Charging scheduling method of engineering vehicle, electronic equipment and storage medium

By obtaining the charging and scheduling conditions of engineering vehicles, and using prediction models and genetic algorithms to optimize charging strategies, the problems of low charging and scheduling efficiency and insufficient intelligent management in the existing technology are solved, and smarter and more efficient charging and scheduling are achieved, and the project is improved. Efficiency and resource utilization.

CN119090180APending Publication Date: 2024-12-06ZOOMLION HEAVY INDUSTRY SCIENCE AND TECHNOLOGY CO LTD
View PDF 0 Cites 1 Cited by

Patent Information

Application Number
CN202411051865.0
Authority / Receiving Office
CN · China
Patent Type
Applications(China)
Current Assignee / Owner
Filing Date
2024-08-01
Publication Date
2024-12-06

AI Technical Summary

Technical Problem

In the prior art, the charging and scheduling of engineering vehicles mainly relies on the driver's subjective speculation, resulting in inefficient efficiency and insufficient intelligent management, and the inability to effectively meet high-intensity work needs.

Method used

By obtaining charging scheduling conditions, using pre-trained charging behavior prediction models and genetic algorithms, the charging scheduling strategy is optimized, the charging time and target charging piles are determined, and the intelligent and efficient charging scheduling is achieved.

Benefits of technology

It improves engineering efficiency and intelligent management of engineering vehicles, ensures fair distribution and responsiveness of charging resources, and reduces charging costs and time-consuming.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN119090180A_ABST
    Figure CN119090180A_ABST
Patent Text Reader

Abstract

The invention relates to an engineering vehicle charging scheduling method, electronic equipment and a storage medium, and the method comprises the steps: obtaining a charging scheduling condition which comprises current time, charging pile information, vehicle information of each engineering vehicle, and construction task data and / or charging preference data of a driver currently corresponding to each engineering vehicle; based on the charging scheduling condition and a pre-trained charging behavior prediction model, the charging behavior of each engineering vehicle is determined, and the charging behavior comprises charging time information and information of a target charging pile; and optimizing the charging behavior of each engineering vehicle by adopting a preset algorithm, and determining a charging scheduling strategy. According to the technical scheme of the invention, the charging scheduling condition related to the engineering vehicle is set, the charging behavior is predicted firstly, and then the prediction result is used as a further optimization object to determine the charging scheduling strategy, so that the charging scheduling strategy can be formulated more intelligently, efficiently and reasonably, and the improvement of the engineering efficiency and the intelligent management of the engineering vehicle are facilitated.
Need to check novelty before this filing date? Find Prior Art

Description

Technical Field

[0001] The present application relates to the technical field of charging scheduling, and in particular to a charging scheduling method, electronic equipment and storage medium for an engineering vehicle. Background Art

[0002] With the continuous development of electric vehicle technology, the application of electric engineering vehicles has also developed. For example, electric concrete machinery vehicle products (including mixer trucks, pump trucks, vehicle-mounted pumps, etc.) are increasingly used in the concrete industry.

[0003] Based on the working characteristics of engineering vehicles, they must ensure sufficient power to meet high-intensity work requirements. However, due to the complexity of the construction tasks of engineering vehicles, it currently still mainly depends on the driver's subjective speculation on whether charging is needed, which affects engineering efficiency and the intelligent management of engineering vehicles. Summary of the invention

[0004] In view of this, the purpose of the present application is to provide a charging scheduling method, electronic device and storage medium for engineering vehicles. By setting charging scheduling conditions related to engineering vehicles, charging behavior prediction is first performed, and then the prediction results are used as the object of further optimization to determine the charging scheduling strategy. The charging scheduling strategy can be formulated more intelligently, efficiently and reasonably, which is beneficial to improving engineering efficiency and intelligent management of engineering vehicles.

[0005] To achieve the above object, the present application provides a charging scheduling method for an engineering vehicle, the method comprising:

[0006] Acquiring charging scheduling conditions, wherein the charging scheduling conditions include current time, charging pile information, vehicle information of each of the engineering vehicles, construction task data and / or charging preference data of a driver currently corresponding to each of the engineering vehicles;

[0007] Based on the charging scheduling conditions and a pre-trained charging behavior prediction model, determining the charging behavior of each engineering vehicle, wherein the charging behavior includes charging time information and information of a target charging pile;

[0008] A preset algorithm is used to optimize the charging behavior of each of the engineering vehicles and determine the charging scheduling strategy.

[0009] In some embodiments, the training process of the charging behavior prediction model includes:

[0010] Acquiring historical charging data, wherein the historical charging data includes historical charging scheduling conditions and historical charging behaviors;

[0011] Processing the historical charging data to generate a training set and a verification set;

[0012] Training two or more machine learning models based on the training set;

[0013] The trained machine learning models are verified and optimized based on the verification set to select one of the machine learning models as the charging behavior prediction model.

[0014] In some embodiments, the preset algorithm is a genetic algorithm, and the use of the preset algorithm to optimize the charging behavior of each engineering vehicle to determine the charging scheduling strategy includes:

[0015] Initializing the population of the genetic algorithm according to the charging time information, so that each chromosome of each individual in the population corresponds to the charging time information of the engineering vehicle;

[0016] Perform evolution iteration on the individuals in the population according to a preset strategy to determine the individual with the highest fitness value, wherein the fitness value is determined based on the charging scheduling condition and the charging behavior corresponding to each individual;

[0017] A charging scheduling strategy is determined according to the individual with the highest fitness value.

[0018] In some embodiments, the initializing the population of the genetic algorithm according to the charging time information so that each chromosome of each individual in the population corresponds to the charging time information of the engineering vehicle, includes:

[0019] Determine the charging status of each engineering vehicle in a preset time period according to the charging time information of each engineering vehicle, wherein the charging status includes being in a charging state and not being in a charging state;

[0020] Encoding the charging status of each engineering vehicle in a preset time period to generate a chromosome corresponding to each engineering vehicle;

[0021] Individuals are generated according to the chromosomes corresponding to the engineering vehicles to obtain an initialized population.

[0022] In some embodiments, performing evolution iteration on the individuals in the population according to a preset strategy to determine the individual with the highest fitness value includes:

[0023] In the process of performing evolution iteration on the individuals in the population according to the preset strategy, the fitness value of each chromosome in the current individual is calculated according to the charging scheduling condition and the charging behavior corresponding to the current individual;

[0024] Calculating the fitness value of the current individual according to the fitness value of each chromosome;

[0025] After the iteration is finished, the individual with the highest fitness value is determined according to the fitness values ​​of all individuals in the population.

[0026] In some embodiments, calculating the fitness value of each chromosome in the current individual according to the charging scheduling condition and the charging behavior corresponding to the current individual includes:

[0027] Calculate the charging cost corresponding to each chromosome according to the current electricity price in the charging pile information, calculate the charging time corresponding to each chromosome according to the charging pile location in the target charging pile information and the charging time information, and calculate the charging urgency corresponding to each chromosome according to the current power in the vehicle information and the construction time in the construction task data;

[0028] The fitness value of each chromosome in the current individual is calculated according to the charging cost, consumption time and charging urgency.

[0029] In some embodiments, the fitness function used to calculate the fitness value of each of the chromosomes is expressed as follows:

[0030]

[0031] Among them, F represents the fitness value; C is the charging cost, which is calculated according to the current electricity price in the charging pile information; T is the consumption time, which is calculated according to the charging pile position in the information of the target charging pile and the charging time information, and the sum of the time for the engineering vehicle to travel to the charging pile, the waiting time in the charging pile queue, and the charging time; U is the urgency, which is calculated according to the current power in the vehicle information and the construction time in the construction task data, and the inverse of the sum of the current power and the current time from the construction start time is obtained, wherein the current power and the current time from the construction start time are normalized; α, β, γ are preset weight values.

[0032] In some embodiments, the step of performing evolution iteration on the individuals in the population according to a preset strategy includes:

[0033] Determine whether the current number of iterations reaches the preset number of iterations;

[0034] If the current number of iterations reaches the preset number of iterations, the evolution iteration is stopped;

[0035] If the current number of iterations does not reach the preset number of iterations, the chromosomes in the current individual are crossovered and mutated according to the fitness values ​​of each chromosome in the current individual to iteratively generate a new individual, and the charging behavior corresponding to the new individual is different from the charging behavior corresponding to the current individual.

[0036] In some embodiments, after optimizing the charging behavior of each engineering vehicle using a preset algorithm and determining a charging scheduling strategy, the method further includes at least one of the following:

[0037] Acquire charging data after implementing the charging scheduling strategy, and optimize the charging behavior prediction model according to the charging data;

[0038] Acquire operation data of charging scheduling and / or charging data after implementing the charging scheduling strategy, and adjust the parameters of the genetic algorithm according to the operation data and / or the charging data.

[0039] The present application also provides an electronic device, characterized in that it includes a processor, a memory, and a computer program stored in the memory and executable on the processor, wherein the computer program implements the steps of any of the methods described above when executed by the processor.

[0040] The present application also provides a computer-readable storage medium, on which a computer program is stored. When the computer program is executed by a processor, the steps of any of the above methods are implemented.

[0041] From the above, the charging scheduling method, electronic device and storage medium of the engineering vehicle of the present application include: obtaining charging scheduling conditions, the charging scheduling conditions include the current time, charging pile information, vehicle information of each engineering vehicle, and the construction task data and / or charging preference data of the driver currently corresponding to each engineering vehicle; based on the charging scheduling conditions and the pre-trained charging behavior prediction model, determining the charging behavior of each engineering vehicle, the charging behavior includes the charging time information and the information of the target charging pile; using a preset algorithm to optimize the charging behavior of each engineering vehicle and determine the charging scheduling strategy. The technical solution of the present application, by setting the charging scheduling conditions related to the engineering vehicle, first predicts the charging behavior, and then uses the prediction results as the object of further optimization to determine the charging scheduling strategy, can formulate the charging scheduling strategy more intelligently, efficiently and reasonably, which is beneficial to improving the engineering efficiency and the intelligent management of engineering vehicles. BRIEF DESCRIPTION OF THE DRAWINGS

[0042] In order to more clearly illustrate the technical solutions of the embodiments of the present application, the drawings required for use in the embodiments will be briefly introduced below. It should be understood that the following drawings only show certain embodiments of the present application and therefore should not be regarded as limiting the scope. For ordinary technicians in this field, other related drawings can be obtained based on these drawings without paying creative work.

[0043] Figure 1 The figure is a flow chart of a charging scheduling method for an engineering vehicle according to an embodiment.

[0044] Figure 2 The figure is a schematic diagram of a specific flow chart of a charging scheduling method for an engineering vehicle provided according to an embodiment. DETAILED DESCRIPTION

[0045] The specific embodiments of the present application will be described in detail below in conjunction with the accompanying drawings. Obviously, the described embodiments are only part of the embodiments of the present application, rather than all of the embodiments. Based on the description of the present application, all other embodiments obtained by ordinary technicians in the field without creative work are within the scope of protection of the present application.

[0046] In the description of this application, unless otherwise clearly specified and limited, the terms "set", "install", "connect", etc. should be understood in a broad sense, for example, it can be a fixed connection, a detachable connection, or an integral connection; it can be a mechanical connection or an electrical connection; it can be a direct connection or an indirect connection through an intermediate medium. For ordinary technicians in this field, the specific meanings of the above terms can be understood according to specific circumstances.

[0047] The terms "first", "second", "third", etc. are merely used to distinguish values ​​or elements of similar attributes, rather than to indicate or imply relative importance or a particular order.

[0048] The terms "comprises," "comprising," or any other variations thereof, are intended to cover a non-exclusive inclusion of the elements listed and may also include additional elements not expressly listed.

[0049] Figure 1 FIG. 1 is a flow chart of a charging scheduling method for an engineering vehicle according to an embodiment. Figure 1 As shown, the charging scheduling method for engineering vehicles of the present application includes the following steps:

[0050] S1, obtaining charging scheduling conditions, where the charging scheduling conditions include current time, charging pile information, vehicle information of each engineering vehicle, construction task data and / or charging preference data of the driver currently corresponding to each engineering vehicle;

[0051] S2, based on the charging scheduling conditions and the pre-trained charging behavior prediction model, determine the charging behavior of each engineering vehicle, the charging behavior includes the charging time information and the target charging pile information;

[0052] S3, using a preset algorithm to optimize the charging behavior of each engineering vehicle and determine the charging scheduling strategy.

[0053] In the above way, the charging behavior prediction model predicts charging behavior through the scheduling conditions related to engineering vehicles, which is suitable for industry needs. Since the model prediction results are used as the optimization object, the optimization efficiency can be improved, and then the efficiency of strategy formulation can be improved. By combining the prediction model and algorithm optimization to formulate charging scheduling strategies, it can better meet the characteristics of engineering vehicles, without the need for manual strategy formulation, and is more intelligent and reasonable. Therefore, a more intelligent, efficient and reasonable charging scheduling strategy is achieved, which is conducive to improving engineering efficiency and intelligent management of engineering vehicles.

[0054] In some embodiments, the charging pile information includes at least one of the current electricity price, charging pile location, capacity, and utilization rate. The vehicle information of the engineering vehicle includes at least one of the vehicle type (such as pump trucks, mixer trucks, etc.), battery capacity, current power, and vehicle usage frequency, wherein the current power can be expressed in terms of SOC level, where SOC is the percentage of the remaining battery power to the fully charged state power. The driver's construction task data includes at least one of the departure time, construction time, construction location, construction quantity, and construction duration. The charging preference data includes at least one of the preferred charging time and the preferred charging pile.

[0055] In some embodiments, the pre-trained charging behavior prediction model may be one of a decision tree, a random forest, a gradient boosting machine, a support vector machine, and a neural network. The preset algorithm may be one of a genetic algorithm, an ant colony algorithm, a particle swarm algorithm, and a simulated annealing algorithm. The charging time information includes the charging start time and the charging duration. The target charging pile is the charging pile selected for charging each engineering vehicle, and the information of the target charging pile includes the location of the charging pile. The charging scheduling strategy includes the charging time information corresponding to each engineering vehicle and the location of the target charging pile.

[0056] In some embodiments, the training process of the charging behavior prediction model includes:

[0057] Obtain historical charging data, including historical charging scheduling conditions and historical charging behaviors;

[0058] Process historical charging data to generate training sets and validation sets;

[0059] Training two or more machine learning models based on the training set;

[0060] The trained machine learning models are verified and optimized based on the validation set to select one of the machine learning models as the charging behavior prediction model.

[0061] The historical charging data includes historical charging scheduling conditions and historical charging behaviors. The historical charging scheduling conditions are historical data corresponding to the current charging scheduling conditions. The historical charging behaviors are obtained by collecting historical charging records of engineering vehicles, such as historical charging time, selected charging piles, charging duration, etc. The historical charging data is cleaned and preprocessed, such as filling missing values, removing abnormal data, and normalizing.

[0062] Next, based on the preprocessed data, features that are helpful in predicting charging behavior (i.e., historical charging scheduling conditions) are extracted and selected, such as time, electricity price, SOC level, vehicle usage frequency, construction time, charging pile location, etc., and features corresponding to historical charging behavior are extracted to obtain training data, which includes a training set and a validation set.

[0063] In some embodiments, two or more machine learning models are selected for training, wherein two or more machine learning models are trained based on a training set, and the trained machine learning models are verified and optimized based on a validation set. In the process of verifying and optimizing the trained machine learning models based on the validation set, performance evaluation indicators of the models are obtained, such as the accuracy, recall rate, F1 score, etc. of the models, and one of the machine learning models is selected as the charging behavior prediction model based on the performance evaluation indicators of the models. In actual implementation, a machine learning model can also be directly selected for training to obtain a charging behavior prediction model.

[0064] In some embodiments, the preset algorithm is a genetic algorithm. Step S3, using the preset algorithm to optimize the charging behavior of each engineering vehicle to determine the charging scheduling strategy, includes:

[0065] Initialize the population of the genetic algorithm according to the charging time information, so that each chromosome of each individual in the population corresponds to the charging time information of a construction vehicle;

[0066] Perform evolution iteration on the individuals in the population according to a preset strategy to determine the individual with the highest fitness value, wherein the fitness value is determined based on the charging scheduling conditions and the charging behavior corresponding to each individual;

[0067] The charging scheduling strategy is determined based on the individual with the highest fitness value.

[0068] Among them, each chromosome of each individual in the population corresponds to the charging time information of an engineering vehicle, and each individual in the population corresponds to a charging scheduling strategy (i.e., a set of charging time information) for all engineering vehicles. The initial population includes an individual, which is generated based on the charging time information predicted by the charging behavior prediction model. After the individuals in the population are evolved and iterated, the number of individuals in the population increases. At this time, the population corresponds to multiple charging scheduling strategies for all engineering vehicles.

[0069] In some embodiments, the population of the genetic algorithm is initialized according to the charging time information so that each chromosome of each individual in the population corresponds to the charging time information of an engineering vehicle, including:

[0070] Determine the charging status of each engineering vehicle in a preset time period according to the charging time information of each engineering vehicle, where the charging status includes being in a charging state and not being in a charging state;

[0071] Encode the charging status of each engineering vehicle in a preset time period to generate a chromosome corresponding to each engineering vehicle;

[0072] Individuals are generated according to the chromosomes corresponding to each engineering vehicle to obtain an initialized population.

[0073] Among them, a coding scheme is designed, such as using binary coding to represent the charging status of each engineering vehicle in a specific time period. For example, a day is divided into 24 hours, and the code represents whether the engineering vehicle is charged in each hour, that is, the chromosome can be represented as: [0, 1, 0, 0, 1, ..., 0], where 1 indicates that the vehicle is in a charging state during this time period, and 0 indicates that the vehicle is not in a charging state during this time period.

[0074] In some embodiments, the individuals in the population are iterated according to a preset strategy, including:

[0075] Determine whether the current number of iterations reaches the preset number of iterations;

[0076] If the current number of iterations reaches the preset number of iterations, the evolution iteration is stopped;

[0077] If the current number of iterations does not reach the preset number of iterations, the chromosomes in the current individual are crossovered and mutated according to the fitness values ​​of each chromosome in the current individual to iteratively generate a new individual, and the charging behavior corresponding to the new individual is different from the charging behavior corresponding to the current individual.

[0078] Among them, the parent chromosomes in the current individual can be selected by roulette or tournament, and new offspring can be generated by single-point crossover or uniform crossover, and some bits on the chromosome can be randomly changed for mutation processing to obtain new chromosomes, and then generate new individuals to increase the diversity of the population. The charging behavior of each engineering vehicle is adjusted in the iteration process of the individual. Each chromosome in the new individual corresponds to the new charging time information of an engineering vehicle. Therefore, the charging behavior corresponding to the new individual is different from the charging behavior corresponding to the current individual.

[0079] The fitness value is determined based on the charging scheduling conditions and the charging behavior corresponding to each individual. It is an indicator used to judge the quality of the individual, that is, an indicator used to judge the quality of the charging scheduling strategy. After each evolutionary iteration of the individual, the fitness value of the individual obtained by the iteration is calculated. Then, after the iteration, the individual with the highest fitness value can be determined, and then the optimal charging scheduling strategy can be determined.

[0080] In some embodiments, performing evolution iteration on individuals in a population according to a preset strategy to determine an individual with the highest fitness value includes:

[0081] In the process of evolving and iterating the individuals in the population according to the preset strategy, the fitness value of each chromosome in the current individual is calculated according to the charging scheduling conditions and the charging behavior corresponding to the current individual;

[0082] Calculate the fitness value of the current individual according to the fitness value of each chromosome;

[0083] After the iteration, the individual with the highest fitness value is determined based on the fitness values ​​of all individuals in the population.

[0084] Among them, the charging time information predicted by the charging behavior prediction model corresponds to each chromosome. For the same engineering vehicle, the charging time information matches the target charging pile. Therefore, there is also a corresponding relationship between the chromosome and the charging behavior. The charging behavior corresponding to the current individual is the set of charging behaviors corresponding to each chromosome. Therefore, the fitness value of each chromosome in the current individual can be calculated first, and then the fitness value of the current individual can be calculated based on the fitness value of each chromosome. In some embodiments, the fitness value of the current individual is the sum of the fitness values ​​of all chromosomes in the individual.

[0085] In some embodiments, according to the charging scheduling conditions and the charging behavior corresponding to the current individual, the fitness value of each chromosome in the current individual is calculated, including:

[0086] The charging cost corresponding to each chromosome is calculated according to the current electricity price in the charging pile information, the charging time corresponding to each chromosome is calculated according to the charging pile location and charging time information in the target charging pile information, and the charging urgency corresponding to each chromosome is calculated according to the current power in the vehicle information and the construction time in the construction task data;

[0087] According to the charging cost, consumption time and charging urgency, the fitness value of each chromosome in the current individual is calculated.

[0088] Among them, the design of the fitness value is combined with the engineering task and is suitable for electric engineering vehicles. Specifically, considering the urgent needs of electric engineering vehicles based on the construction time, vehicles with more urgent engineering tasks can be given priority for charging, ensuring that the vehicles can be put into use in time, improving responsiveness and efficiency, and ensuring fair distribution of charging resources. Taking into account the charging cost, low-peak charging can be achieved to reduce the charging cost. Taking into account the charging time, a charging solution that consumes less time can be selected to improve the efficiency of scheduling.

[0089] Among them, according to the charging pile location and charging time information in the target charging pile information, the time for the engineering vehicle to drive to the charging pile, the waiting time in the charging pile queue, and the sum of the charging time are calculated to obtain the charging time corresponding to each chromosome. According to the current power in the vehicle information and the construction time in the construction task data, the reciprocal of the sum of the current power and the current time from the start of the construction is calculated to obtain the charging urgency.

[0090] In some embodiments, the fitness function used to calculate the fitness value of each chromosome is expressed as follows:

[0091]

[0092] Among them, F represents the fitness value; C is the charging cost, which is calculated according to the current electricity price in the charging pile information; T is the consumption time, which is calculated by the sum of the time it takes for the engineering vehicle to travel to the charging pile, the waiting time in the charging pile queue, and the charging time according to the charging pile location and charging time information in the target charging pile information; U is the urgency, which is calculated according to the current power in the vehicle information and the construction time in the construction task data, and the inverse of the sum of the current power and the current time from the start time of construction, among which the current power and the current time from the start time of construction are normalized; α, β, γ are preset weight values.

[0093] In some embodiments, after optimizing the charging behavior of each engineering vehicle using a preset algorithm and determining the charging scheduling strategy, the method further includes at least one of the following:

[0094] Obtain charging data after implementing the charging scheduling strategy, and optimize the charging behavior prediction model based on the charging data;

[0095] The operation data of the charging scheduling and / or the charging data after the charging scheduling strategy is implemented are obtained, and the parameters of the genetic algorithm are adjusted according to the operation data and / or the charging data.

[0096] Among them, after determining the charging scheduling strategy, the charging scheduling strategy is implemented, and the charging data after the implementation of the charging scheduling strategy is obtained. These data can be used to retrain and optimize the charging behavior prediction model to further improve the accuracy of the prediction. In addition, the operating data of the charging scheduling system, such as the computing load of the operation, etc., can be monitored regularly, and the parameters of the genetic algorithm, such as the weight in the fitness function, the population size, the crossover and the mutation rate, etc., can be adjusted according to the operating data and / or the charging data. For example, when the computing load of the system is large, the population size can be reduced, and the crossover and mutation rates can be adjusted to reduce the system load.

[0097] Please refer to Figure 2 Next, the charging scheduling method of this embodiment is specifically described in combination with the steps executed by each module in the charging scheduling system for engineering vehicles.

[0098] The charging scheduling system for engineering vehicles includes data collection and preprocessing module, model training module, algorithm optimization module, charging scheduling strategy generation module, implementation and feedback module, and monitoring and adjustment module. The specific steps for each module are as follows:

[0099] 1. Data collection and preprocessing module:

[0100] Collect historical charging data, including vehicle information, charging pile information, driver information, etc. of each engineering vehicle. Vehicle information includes: type, battery capacity, historical charging records. Charging pile information, including charging pile location, capacity, usage rate, etc. Driver information, including work schedule, historical task order information, charging time preference, etc. Clean and preprocess these data, such as filling missing values, removing abnormal data, and normalizing.

[0101] 2. Model training module:

[0102] (1) Feature engineering: Extract and select features that are helpful in predicting charging behavior to form training data, including current time, electricity price, SOC level, vehicle usage frequency, construction time, charging station location, etc.

[0103] (2) Model selection: Select an appropriate machine learning model, such as random forest, gradient boosting, deep learning, etc.

[0104] (3) Model training: Use historical charging data to train the model and optimize the model performance by adjusting parameters. For example, you can train a random forest model, a gradient boosting model, etc.

[0105] (4) Model validation: Use cross-validation to evaluate the generalization ability of each model and adjust model parameters to optimize performance. By comparing the accuracy, recall rate, F1 score and other indicators of different models, the best model is selected as the charging behavior prediction model.

[0106] (5) Real-time prediction: Integrate the best machine learning model into the charging scheduling system to predict the charging behavior of each engineering vehicle in real time, so that the charging scheduling system can dynamically adjust the charging plan based on real-time data and prediction results.

[0107] 3. Algorithm optimization module:

[0108] (1) Design a coding scheme to convert the charging scheduling problem into a format that can be processed by the genetic algorithm. For example, a day is divided into 24 hours, and the code represents whether the engineering vehicle is charged at each hour. That is, a chromosome can be represented as: [0, 1, 0, 0, 1, ..., 0], where 1 indicates that the vehicle is in the charging state during this time period, and 0 indicates that the vehicle is not in the charging state during this time period.

[0109] (2) Initialize the population: The prediction model uses historical charging data and current charging scheduling conditions (including current time, electricity price, SOC level, vehicle usage frequency, construction time, etc.) to predict the charging time and charging duration of each construction vehicle. These prediction results are used to initialize the population of the genetic algorithm, that is, to generate a set of possible charging scheduling schemes. Each chromosome of each individual in the population corresponds to the charging time information of a construction vehicle. In this way, the individuals in the initial population are already obtained based on the prediction results of the prediction model, and subsequent optimization is more likely to approach the optimal solution faster.

[0110] (3) Define the fitness function and consider factors such as charging cost, charging time consumption, and charging urgency to evaluate the quality of individuals.

[0111] (4) Genetic manipulation: Use roulette wheel or tournament selection to select parent chromosomes, use single-point crossover or uniform crossover to generate new offspring, and randomly change certain bits on the chromosomes to perform mutation processing to obtain new individuals to increase the diversity of the population.

[0112] 4. Charging scheduling strategy generation module:

[0113] Run the genetic algorithm and iterate for a preset number of times. After the iteration is completed, select the individual with the highest fitness as the current charging scheduling strategy.

[0114] 5. Implementation and feedback module:

[0115] Apply the charging scheduling strategy to the charging scheduling system and collect data after implementation. This data can be used to retrain the machine learning model to further improve the accuracy of the prediction.

[0116] 6. Monitoring and adjustment module:

[0117] Regularly monitor the operation and performance of the charging scheduling system, and adjust the parameters of the genetic algorithm, such as the weight of the fitness function, population size, crossover and mutation rate, etc., based on the data and feedback after implementation.

[0118] From the above, the charging scheduling method for engineering vehicles of the present application, by setting charging scheduling conditions related to engineering vehicles, first predicts the charging behavior, and then uses the prediction results as the object of further optimization to determine the charging scheduling strategy. It can formulate the charging scheduling strategy more intelligently, efficiently and reasonably, which is beneficial to improving engineering efficiency and intelligent management of engineering vehicles.

[0119] This application effectively solves the problems of inefficiency, high cost, unfairness, insufficient response to emergency needs, and lack of intelligent management in the charging behavior of multiple engineering vehicles by combining machine learning prediction models and genetic algorithms. It can improve charging efficiency, reduce charging costs (using machine learning models to predict charging needs and combining genetic algorithms to optimize charging scheduling strategies to achieve low-peak charging), improve fairness (taking into account the driver's charging preferences and emergency needs to ensure fair allocation of charging resources), and adaptability (dynamically adjusting charging scheduling strategies based on real-time data and prediction results to adapt to changing environments and needs).

[0120] The present application also provides an electronic device, characterized in that it includes a processor, a memory, and a computer program stored in the memory and executable on the processor, wherein the computer program implements the steps of any of the methods described above when executed by the processor.

[0121] The present application also provides a computer-readable storage medium, on which a computer program is stored. When the computer program is executed by a processor, the steps of any of the above methods are implemented.

[0122] The above is only a specific implementation of the present application, but the protection scope of the present application is not limited thereto. Any changes or substitutions that can be easily thought of by a person skilled in the art within the technical scope disclosed in the present application should be included in the protection scope of the present application. Therefore, the protection scope of the present application shall be subject to the attached claims.

Claims

1. A charging scheduling method for engineering vehicles, characterized in that: The method comprises: Acquiring charging scheduling conditions, wherein the charging scheduling conditions include current time, charging pile information, vehicle information of each of the engineering vehicles, construction task data and / or charging preference data of a driver currently corresponding to each of the engineering vehicles; Based on the charging scheduling conditions and a pre-trained charging behavior prediction model, determining the charging behavior of each engineering vehicle, wherein the charging behavior includes charging time information and information of a target charging pile; A preset algorithm is used to optimize the charging behavior of each of the engineering vehicles and determine the charging scheduling strategy.

2. The method according to claim 1, characterized in that The training process of the charging behavior prediction model includes: Acquiring historical charging data, wherein the historical charging data includes historical charging scheduling conditions and historical charging behaviors; Processing the historical charging data to generate a training set and a verification set; Training two or more machine learning models based on the training set; The trained machine learning models are verified and optimized based on the verification set to select one of the machine learning models as the charging behavior prediction model.

3. The method according to claim 1, characterized in that The preset algorithm is a genetic algorithm, and the preset algorithm is used to optimize the charging behavior of each engineering vehicle to determine the charging scheduling strategy, including: Initializing the population of the genetic algorithm according to the charging time information, so that each chromosome of each individual in the population corresponds to the charging time information of the engineering vehicle; Perform evolution iteration on the individuals in the population according to a preset strategy to determine the individual with the highest fitness value, wherein the fitness value is determined based on the charging scheduling condition and the charging behavior corresponding to each individual; A charging scheduling strategy is determined according to the individual with the highest fitness value.

4. The method according to claim 3, characterized in that Initializing the population of the genetic algorithm according to the charging time information so that each chromosome of each individual in the population corresponds to the charging time information of the engineering vehicle, comprises: Determine the charging status of each engineering vehicle in a preset time period according to the charging time information of each engineering vehicle, wherein the charging status includes being in a charging state and not being in a charging state; Encoding the charging status of each of the engineering vehicles in a preset time period to generate a chromosome corresponding to the charging time information of each of the engineering vehicles; Individuals are generated according to the chromosomes corresponding to the engineering vehicles to obtain an initialized population.

5. The method according to claim 3, characterized in that The step of performing evolution iteration on the individuals in the population according to a preset strategy to determine the individual with the highest fitness value includes: In the process of performing evolution iteration on the individuals in the population according to the preset strategy, the fitness value of each chromosome in the current individual is calculated according to the charging scheduling condition and the charging behavior corresponding to the current individual; Calculating the fitness value of the current individual according to the fitness value of each chromosome; After the iteration is finished, the individual with the highest fitness value is determined according to the fitness values ​​of all individuals in the population.

6. The method according to claim 5, characterized in that The step of calculating the fitness value of each chromosome in the current individual according to the charging scheduling condition and the charging behavior corresponding to the current individual includes: Calculate the charging cost corresponding to each chromosome according to the current electricity price in the charging pile information, calculate the charging time corresponding to each chromosome according to the charging pile location in the target charging pile information and the charging time information, and calculate the charging urgency corresponding to each chromosome according to the current power in the vehicle information and the construction time in the construction task data; The fitness value of each chromosome in the current individual is calculated according to the charging cost, consumption time and charging urgency.

7. The method according to claim 3, characterized in that The step of performing evolution iteration on the individuals in the population according to a preset strategy includes: Determine whether the current number of iterations reaches the preset number of iterations; If the current number of iterations reaches the preset number of iterations, the evolution iteration is stopped; If the current number of iterations does not reach the preset number of iterations, the chromosomes in the current individual are crossovered and mutated according to the fitness values ​​of each chromosome in the current individual to iteratively generate a new individual, and the charging behavior corresponding to the new individual is different from the charging behavior corresponding to the current individual.

8. The method according to any one of claims 1 to 7, characterized in that After optimizing the charging behavior of each engineering vehicle by using a preset algorithm and determining the charging scheduling strategy, the method further includes at least one of the following: Acquire charging data after implementing the charging scheduling strategy, and optimize the charging behavior prediction model according to the charging data; Acquire operation data of charging scheduling and / or charging data after implementing the charging scheduling strategy, and adjust the parameters of the genetic algorithm according to the operation data and / or the charging data.

9. An electronic device, characterized in that: The method comprises a processor, a memory, and a computer program stored in the memory and executable on the processor, wherein the computer program implements the steps of the method according to any one of claims 1 to 8 when executed by the processor.

10. A computer-readable storage medium, characterized in that: The computer-readable storage medium stores a computer program, and when the computer program is executed by a processor, the steps of the method according to any one of claims 1 to 8 are implemented.

Citation Information

Cited By

  • Dynamic power distribution regulation and control method for tunnel construction of electric loader and dumper

    CN120806519A