Multi-population Genetic Algorithm Research Articles

Research on Credit Default Prediction Model Based on TabNet-Stacking.

With the development of financial technology, the traditional experience-based and single-network credit default prediction model can no longer meet the current needs. This manuscript proposes a credit default prediction model based on TabNeT-Stacking. First, use the PyTorch deep learning framework to construct an improved TabNet structure. The multi-population genetic algorithm is used to optimize the Attention Transformer automatic feature selection module. The particle swarm algorithm is used to optimize the hyperparameter selection and achieve automatic parameter search. Finally, Stacking ensemble learning is used, and the improved TabNet is used to extract features. XGBoost (eXtreme Gradient Boosting), LightGBM (Light Gradient Boosting Machine), CatBoost (Category Boosting), KNN (K-NearestNeighbor), and SVM (Support Vector Machine) are selected as the first-layer base learners, and XGBoost is used as the second-layer meta-learner. The experimental results show that compared with original models, the credit default prediction model proposed in this manuscript outperforms the comparison models in terms of accuracy, precision, recall, F1 score, and AUC (Area Under the Curve) of credit default prediction results.

Multi-population genetic algorithm with crowding-based local search for fuzzy multi-objective supply chain configuration

Supply chain configuration is often fuzzy and involves multiple objectives in real-world scenarios, but existing researches lack the exploration in the fuzzy aspect. Therefore, this paper establishes a fuzzy multi-objective supply chain configuration problem model to minimize the lead time and product cost oriented towards real supply chain environments. To solve the fuzzy problem, the theories of membership and closeness degree in fuzzy mathematics are adopted, and a multi-population genetic algorithm (MPGA) with crowding-based local search method is proposed. The MPGA algorithm uses two populations for optimizing the two objectives separately and effectively, and is characterized by three main innovative aspects. Firstly, a radical-and-radial selection operator is designed to balance the convergence speed and diversity of population. In the early stage of the algorithm, two populations are both optimized towards the ideal knee point, and then are separately optimized towards the two ends of the Pareto front (PF). Secondly, an elitist crossover operator is devised to promote information exchange within two populations. Thirdly, a crowding-based local search is proposed to speed up convergence by improving the crowded solutions, and to enhance diversity by obtaining new solutions around the uncrowded ones. Comprehensive experiments are tested on a fuzzy dataset with different sizes, and the integral of the hypervolume of PF is used for the evaluation of the fuzzy PF. The results show that MPGA achieves the best performance over other comparative algorithms, especially on maximum spread metric, outperforming all others by an average of 39 % across all test instances.

Application of improved multi-population genetic algorithm in structural optimization of automotive electrical equipment

Application of improved multi-population genetic algorithm in structural optimization of automotive electrical equipment

Few shot intelligent assessment method for compaction quality of earthwork considering uncertainty

Current practices lack reliable models for continuously assessing compaction quality from limited data, which hinders effective compaction quality control and the acceptance of earthwork. Using material source parameters as inputs for assessment models makes it infeasible to achieve a real-time assessment of the compaction quality in earthwork. These issues have hindered the popularization of emerging intelligent compaction (IC) and intelligent rolling compaction (IRC) technologies. Therefore, this study addresses the need for small-sample training by introducing an uncertainty-based multi-population genetic algorithm fused with a backpropagation neural network model (UB-MPGA-BP), enabling the real-time intelligent assessment of compaction quality based on few-shot learning. In case studies involving high- and low-liquid limit silts, existing models trained on large datasets were notably inadequate for small-sample evaluation tasks. However, the proposed method not only performed well in silt cases but also exhibited high performance in rockfill materials. The analysis revealed that in small-sample scenarios without reliance on material source parameters, the proposed method has the advantages of real-time accuracy, generalization, few-shot data-driven, and simple input parameters, which enable the rapid assessment of compaction quality. The model can potentially facilitate IC/IRC applications in other engineering practices.

Ship scheduling problem in an anchorage-to-quay channel with water discharge restrictions

The optimization of ship scheduling in water-land transshipment plays a crucial part in enhancing transportation efficiency and alleviating traffic congestion on inland waterways, particularly when the service capacity of locks is insufficient. This study focuses on the ship scheduling problem in an anchorage-to-quay channel, where the navigation of ships is influenced by the water discharge during flood seasons. A new mixed-integer linear programming (MILP) model is formulated for this problem, in which the interconnected decision of ships’ navigation and berthing activities, the variation of water discharge in the waterway channel, and the change of navigation scenarios (i.e., normal or flood impact) are explicitly taken into account. Then a multi-population genetic algorithm (MPGA) is applied to solve this model. Numerical experiments conducted for the Three Gorges Dam (TGD) demonstrated that the proposed heuristic method can tackle the ship scheduling problem within a reasonable time, which is also effective in assisting ships to match acceptable water discharge time window by slightly adjusting the departure time of ships at the anchorage or quay. Furthermore, the proposed method can ensure the navigational safety of ships and improve the efficiency of ship transportation in different scenarios, and quantitatively evaluate the water discharge impact.

Seismic performance of soft soil foundation with a new type of assembled wall tuned mass damper.

The design of tuned mass damper (TMD) parameters is influenced by the soil-structure-TMD coupling system; thus, it is important to consider the soil-structure interaction (SSI) for the vibration control effect of the TMD. Recently, the acquisition of TMD parameters considering soil-structure interactions has only remained at the theoretical stage, lacking relevant experimental verification. Traditional TMD face the problems of occupying a large building space, increasing construction costs, and non-replaceable components. In this study, an assembled wall-type damping TMD was designed. By comparing the dynamic response of the uncontrolled and controlled structures equipped with the newly assembled wall-type damping TMD in the shaking table test on a soft soil foundation, we analyzed whether the SSI effect was considered in the TMD design parameters on the damping effect of the newly assembled wall-type tuned mass damper. The TMD parameters optimized using the artificial intelligence algorithm were verified experimentally. The results indicated that the traditional TMD design parameters were discordant because the SSI effect was not considered. The SSI effect in the soil effectively reduces the dynamic response of the superstructure. By considering the SSI effect and improving the multi-population genetic algorithm, a wall-type damping TMD with optimized parameters can achieve a good damping effect.

Flexible job-shop scheduling problem with parallel batch machines based on an enhanced multi-population genetic algorithm

AbstractThe flexible job-shop scheduling problem (FJSP) with parallel batch processing machine (PBM) is one of those long-standing issues that needs cutting-edge approaches. It is a recent extension of standard flexible job shop scheduling problems. Despite their wide application and prevalence in practical production, it seems that current research on these types of combinatorial optimization problems remains limited and uninvestigated. More specifically, existing research mainly concentrates on the flow shop scenarios in parallel batch machines for job shop scheduling but few literature emphasis on the flexible job shop integration in these contexts. To directly address the above mentioned problems, this paper establishes an optimization model considering parallel batch processing machines, aiming to minimize the maximum completion time in operating and production environments. The proposed solution merges variable neighborhood search with multi-population genetic algorithms, conducting a neighborhood search on the elite population to reduce the likelihood of falling into local optima. Subsequently, its applicability was evaluated in computational experiments using real production scenarios from a partnering enterprise and extended datasets. The findings from the analyses indicate that the enhanced algorithm can decrease the objective value by as much as 15% compared to other standard algorithms. Importantly, the proposed approach effectively resolves flexible job shop scheduling problems involving parallel batch processing machines. The contribution of the research is providing substantial theoretical support for enterprise production scheduling.

Hybrid optimization algorithm for estimating soil parameters of spoil hopper deposition model for trailing suction hopper dredgers

Trailing suction hopper dredger is a kind of hydraulic dredger, it has the characteristics of self-propelled, selfloading, self-dredging, self-unloading, it is the main force in dredging and blowing works, it is widely used in the world, it can be said that where there is a big dredging project where there is a trailing suction hopper dredger’s figure. The loading optimization process of trailing suction hopper dredger contains a lot of dredging parameters related to soil type, and the soil type under different working conditions is not very clear. In this study, we present a hybrid optimization technique based on simulated annealing and multi-population genetic algorithm to enhance the loading efficiency of a trailing suction hopper dredger and to examine the variation of dredged soil parameters. The soil parameters of the spoil hopper deposition model were estimated using this hybrid optimization algorithm. The experimental results show that the soil parameters are successfully estimated and verified by our measured construction data of a trailing suction hopper dredger. In addition, our proposed method has the highest accuracy of soil parameter estimation, the fastest algorithm convergence, and excellent robustness compared to the other three intelligent optimization methods. In addition, our method successfully avoids the phenomenon of premature convergence that usually occurs in traditional genetic algorithms, and the parameters show strong adaptability to different vessels under the same dredging area.

Test data generation for covering mutation-based path using MGA for MPI program

Message Passing Interface (MPI) is a communication protocol used for parallel programming in various languages, valued for its reliability and broad applicability. Mutation testing is a software testing method for systematically simulating software faults. However, the significant number of inserted mutation branches in the program escalates the testing cost. To address this issue, we propose the comprehensive PMTDGM framework. The framework first generates mutation-based paths based on the relevancy of mutant branches and the difficulty of mutant branch coverage, followed by the establishment of a multitask model of path coverage. Finally, we employ a multi-population genetic algorithm (MGA) to generate test data. Our experiments, performed on six MPI programs of varying sizes and structures, demonstrate that the mutation-based paths have the small test sets and are easy to cover. Additionally, the multitask model and MGA can significantly improve the efficiency of generating test data and reduce the cost of mutation testing for MPI programs compared to traditional methods.

Intelligent optimization of axial-flow pump using physics-considering machine learning

Abstract To address the significant energy waste generated by axial flow pumps, this paper proposes an intelligent optimization method based on physics-considering machine learning. First, a highly parameterized geometric design theory is constructed using six featured variables to achieve a complete three-dimensional modeling of the blade geometry. Four hundred preliminary cases are studied using the computational fluid dynamics method with various combinations of these featured variables to obtain a preliminary solution. The best preliminary design has an efficiency of 83.33%, and a head of 5.495 m. To further improve this performance, this paper also presents a high-precision prediction model for the energy performance of axial flow pump based on back-propagation neural network and the encoding layers of random sampling and local feature aggregator network created. Afterwards, a multi-population genetic algorithm is used to quickly find the optimal solution within the prediction mode range. The algorithm achieved a highest efficiency of 86.373% and was validated by numerical simulation with a value of 86.057% and a prediction error of 0.316%. Compared with the preliminary solution, the efficiency of the optimized axial flow pump is increased by 1.615%, with a wider high-efficiency range and an optimal operating point closer to the design conditions. Overall, this intelligent optimization method has the potential to significantly reduce the design time of axial pumps and increase their performance.

A voltage control method for distribution networks based on TCN and MPGA under cloud edge collaborative architecture

A distribution network voltage control method based on TCN and MPGA under cloud edge collaborative architecture is proposed to address the issues of heavy computational burden and inability to balance economic efficiency in traditional decentralized and centralized voltage control methods. Firstly, under the cloud edge collaboration architecture, the power prediction model based on time convolutional networks is trained in the cloud, and the edge predicts the load demand and new energy active output in each region based on the TCN Attention power prediction model issued by the cloud. Then, considering safety and economy comprehensively, with the goal of minimizing the daily network loss and voltage regulation cost in the distribution network, the Pareto solution set for the optimal operation point of each voltage regulation equipment in the entire distribution network was solved based on an improved multi population genetic algorithm at the distribution cloud main station. Finally, the optimal distribution network voltage control scheme is obtained based on the Pareto solution set. Based on the simulation system, experimental testing was conducted on the proposed method, and the results showed that its daily voltage regulation cost and network loss were 863.4 yuan and 43.73 MW h, respectively. The average absolute percentage error of the prediction model was 4.51 %, which was superior to other comparative methods.

Orthotropic Material Parameters Identification Method of Stator Core and Windings in Electric Motors

The accurate equivalent orthotropic material parameters (OMPs) of stator core and windings are the basis of modal analysis in electric motors and the key to precisely predicting electromagnetic vibration and noise. A CNN (convolutional neural network)-MPGA (multi-population genetic algorithm) based OMPs identification method of stator core and windings for electric motors is proposed in this study. Firstly, the surrogate models between the material parameters and eigenmode frequencies are trained with CNN for stator core and windings. Then, MPGA is utilized to identify the equivalent OMPs of stator core and windings combined with tested eigenmode frequencies. Finally, two different classes of electric machines are selected to verify the method. The CNN-MPGA based method in this study can quickly and accurately identify OMPs of stator core and windings. It is significant for the calculation and suppression of electromagnetic noise for electric motors.

Unmanned surface vehicles (USVs) scheduling method by a bi-level mission planning and path control

To enhance the operational efficiency of container terminals, we propose a bi-level scheduling method that employs unmanned surface vehicles (USVs) to transport containers exclusively between different berths on the waterside of terminals. In the upper level, Considering time windows, berth coordination, energy replenishment, etc., we formulate the mission decision model to obtain a cost-effective USVs shipping solution by minimizing unplanned delay and total execution cost. In the lower level, considering path distance, path smoothness, and motion constraints, we develop the USV path control model to achieve the USV path planning and reduce path tracking errors. Finally, we integrate the advantages of the chaotic Electron Search and multi-population Genetic Algorithm (ES-mGA) to solve the bi-level USVs scheduling model. Experimental results demonstrate the effectiveness of our approach and algorithm in guiding the USVs to efficiently accomplish container transshipment and acquire stable USV sailing states.

A novel flow channel design to achieve high temperature homogenization in solid oxide fuel cell

The thermal stress, which has a significant impact on the stable and prolonged operation of Solid Oxide Fuel Cells (SOFCs), can be further aggravated by the presence of local hot spots. Improving the distribution uniformity of physical quantities is of great significance to improving the internal temperature distribution of the SOFC. In this study, to improve the uniformity of temperature distribution, the thermal and flow distribution of three kinds of flow fields are investigated including Z, U, and L flow fields. The results show that the temperature distribution is dominated by the flow rate distribution. To optimize the physical quantity distribution of the designed flow channel, a novel, and easy-to-manufacture thermal management is proposed. This method is to adopt the extended ridge for controlling the flow rate into each flow channel. Moreover, a multi-population genetic algorithm is adopted to optimize the length of each extended ridge. The optimized results show that the temperature deviation in the Z, U, and L flow fields are reduced by about 64.20 %, 26.06 %, and 23.45 % respectively. This novel thermal management method enables more uniform temperature distribution and it is expected to be used in large-scale manufacturing due to its simple but subtle design.

Machine learning-driven optimization design of hydrogel-based negative hydration expansion metamaterials

Hydrogel-based negative hydration expansion (NHE) metamaterials are composite structures composed of responsive hydrogels and polymers, and their properties depend on their unique structures. In this paper, an optimization method based on the combination of the back-propagation neural network (BPNN) and the multi-population genetic algorithm (MPGA) is developed to rapidly design isotropic and anisotropic hydrogel-based metamaterials with specific NHE effects. In this method, several dimensionless design parameters are introduced to describe the structural characteristics of the metamaterial. The initial dataset is constructed based on the finite element method simulation results, and the mapping relationship between the design parameters and the equivalent linear strain is constructed by the BPNN, and the metamaterial with specific effect is efficiently optimized by combining the MPGA. The method is proved to have high accuracy and efficiency, and is applied to design many novel 2D and 3D metamaterials. The 3D metamaterial designed by this method has an ultra-large NHE ratio about 82 %. Compared with the topology optimization method, this method can significantly reduce the amount of computation, and can effectively avoid falling into the local optimum. The results show that the optimization method based on machine learning is an efficient means to design hydrogel-based metamaterials.

Multi-population genetic algorithm with greedy job insertion inter-factory neighbourhoods for multi-objective distributed hybrid flow-shop scheduling with unrelated-parallel machines considering tardiness

Distributed manufacturing is gradually becoming the future trend. The fierce market competition makes manufacturing companies focus on productivity and product delivery. The hybrid flow shop scheduling problem (HFSP) is common in manufacturing. Considering the difference of machines at the same stage, the multi-objective distributed hybrid flow shop scheduling problem with unrelated parallel machines (MODHFSP-UPM) is studied with minimum makespan and total tardiness. An improved multi-population genetic algorithm (IMPGA) is proposed for MODHFSP-UPM. The neighbourhood structure is essential for meta-heuristic-based solving algorithms. The greedy job insertion inter-factory neighbourhoods and corresponding move evaluation method are designed to ensure the efficiency of local search. To enhance the optimisation ability and stability of IMPGA, sub-regional coevolution among multiple populations and re-initialisation procedure based on probability sampling are designed, respectively. In computational experiments, 120 instances (including the same proportion of medium and large-scale problems) are randomly generated. The IMPGA performs best in all indicators (spread, generational distance, and inverted generational distance), significantly outperforming existing efficient algorithms for MODHFSP-UPM. Finally, the proposed method effectively solves a polyester film manufacturing case, reducing the makespan and total tardiness by 40% and 60%, respectively.

General modeling and energy management optimization for the fuel cell electric tractor with mechanical shunt type

The fuel cell electric tractor has emerged as a new development direction in modern agriculture due to its energy-controllable and emission-reducing advantages. This paper focuses on improving the universality of basic research by studying the fuel cell electric tractor with mechanical shunt type (FCETMST). The bond graph model has the benefits of cross-physical system integration. To fill the gap in general modeling of electric tractors, an electromechanical hydraulic integrated machine model by utilizing the bond graph model, which includes energy system, drive system, lifting system and power take-off (PTO) system, is developed. The general power flow model of the electric tractor with mechanical shunt type is also derived. Lastly, the accuracy of the model is validated by analyzing the tractor’s traction characteristics. To further enhance energy efficiency, this paper proposes an energy management strategy (EMS) utilizing a multi-population genetic algorithm (MPGA) and compares it with a power following control strategy (PFCS) for a fuel cell/lithium battery composite energy system. The results base on the co-simulation by using AVL-Cruise and Matlab/Simulink to build a vehicle simulation model and EMS models, respectively, show that under ploughing and transportation conditions, compared with the traditional PFCS, using MPGA leads to reduce state of charge (SOC) fluctuations and the remaining SOC is 11.22 % and 6.1 % higher, respectively. The theoretical hydrogen fuel consumption is decreased by about 26.49 % and 36.21 %, respectively. This study can provide theoretical support for the modeling and energy management optimization of the electric tractor system.

Retracted: Application of Multipopulation Genetic Algorithm in Industrial Special Clothing Design

Retracted: Application of Multipopulation Genetic Algorithm in Industrial Special Clothing Design

Path Planning for the Rapid Reconfiguration of a Multi-Robot Formation Using an Integrated Algorithm

Path planning is crucial in the scheduling and motion planning of multiple robots. However, solving multi-robot path-planning problems efficiently and quickly is challenging due to their high complexity and long computational time, especially when dealing with many robots. This paper presents a unified mathematical model and algorithm for the path planning of multiple robots moving from one formation to another in an area with obstacles. The problem was initially simplified by constructing a cost matrix, and then the route planning was achieved by integrating an elite enhanced multi-population genetic algorithm and an ant colony algorithm. The performance of the proposed planning method was verified through numerical simulations in various scenarios. The findings indicate that this method exhibits high computational efficiency and yields a minimal overall path distance when addressing the path-planning problem of a multi-robot formation reconstruction. As a result, it holds promising potential for the path-planning problem of a multi-robot formation reconstruction.

Research on Multi-AGV Task Allocation in Train Unit Maintenance Workshop

In the context of the continuous development and maturity of intelligent manufacturing and intelligent logistics, it has been observed that the majority of vehicle maintenance in EMU trains still relies on traditional methods, which are characterized by excessive manual intervention and low efficiency. To address these deficiencies, the present study proposes the integration of Automatic Guided Vehicles (AGVs) to improve the traditional maintenance processes, thereby enhancing the efficiency and quality of vehicle maintenance. Specifically, this research focuses on the scenario of the maintenance workshop in EMU trains and investigates the task allocation problem for multiple AGVs. Taking into consideration factors such as the maximum load capacity of AGVs, remaining battery power, and task execution time, a mathematical model is formulated with the objective of minimizing the total distance and time required to complete all tasks. A multi-population genetic algorithm is designed to solve the model. The effectiveness of the proposed model and algorithm is validated through simulation experiments, considering both small-scale and large-scale scenarios. The results indicate that the multi-population genetic algorithm outperforms the particle swarm algorithm and the genetic algorithm in terms of stability, optimization performance, and convergence. This research provides scientific guidance and practical insights for enterprises adopting task allocation strategies using multiple AGVs.