Search

Aiming at the problems that traditional genetic algorithm is prone to fall into local optimal solution, parameters cannot be adjusted intelligently, and local search ability is poor when solving flexible job-shop scheduling problems, a flexible job-shop scheduling model with the goal of minimizing the maximum completion time was established. A reinforcement learning improved genetic algorithm (RLIGA) based on reinforcement learning was proposed to solve the model. Firstly, in the iterative process of genetic algorithm, reinforcement learning was used to dynamically adjust key parameters. Secondly, the discrete Lévy flight mechanism based on process coding distance was introduced to improve the solution space. Finally, the variable neighborhood search mechanism was introduced to improve the local development ability of the algorithm. PyCharm was used to run Brandimarte examples to verify the solving performance of the proposed algorithm. The experiment proves that the proposed algorithm has higher solving efficiency, stronger ability to jump out of the local optimal, and better solving results. [ABSTRACT FROM AUTHOR]

Due to the uncertainty and information delay of disaster situations, emergency relief material distribution and vehicle scheduling often face complex environments where demand point information dynamically changes during the distribution process. It is particularly important to quickly respond and accurately and efficiently achieve emergency material rescheduling in a dynamically changing environment. A dynamic emergency logistics model was established for emergency supplies in a complex environment with dynamic changes in demand point information, and a dynamic optimization algorithm was proposed based on cluster-distribution strategy. Corresponding route adjustment strategies were designed to ensure the feasibility of the plan. Finally, genetic algorithm was used to solve the problem. The numerical experimental results demonstrate the effectiveness of the proposed method. [ABSTRACT FROM AUTHOR]

In order to improve the efficiency of logistics distribution in urban environment, taking delivery time and delivery cost as optimization objectives, a mathematical model of “vehicle-drone”route optimization was established, and a third-order algorithm based on clustering-Floyd-genetic algorithm was proposed. The experimental results show that the algorithm can effectively reduce the computational load through multi-order data processing, and overcome the slow convergence rate of genetic algorithm, easy to fall into the local optimal problem. The sensitivity analysis of drone capacity shows that the delivery capability of drone increases significantly with the increase of load, and the simultaneous increase of load and maximum range can bring the delivery capability of drone into full play. Compared with the vehicle-only mode, the total cost and time of “ vehicle-drone” mode decrease by 36. 1% and 34. 5% respectively. It is proved that the algorithm has some practical value in city logistics distribution. [ABSTRACT FROM AUTHOR]

Traditional machine learning models for groundwater potential prediction do not consider the optimal combination of factors, which can adversely affect groundwater potential mapping. For this reason, a groundwater potential prediction method based on genetic algorithm optimization support vector machine was proposed. Taking Yiliang County of Yunnan Province as the study area, a total of 15 influencing factors were selected from topography, hydrology, soil, geology, etc. Considering the model performance and the roles of the influencing factors, the optimal factor combinations containing 11 influencing factors were screened using the genetic optimization algorithm. Then the groundwater potential prediction model was constructed using the support vector machine method. Finally, the model accuracies and areas under the curves of the receiver operating characteristic (AUCs) before and after the optimization of the factors were calculated, and the receiver operating characteristic (ROC) curves of the model and the groundwater potential prediction maps were plotted. The results show that the accuracy of the model before factor optimization is 0. 774, the AUC of the validation set is 0. 789, the accuracy of the model after factor optimization is 0. 777, and the AUC of the validation set is 0.806, which are increased by 0.003 and 0.017 respectively. It can be seen that the accuracy and reliability of the proposed method are superior to the traditional support vector machine method, and the results can provide scientific reference for regional hydrogeological survey and groundwater resource management and planning. [ABSTRACT FROM AUTHOR]

The prediction of subway passengers' walking time during exit is an important basis for urban traffic operation and management. Accurate prediction can help alleviate subway congestion, optimize subway services, and improve passenger satisfaction. Firstly, video analysis software was used to extract the walking time and several characteristic variables of passengers during exit from surveillance videos. Secondly, in order to screen out the factors that have a significant impact on walking time, correlation analysis and optimal scaling regression model were used for factor analysis, and genetic algorithm was used to extract the optimal feature combination. Finally, the extracted features were used as input vectors and the Extreme Gradient Boosting model was used to predict walking time, with mean absolute error as the evaluation index. The experimental results show that the method proposed in this paper has good effect in predicting the behavior of subway passengers during exit, with a mean absolute error of 1. 55 s, lower than the unoptimized Extreme Gradient Boosting model (1. 87 s), support vector machine (2. 03 s) and random forest (1. 96 s) models. [ABSTRACT FROM AUTHOR]

Air-rail intermodal freight transportation offers a promising solution to address the increasing demand for high-value and time-sensitive goods. In the transformation of transport mode, the connecting link was affected by the transport cost and transport efficiency of combined transport goods to a certain extent. Based on the external transfer connection mode of “ all-cargo aircraft + highspeed rail freight train”, a model was constructed for optimizing the connection timetable and ferry scheduling scheme in air-rail intermodal freight transportation to further optimize the air-rail intermodality scheme. To solve this model, an approach based on genetic algorithms was developed. Connection cost and connection time were further analyzed. The model and approach were tested in a realworld case of air-rail freight transportation using the timetable data of Hongqiao Airport T2 Terminal and Shanghai Hongqiao High-speed Railway Station on a specific day. The simulation results show that three ferry cars are needed to transport the goods in the air-rail connection. In this scheme, the workload of each ferry car is reasonably distributed, and the cost of each batch of goods is the smallest. The connection cost accounts for about 20% of the total cost, and the cost advantage of air-rail intermodality in long-distance transportation is more obvious. The efficiency of loading and unloading is the key factor affecting the connection time of air-rail intermodality. It can provide a guidance for the future operation of long-distance air-rail intermodality. [ABSTRACT FROM AUTHOR]

Aiming at the problems of long time and low positioning accuracy of indoor positioning algorithm, an indoor positioning algorithm based on improved LightGBM (light gradient boosting machine) algorithm was proposed. The algorithm first preprocessed the data in the fingerprint database, and removed outliers and outliers through KNN (K-nearest neighborhood ) algorithm to reduce environmental noise interference and improve data reliability. Next, the algorithm divided the sample set into training set and test set, and used LightGBM algorithm to model. At the same time, genetic algorithm was used to adjust the parameters in LightGBM algorithm, and the optimal parameters were found according to the fitness function to obtain the LightGBM + GA (genetic algorithm) coordinate prediction model. Finally, a prediction model was established according to the optimized parameters to realize coordinate prediction. The experimental results show that the algorithm improves the accuracy of WiFi positioning by 0. 1 m compared with XGBoost (extreme gradient boosting,XGBoost) algorithm and 0. 19 m compared with GBDT algorithm. In terms of positioning time, LightGBM + GA algorithm is 5. 10 s faster than GBDT (gradient boosting decision tree) algorithm and 5. 97 s faster than XGBoost algorithm, which has good practicability. [ABSTRACT FROM AUTHOR]

In view of the lack of meteorological visibility data in Anhui Province, meteorological data from four automatic meteorological stations (Mount Huangshan Station, Lingbi Station, Shannan Xigu Station, and Baize Lake Station) under different terrain conditions in Anhui Province were selected from the years 2017 to 2019. The meteorological elements closely related to visibility were first identified using the gray correlation analysis method. Subsequently, a hybrid algorithm of genetic algorithm (GA) and particle swarm optimization algorithm (PSO) was employed to optimize the prediction model of the back propagation (BP) neural network. The visibility of automatic weather stations under four different terrain conditions was predicted, and the prediction results were compared with the RF prediction model and XGBoost prediction model. The results indicate that, under any terrain condition, the GA-PSO-BP neural network prediction model exhibits smaller prediction errors and higher model accuracy. [ABSTRACT FROM AUTHOR]

In order to analyze the optimization problem of bus route departure time, two intersecting bus routes with transfers were taken as the research object. Taking into account passenger congestion, a bus timetable optimization model was constructed with the objective function of minimizing the total waiting time, transfer waiting time, and non transfer passenger waiting time. Two city bus routes in Hohhot City, namely Line 4 “JINYU Global Center to Affiliated Hospital of Medical University” section and Line 78 “Hohhot Bus to Saihan District Education Bureau” section, were studied, and the optimization model was solved using genetic algorithm. The results show that the proposed city bus timetable optimization scheme can shorten the transit waiting time and total waiting time by 18.1% and 10.7% respectively, thereby validating the proposed optimization model’s effectiveness in improving the operating efficiency of intersecting routes. [ABSTRACT FROM AUTHOR]

In order to improve the stability of power grid operation, it is necessary to build a stable and reliable power grid dispatching system, and an active-active dispatching system considering the characteristics of parallel systems was proposed. By building two data centers, active and standby. According to the application business volume, adjustments were made between the primary and backup adjustments, achieving full utilization of resources. In the event of a failure, another functioning center will take on all the business. The parallel system was regarded as the virtual operation system of the power grid, and the active-active dispatching system of the power grid was obtained after combining the real and virtual systems, with functions such as management and control module, experiment and evaluation module, and learning and training module. By initializing the data, calculating the power distribution results, establishing a two-layer task scheduling model, and using genetic algorithms to solve the model, the optimization scheduling of the dual active scheduling system can be achieved. The test results show that the method proposed has the advantages of short time, high efficiency, large total amount of grid information dispatching, strong dispatching ability and high security of dispatching information to reach the optimal distribution value. It shows that the designed active-active dispatching system is feasible and beneficial to the optimal dispatching of regional power grid. [ABSTRACT FROM AUTHOR]

In order to improve network routing performance, a QoS (quality of service) multicast routing algorithm based on genetic ant colony optimization algorithm was proposed and designed. Firstly, an adaptive frequency conversion acquisition strategy was designed to collect network and node information, in order to obtain the status of the network and nodes and provide data support for subsequent routing optimization. Secondly, the path cost was calculated, the minimum path cost as the optimization objective was set, a QoS multicast routing optimization model was established, and relevant constraints was set. Finally, a genetic ant colony optimization algorithm was proposed by combining genetic algorithm and ant colony algorithm to solve the above model, output the optimal path, and complete routing optimization. The experimental results show that the proposed algorithm can effectively reduce path length and cost, improve search efficiency and routing request success rate, and optimize the routing delay jitter [ABSTRACT FROM AUTHOR]

A kernel based extreme learning machine (KELM) method was proposed for predicting joint moments of lower limb rehabilitation robots to address the problem that random initialization of input weights and bias affect the accuracy of the model when predicting lower limb joint moments by extreme learning machine (ELM). The method integrated Gaussian kernel function with ELM and used genetic algorithm ( GA) combined with particle swarm optimization ( PSO) of genetic algorithm-particle swarm optimization (GAPSO) to optimize the parameters of KELM. Firstly, the motion data of a patient with right lower limb hemiplegia walking on a treadmill at 5 different speeds of 0. 4, 0. 5, 0. 6, 0. 7 and 0. 8 m/ s were collected and preprocessed. Secondly, the KELM was optimized by GAPSO to obtain the optimal regularization coefficient C and kernel function width parameter S. The output joint moments were compared with the joints calculated by inverse biomechanical analysis. Finally, the root mean square (RMSE) and correlation coefficient (P) were used to evaluate the superiority of the algorithm. The experimental results show that the average maximum root mean square error of KELM algorithm based on GAPSO(GAPSO-KELM) optimization is 14%, 18% and 28% lower than that of PSO-KELM algorithm, KELM algorithm and ELM algorithm, respectively, and the minimum P is 0. 84 except for 0. 8 m/ s right ankle inversion, which is 0. 79. The GAPSO-KELM algorithm further improves the prediction accuracy, making it a more effective algorithm support for rehabilitation treatment. [ABSTRACT FROM AUTHOR]

The salt cavern gas storage facility, as one of the crucial means for natural gas peak shaving in our country, is tasked with a strategic reserve role. In order to enhance the operational optimization of the Sichuan-to-East transmission pipeline network in this region, suitable gas injection and production schemes were proposed for Jintan salt cavern gas storage facility, with the primary objective at this stage being the reduction of compressor energy consumption. Following an analysis of various factors including geographical environment, pipeline parameters, gas storage well parameters, and compressor energy consumption of Jintan salt cavern gas storage facility, the injection process of ten wells in the region is taken as an example. The energy consumption analysis of the distribution scheme was established as the main goal. While adhering to constraints on the total gas storage capacity and individual well injection, a genetic algorithm was employed to optimize the injection task scheme. After evaluating the optimization process using five sets of genetic algorithm iterations, the results indicate that convergence of total energy consumption is achieved at around 25 iterations, with an average runtime of 4. 57 seconds per iteration. Further analysis reveals the trend of gas injection volumes for each individual well during the optimization process. In compliance with on-site conditions and injection requirements, the final simulated and optimized production scheme demonstrates a 33% reduction in compressor energy consumption compared to the original scheme. The actual on-site reduction in compressor energy consumption generally falls between 30% and 35%, significantly mitigating energy wastage in compressor operation. This holds substantial significance for practical production applications. [ABSTRACT FROM AUTHOR]

Repetitive projects are usually interfered by uncertainties, and the most important challenge for project managers is how to quickly repair the interfered plan to the baseline scheduling at minimum cost. The reactive scheduling model and algorithm for repetitive projects were studied. Firstly, a new model of repairing strategy was proposed to quickly repair the plan to the baseline schedule at a minimum cost. Then a hybrid algorithm combining Q-learning and genetic algorithm was designed to solve the optimization problem. Finally, a highway project and Monte Carlo simulation were used to illustrate the effectiveness of the model and algorithm. Results show that the repairing strategy proposed can reduce reactive scheduling costs, and increasing the extent of repairing can effectively reduce the reactive scheduling costs. Moreover, the proposed hybrid algorithm is superior to the genetic algorithm in terms of solution quality and efficiency for this problem. It can help project managers to deal with reactive scheduling problems of repetitive projects. [ABSTRACT FROM AUTHOR]

Spare parts are the material basis for equipment maintenance in nuclear power plants. However, holding excessive spare parts inventory will increase the operating cost of nuclear power plants. By setting reasonable spare parts inventory parameters, a reasonable balance between supply guarantee and inventory control can be found. To solve the problem of spare parts inventory parameter setting, genetic algorithm was introduced into the field of spare parts inventory parameter optimization. By establishing the evaluation method to evaluate the quality of spare parts inventory parameters, the setting of spare parts inventory parameters was transformed into a multivariable optimization problem, and the genetic algorithm was used to calculate the optimal inventory parameters. The spare parts of nuclear power plants were selected for simulation testing. When the spare parts are added with multiple consumption fluctuations, the spare parts inventory parameters calculated by genetic algorithm are better than those in the current database. The results show that genetic algorithm has important practical value in the field of spare parts inventory parameter optimization. [ABSTRACT FROM AUTHOR]

In view of the phenomenon of waste of logistics volume caused by unreasonable workshop layout, a study was conducted on forging plants, and a single objective optimization function for minimizing logistics volume was established. The original layout was optimized using the systematic layout planning( SLP) method to obtain three layout plans. A combined weighting model was constructed using the analytic hierarchy process (AHP) and entropy weight method, Then, the technique for order preference by similarity to an ideal solution ( TOPSIS) was used to rank and optimize the three options, forming a multi indicator decision evaluation model. Finally, the genetic algorithm was combined with the harmony search algorithm to perform secondary optimization on the layout of the forging plant workshop and obtain the optimal layout plan. The case study shows that this research method reduces the flow of goods by 28. 82%, with significant improvement effects. At the same time, it also verifies the superiority of the model and provides certain reference value for workshop layout design in related industries. [ABSTRACT FROM AUTHOR]

In order to improve the measurement accuracy of green view index, find out the factors affecting green view index and the measures to improve road green vision rate, the main road of Zibo City was taken as the research object, and the green view index was taken as the evaluation index of road greening effect. For the first time, genetic algorithm was used to screen clear and unshielded street view pictures, MATLAB software was used to calculate the green view index and ArcMap software was used to quantify the effect of road greening in Zibo City. The results show that the average green view index in Zibo City reaches 20. 483%. The sampling points of low level and general level are mainly caused by wide pavement width and unreasonable collocation of green patterns. The relationship table between afforestation pattern and road type is established. It is found that the open forest shrub pattern significantly improves the green view index and is more suitable for urban road greening. The comparison of green rate of lane number is: two-way two-lane > two-way three-lane > two-way single-lane > two-way four-lane. A green belt separating non-motor lane and motor lane is set on the road. The green belt adopts the type of sparse forest and shrub, which will significantly improve the green view index. [ABSTRACT FROM AUTHOR]

Aiming at the characteristics of the high cost of hazardous chemicals logistics control, the difficulty in detecting potential safety hazards and many uncertain factors, an overall framework of a hazardous chemicals logistics distribution system driven by twin data was established based on digital twin technology, which can realize the real-time monitoring of parameters such as fatigue status of drivers and fault warning of vehicles. Therefore, accurate simulation and iterative optimization of the physical space of hazardous chemical vehicle scheduling could be realized based on physical parameters. Besides, a path optimization model with time window constraints for a single distribution center was constructed. The minimum total cost of distribution was taken as the objective function and the customer service time was dynamically adjusted by combining twin data, and the mathematical model was solved by using MATLAB tools with the genetic algorithm kernel. The fluctuations in the transportation cost of hazardous chemicals caused by dynamic unsafe factors in the logistics scheduling process can be effectively solved by adopting this method. [ABSTRACT FROM AUTHOR]

Aiming at the problems of limited search area and inadequate evaluation methods for automatic highway alignment selection junction, based on genetic algorithm, the highway alignment constraint model, automatic highway alignment optimization model and operation speed prediction model were constructed to realize automatic highway three-dimensional spatial alignment solution search and safety evaluation. The results show that the automatic alignment selection and safety evaluation method based on genetic algorithm can overcome the deficiency that alignment junction can only be searched for in a given section under the given design area, proposed technical, economic and safety conditions, and achieve the automatic convergence and optimization of the design scheme. Under the condition of alignment safety, the alignment junction can be guaranteed to be searched automatically in the proposed alignment selection area globally, which verifies the feasibility of the automatic highway alignment selection method based on genetic algorithm and improves the efficiency of the alignment selection work. The method can realize the evaluation of road alignment safety with speed consistency as the index, and improve the safety design level of the alignment selection scheme. The study can provide reference for improving the theoretical basis of automatic road alignment selection and continuing the research on automatic road alignment selection methods. [ABSTRACT FROM AUTHOR]

Aiming at the output power fluctuation and unbalanced impeller load caused by wind shear effect in the high wind speed region of large wind turbines, an independent pitch control strategy based on the power and load dual-objective synergy of power and load based on the weight coefficient method was proposed for NREL's 5 MW land-based wind turbine from both power and load dimensions. First, the PID(proportional integral derivative) optimized by genetic algorithm was used as the main controller to get the uniform variable pitch angle through the wind turbine speed feedback. Then, the dynamic weight coefficient of each blade was obtained according to the azimuth angle feedback to adjust the pitch angle of each blade. Finally, the blade root waving moment was fed to the fuzzy PID controller to get the pitch angle fine-tuning under the influence of load, and the pitch angle was corrected. The proposed control strategy was compared and analyzed with the unified pitch control and independent pitch PI control through the joint simulation of FAST and MATLAB/ Simulink platforms. The results show that the proposed control strategy can effectively stabilize the output power near the rated power and has a significant effect on reducing the unbalanced load of the wind turbine. [ABSTRACT FROM AUTHOR]

The sensitivity of parameters to the maximum takeoff weight within the optional ranges of V1 / VR and V2 / VS was analyzed, and the improvement effect of V1 / VR and V2 / VS on the maximum takeoff weight within the optimal range under typical operating conditions was discussed to enhance the competitiveness of domestic civil aircraft and further improve the performance of the ARJ 21 aircraft. Wherein, V1 is the takeoff decision speed, V2 is the takeoff safety speed, VR is the rotation speed, and VS is the reference stalling speed. The feasibility and effectiveness of implementing speed full range optimization using the golden section method and genetic algorithm method were demonstrated, and the optimization error and efficiency of these two methods were compared. It was found through research that a weight limit lifting effect of 1. 0% to 5. 5% can be achieved by optimization under field length or climb gradient limitations. At the same time, the time required to construct a conventional takeoff analysis table is better with genetic algorithm method when the genetic evolution algebra is 5, 10 and 20, with 102, 187 and 357 min, respectively, compared to the golden section method. The results demonstrate that the takeoff commercial load capacity of ARJ 21 aircraft can be effectively improved through the implementation of global optimization of speed, and the efficiency of optimization can be significantly improved by the reasonable selection of optimization algorithm. [ABSTRACT FROM AUTHOR]

To solve the problem of high operation cost of mobile charging platform, the mixed integer non-linear programming model with the minimum sum of vehicle start-up cost, driving cost and penalty cost of violating the time window of emergency service points was established, considering the constraints of emergency service points time window, flow balance and charging vehicles mileage limitation, the mobile charging vehicles path optimization on half-open multi depots was used to investigate by branch and bound method and genetic algorithm. The results show that mobile charging vehicles path optimization problem on half-open multi depots is solved 10 times, and the average deviation gap between the solution obtained in each run and the optimal solution in 10 times is 1. 22%, which illustrates the genetic algorithm has good stability. Compared with the independent service mode of single depot, the half-open multi depots joint service can provide more routing choices for mobile charging vehicles, reduce the mileage of charging vehicles and the operating cost of the platform. Compared with the centralized distribution of customer points, the cost saving effect of mobile charging platform is more significant under the random distribution and mixed distribution of customer points. [ABSTRACT FROM AUTHOR]

In order to reduce the increasing adverse impacts on the life and traffic of residents in the region during the implementation of traffic microcycle, minimizing road connectivity, minimizing travel time of slow traffic, minimizing environmental pollution and minimizing road saturation were taken as optimization objectives. The constraints of arterial road saturation, average speed limit of slow traffic and density of slow road network were considered. A two-layer programming model of multi-objective regional traffic path optimization was established. Genetic algorithm was used to solve the problem, roulette operator and non-uniform mutation operator were used to operate, and the optimal organization of microcirculation road network was obtained. Through the analysis of the example, compared with the road network before the implementation of microcirculation, the road network saturation decreases obviously after the implementation of microcirculation, which reduces the impact on residents and the destruction of ecological environment in the street area, and a reasonable street traffic microcirculation network is obtained. [ABSTRACT FROM AUTHOR]

When using an intelligent optimization algorithm to identify structural damage, the actual damage location is usually assumed to be consistent with the division form of finite element model elements. In order to study the influence on the recognition effect when there is a difference between them, a new damage identification method was proposed. The selection, crossover, and mutation operators of the traditional genetic algorithm were improved, and the catastrophe and neighborhood search mechanism was introduced. Through a project case, several independent repeated calculations were carried out for different damage conditions, and the stability and convergence speed of the algorithm was analyzed. In addition, the effects of noise, the number of measuring points, and train speed on the recognition results were also discussed. The results show that even if the damage location is different from the element division form of the finite element model, the proposed method can locate the damage location accurately, but it will affect the quantitative results to a certain extent, and the convergence speed of this method is fast. The computational efficiency is higher than that of the traditional primary identification method. [ABSTRACT FROM AUTHOR]

Restricted by terrain, there are some areas in mountainous areas that are very close in space, but the actual route needs to be detoured farther. In order to improve the convenience of traffic between regions, the coordinated construction of the tunnel and the winding mountain highway was studied from the overall road network planning level. Firstly, based on the original road network, according to the shortest travel distance strategy, a network model of the road network in mountainous areas was established to measure the travel convenience of residents under the original road network. On this basis, with the limit of investment quota as the constraint, and the goal of maximizing the shortening of the average annual travel distance of residents under the unit investment cost, a road network optimization model in mountainous areas considering tunnels and winding highways was established, and the solution method of the model was designed based on genetic algorithm. Finally, taking the road network in mountainous areas in the middle of Zhifu District of Yantai as an example, the road network optimization scheme under the coordinated construction of tunnels and winding mountain highways was obtained, and the reduction ratio of each travel path length under this scheme was analyzed. Through sensitivity analysis, the impact of investment quota on investment efficiency was studied and provided a reference for the competent authorities to determine a reasonable investment quota [ABSTRACT FROM AUTHOR]

In order to meet the diversified travel demand of passengers and plan the tourism transportation network, this paper puts forward the perspective of passenger intermodal transport, a new road and short-haul general aviation transportation intermodal (road-air intermodal) mode was proposed. A bilevel programming model of road-air intermodal tourism transportation network was constructed. The upper layer mode aims at the shortest total travel time of passengers and total operating cost of the traffic network to designed the traffic route scheme. The lower layer model aimed at the shortest travel time of individual passenger and allocates passenger flow to the route. The proposed model was solved by genetic algorithm, and the example analysis shows that the road-air intermodal transportation network model is scientific, reasonable and feasible. [ABSTRACT FROM AUTHOR]

The current “first come, first serve” airport deicing vehicle scheduling method is inefficient. Taking minimizing the total distance of the deicing vehicle and the total waiting time of the aircraft as the objective function, the cooperative scheduling model of the stand deicing vehicle and the aircraft was constructed, and an improved genetic algorithm was proposed to solve the model. The simulation experiment was carried out with 142 flight data of Xian Airport, and compared with random scheduling algorithm and greedy algorithm. The results show that compared with random scheduling algorithm and greedy algorithm, the improved genetic algorithm saves 15. 23% and 7. 81% of the total travel distance respectively, and the aircraft waiting time for deicing is greatly reduced. The superiority of the proposed algorithm in guiding the operation of deicing truck is proved. [ABSTRACT FROM AUTHOR]

A map weighted genetic algorithm was proposed to solve the problem of high equipment failure rate and increased oil production cost due to the high contact rate between oil well inspection robot and obstacles. Firstly, the map was rasterized, the raster map model was established, and the weighted setting was carried out. Secondly, genetic algorithm model was introduced to carry out path planning, and the result of each path planning was saved into cells and the path length was calculated. Finally, cells with the shortest path in the maximum weight value were selected and the path was drawn. Set in the parameter under the same conditions, using the genetic algorithm based on weighted map, compare experiment, the classical genetic algorithm simulation results show that the genetic algorithm based on weighted map priorities is not close to the obstacles of grids completed the path planning tasks, robot and obstacles of exposure rate dropped by 74. 91%, Time and distance increased by only 0. 317 9 s and 32% . [ABSTRACT FROM AUTHOR]

The correct classification of network data plays an important role in the monitoring and maintenance of network environment. It is particularly important to solve the problem of data classification and deal with complex feature relations under the condition of data imbalance. Therefore, a machine learning classification method based on improved SMOTE (synthetic minority over-sampling technique) +GA-XGBoost (genetic algorithm-extreme gradient boosting) was proposed. The local outlier factor was introduced into the SMOTE interpolation process to over sample a few samples and under sample a lot of samples at random, so as to achieve sample rebalancing.At the same time, in order to increase the model fitting degree in the model training process, the genetic algorithm with the advantage of evolutionary iteration was combined with XGBoost to solve the problems of many XGBoost parameters and slow convergence of feature learning. UNSW＿NB15 was used in the experiment dataset, machine learning algorithms such as multi-layer perceptron, K-nearest neighbor, decision tree and unbalanced data training methods such as SMOTE+XGBoost were selected for experimental comparison. The results show that the method has good classification prediction accuracy (97.40%), high average recall rate (70.2%) and average F1 score (68.8%). The data collected on the industrial information security platform of our laboratory are tested, and the classification accuracy is 99%, which further verifies the effectiveness and feasibility of this method. [ABSTRACT FROM AUTHOR]

At present, most of the oilfield centrifugal pump injection stations use traditional manual inspection and other methods for anomaly detection, which wastes a lot of resources and has low detection accuracy. To address this problem, a genetic algorithm optimized one-class support vector machine (GA-OC-SVM) based anomaly detection method was proposed for water injection stations. Firstly, the data of water injection station were standardized, normalized and feature extracted. Secondly, genetic algorithm was used to find the best individual value of the population as a parameter of single class support vector machine, and the detection model was established. Finally, GA-OC-SVM algorithm was compared with isolated forest algorithm, local outlier factor algorithm and other mainstream methods for anomaly detection of test data sets, and the accuracy of the algorithm was analyzed. The receiver operating characteristics (ROC) curve was used for model evaluation. The results show that the proposed GA-OC-SVM algorithm is superior, with a detection accuracy of 99%, and can save a lot of human and material resources at the same time. [ABSTRACT FROM AUTHOR]

In order to predict the drilling rate of a certain oil field in the east accurately, based on the analysis of factors affecting the drilling rate of a well in the oil field, the rate of penetration (ROP) of PDC bits with different diameters was classified according to the field experience. The penetration prediction model was set up using the random forest algorithm, K-nearest Neighbour algorithm, support vector machine (SVM) algorithm, and the model parameters were optimized using the genetic algorithm, the classification and prediction method for ROP was obtained to meet the requirements of construction design and field operation. The results show that the accuracy of prediction for ROP by the random forest model optimized by genetic algorithm is 82.1%, which is significantly higher than the K-nearest neighbor algorithm and support vector machine algorithm. This method can guide the optimization of drilling parameters in this block to improve drilling efficiency. [ABSTRACT FROM AUTHOR]

In order to optimize the aircraft load balance problem encountered by airlines, a multi-objective load balance model with the largest business load and the smallest shift in the center of gravity was established, taking into account the limit of single space. Restrictions such as heavy weight, multi-space weight limit, and center of gravity(CG) limit were used to solve the model using an genetic algorithm. In order to speed up the algorithm's optimization ability and speed up the convergence speed, an evolutionary reversal operation operator was added on the basis of the ordinary genetic algorithm. Finally, using Boeing 757-200F as a simulation optimization example, the calculation results show that the average load capacity can reach 21 281 kg, and the average center of gravity offset is 2. 31% MAC, which verifies the validity and rationality of the model. [ABSTRACT FROM AUTHOR]

Aiming at the problem of leakage detection in landfill, a method of using infrared inspection robot to inspect the bare film of landfill instead of manual inspection is proposed, and the path of the robot is planned. Firstly, according to the landfill environment, the rectangular decomposition method is used to decompose it into several sub regions; Then, the template model method is used to specify the way for the robot to traverse each sub region; Finally, based on genetic algorithm, the process algorithm is improved and optimized, including selection, crossover and mutation, so as to realize the conversion and connection between sub regions, and then realize the full coverage optimal path planning. Based on MATLAB simulation analysis, the results show that a mature and complete full coverage path planning model is formed by combining the improved genetic algorithm with the template model method, which can effectively complete the patrol task, and has faster convergence speed and lower repeated coverage rate. [ABSTRACT FROM AUTHOR]

aking gap adjustment and positive force control are among the key issues in the design and implementation of electromechanical disc brake. The mathematical model including motor, reducer, disc spring, and thread pair was established by analyzing the disc brake of mine hoisting machinery. Genetic algorithm and fuzzy control law were used to optimize proportional integral derivative (PID) parameter selection. In order to control braking gap and positive force, PID with self-tuning function was applied to electromechanical disc brake of mine hoist. The performance of the controller was discussed by comparing the simulation and test results. The test results show that the brake gap adjustment time is 0.8 s, the rise time is 0.4 s, and the steady-state error is 4%. The positive brake pressure is linearly related to the motor voltage, and the fuzzy PID controller is able to control the positive brake pressure within the target range with good tracking performance. [ABSTRACT FROM AUTHOR]

In order to improve the traffic capacity of the road network, alleviate the congestion of the trunk road and improve the utilization rate of the branch road, the influence of non-motor vehicle flow on motor vehicle flow was studied, and the influence model was constructed to improve the road resistance function of motor vehicle flow. The improved road resistance function is used to construct a bi-level programming model of unidirectional traffic organization in microcirculation network. The objective of the upper model is to maximize the traffic efficiency of road network, minimize the mean saturation of trunk road and minimize the mean saturation of micro-circulation branch road. The lower layer model is a traffic flow allocation model based on user balanced allocation. Genetic algorithm was used to solve the model. Finally, a microcirculation system in Xisanqi area of Beijing is taken as an example. The results show that one-way traffic can effectively reduce arterial saturation, improve the utilization rate of branch roads and balance the use of road resources. When the influence of non-motor vehicle flow is not considered, there are 9 sections for one-way traffic organization; when the influence of non-motor vehicle flow is considered, there are 4 sections for one-way traffic organization, and only two sections overlap. In both cases, there are obvious differences in road network organization schemes, so the influence of non-motor vehicle flow should be considered in the actual network optimization. Therefore, the microcirculation network organization optimization model considering the influence of non-motor vehicles can provide theoretical support for the actual one-way traffic organization. [ABSTRACT FROM AUTHOR]

Optimizing the vulnerable units in the traffic network is the premise to ensure the stable operation of the road network. Therefore, the improved evaluation index which could comprehensively reflect the vulnerability of road network from many aspects was selected, the relationship between the vulnerability index and the number of road sections and traffic volume was analyzed, and the bilevel programming model was used to describe the discrete traffic network design considering vulnerability. The upper model took the minimization of road network vulnerability as the optimization goal, and the lower model was the user equilibrium distribution model, by designing genetic algorithm and combining Frank-Wolfe algorithm. The example results show that when the investment amount is 3 million yuan, compared with the scheme taking the vulnerability of a single road section as the optimization objective, the optimization scheme proposed from the perspective of the whole road network can reduce the vulnerability of 13 road sections to the ideal value, while the former is only 7. At this time, the scheme can reduce the vulnerability of the road network by 38.5%, which is 9.1% more than the former; It is also found that the greater the origin-destination (OD) demand, the more construction funds need to be invested to reduce the road network vulnerability to the ideal value. [ABSTRACT FROM AUTHOR]

In order to prevent the local optimal constraints of the fruit fly optimization algorithm and improve the accuracy of muscle fatigue classification, a muscle fatigue classification method on EMG signals was proposed. Fruit fly-genetic optimization algorithm, which realized the accurate detection and classification of muscle fatigue. On the basis of the improved fruit fly optimization algorithm, the step cross and mutation of the genetic algorithm and the fruit fly optimization algorithm concentration were mixed, and combined with the neural network to identify muscle fatigue. Compared with the fruit fly optimization algorithm, the improved algorithm had a stronger ability to jump out of the local optimum and maintain vitality. Combined with the neural network, it had a better effect on fatigue state recognition. The muscle fatigue classification method proposed could prevent muscle damage caused by athletes’ fatigue and achieve accurate muscle fatigue detection and classification. A total of 10 healthy young participants（6 men and 4 women） were recruited for fatigue testing. First, the fatigue level was divided according to the subjective evaluation method. Then, the collected EMG signal data was preprocessed and features were extracted as the input of neural network, the genetic algorithm-neural network, the fruit fly optimization algorithm-neural network, fruit fly-genetic algorithm-neural network. By comparison, the accuracy rate of fruit fly-genetic optimization algorithm-neural network was 94.3%, which is better than other methods. [ABSTRACT FROM AUTHOR]

Systematic layout planning (SLP) method is widely used in workshop layout design, and genetic algorithm (GA) is often used to select satisfactory solutions. In order to reduce the influence of randomness of roulette model in GA, the entropy weight-TOPSIS (ET) evaluation model was combined to recalculate selection operators of GA, thus ET-GA model was formed to improve SLP method. Finally, the improved SLP method was used for layout design of a workshop. Compared with the original design and GA-based SLP method, the improved SLP method based on ET-GA model saves material handling cost by 12. 50% and 5. 88% respectively, increases non-logistics relationship by 3. 72% and 2. 50% respectively, and reduces material handling time by 21. 02% and 14. 75% respectively. [ABSTRACT FROM AUTHOR]