy-saving strategy is proposed for reducing the TEC. The energy-efficient no-idle FSP is transformed to be an energy-efficient permutation FSP to search the idle times. The speed of operations which adjacent are idle times is reduced. The effectiveness of SD-Jaya is tested on 60 benchmark instances. On the quality of the solution, the experimental results reveal that the efficacy of the SD-Jaya algorithm outperforms the other algorithms for addressing HFS-EEDNIFSP.