Intelligent container ship quay crane operation plan generation method based on genetic algorithm

Abstract

The invention discloses a container ship shore bridge operation plan intelligent generation method based on a genetic algorithm, and relates to the technical field of port automatic scheduling. The method comprises the following steps: performing problem modeling coding based on input parameters; selecting individuals from the population based on a selection algorithm to enter a next generation and updating the population, repairing time conflicts through a crossover algorithm, and performing adaptive parameter adjustment; performing digital twinborn verification based on a simulation engine; adjusting parameters of the genetic algorithm, and optimizing the genetic algorithm. According to the method, through efficient solution under dynamic multi-constraint, the NP-hard scheduling problem under multi-dimensional constraint conditions such as uncertain ship berthing time, limited shore bridge movement energy consumption and operation priority conflict is solved; real-time performance and robustness are balanced, wherein generation and dynamic adjustment of a minute-level scheduling scheme are achieved through combination of a genetic algorithm and digital twinborn simulation; key indexes such as quay crane utilization rate, total operation duration and energy consumption are synchronously optimized, and the overall operation efficiency of a port is improved.

Description

Technical Field

[0001] The present invention belongs to the technical field of port automation scheduling, and in particular relates to a method for intelligently generating a container ship quay crane operation plan based on a genetic algorithm. Background Art

[0002] Against the backdrop of global trade pressures, increasing port congestion, accelerated intelligent transformation, and the "dual carbon" goals, container terminals face unprecedented challenges in efficiency, cost, safety, and environmental protection. The global supply chain is highly dependent on maritime transport, with larger container ships becoming increasingly popular and single-vessel loading and unloading volumes surging, posing unprecedented challenges to port operational efficiency. Port congestion has become a global issue, with extended ship stays leading to high demurrage costs and the risk of supply chain disruption. Efficient and intelligent quay crane scheduling is a key bottleneck in alleviating port congestion and enhancing ports' core competitiveness.

[0003] For traditional manual scheduling mode: Strong reliance on experience, poor replicability and scalability: Relying on the experience of "old masters", it is difficult to form a standardized, quantifiable and inheritable scheduling knowledge base, and the scheduling quality fluctuates greatly.

[0004] Serious lag in dynamic response: Frequent sudden disturbances caused by weather, sudden malfunctions of quay cranes / trucks, dynamic shifts in ship arrival times, poor truck turnover due to yard congestion, special operations, and the dismantling of large items often result in lengthy and inefficient scheduling. Weak information integration capabilities: It is difficult to integrate and process heterogeneous data from multiple sources such as ship loading plans, yard management systems, equipment status monitoring, real-time meteorological data, and traffic control information in real time, resulting in insufficient basis for decision-making.

[0005] Subjectivity and potential conflicts: Manual scheduling is easily influenced by subjective factors, and different dispatchers' plans vary greatly. Moreover, it is difficult to handle multi-party conflicts fairly and transparently when resources are tight.

[0006] For the current static / heuristic algorithm mode: The conflict between "combinatorial explosion" and real-time performance: Faced with the complex matching problem of dozens of quay cranes and thousands of containers, the calculation time of precise algorithms is unacceptable. Traditional heuristic algorithms, while fast, have limited solution quality and are prone to falling into local optimality. Rigid planning lacks adaptability and cannot respond to the aforementioned sudden disturbances online, quickly, and with high quality during plan execution. It also struggles to effectively incorporate dynamically changing constraints. Oversimplification of the model: To reduce complexity, real constraints are often oversimplified, resulting in the generated plan being infeasible or inefficient in practice.

[0007] For example, pursuing the shortest time for a single ship may result in an increase in energy consumption due to frequent movement of the shore bridge; strictly adhering to all safety distances may reduce the utilization rate of the shore bridge; and fully meeting the highest priority may disrupt the overall operation balance. The existing method lacks effective means to balance these conflicting goals in a complex and dynamic environment. The scheduling system based on traditional methods or primary informatization has been unable to meet the requirements of efficient, flexible and resilient smart ports.

[0008] The popularization of technologies such as Internet of Things, 5G, Beidou / GPS positioning, etc. makes it possible to obtain real-time shore bridge position state, truck dynamics, container position, environmental information, etc., laying a foundation for data-driven intelligent dynamic scheduling; evolutionary algorithms represented by genetic algorithms, reinforcement learning and other AI technologies show great potential in handling complex combinatorial optimization, multi-objective decision-making and dynamic adaptability, providing a technical path to break through the limitations of traditional methods.

[0009] Therefore, the present application provides a container ship shore bridge operation plan intelligent generation method based on genetic algorithm. SUMMARY

[0010] The present application aims to provide a container ship shore bridge operation plan intelligent generation method based on genetic algorithm, which optimizes key indicators such as shore bridge utilization rate, total operation time, energy consumption, etc. at the same time, improves the overall operation efficiency of the port, and solves the NP-hard scheduling problem under the multi-dimensional constraint conditions such as uncertain ship berthing time, shore bridge movement energy consumption limit and operation priority conflict.

[0011] To solve the above technical problems, the present application is realized by the following technical scheme: The present application is a container ship shore bridge operation plan intelligent generation method based on genetic algorithm, comprising the following steps: Based on the input parameters, the problem is modeled, and a hybrid coding strategy is adopted to encode the problem solution into a chromosome structure that can be processed by genetic algorithm; Based on the selection algorithm, individuals are selected from the population to enter the next generation and update the population, time conflicts are repaired through the crossover algorithm, and adaptive parameter adjustment is performed; Based on the simulation engine, digital twin verification is performed to evaluate whether the repaired chromosome structure meets the constraint conditions and optimization objectives; Based on the simulation verification result and the optimization objective calculation deviation, the genetic algorithm parameters are adjusted to optimize the genetic algorithm.

[0012] Further, the input parameters include the gantry crane basic information, the tidal height and ship draft information, the height of the landing pad, the route operation time information, the container loading and unloading priority information, and the ship import and export message data. Specifically, the gantry crane basic information includes the number, position, working efficiency, and load limit; the tidal height and ship draft affect the operation safety and feasibility; the height of the landing pad affects the efficiency of the container hoisting operation; the route operation time is the standard operation time of different routes; the container loading and unloading priority is determined according to the destination and cargo type; and the ship import and export message data includes the cargo information and loading and unloading sequence requirements.

[0013] Further, the method of encoding the problem into a chromosome structure that can be processed by the genetic algorithm includes: The gantry crane allocation part: each gene represents a container, and the value is the allocated gantry crane number; The operation time sequence part: each gene represents the start operation time of the corresponding container; The target optimization is based on minimizing the total operation time, penalizing conflicts, and violating priorities.

[0014] Further, the selection algorithm is the tournament selection method, and the method of selecting individuals from the population into the next generation based on the selection algorithm and updating the population includes: Randomly select K individuals from the population, K being a preset value; Compare the fitness values of the K individuals, and select the optimal individual into the next generation; Repeat the selection process until the new population size meets the requirements.

[0015] Further, the method of repairing time conflicts through the crossover algorithm includes: Randomly select two crossover points based on the two-point crossover operator for the operation sequence part; Exchange the gene segments of the two parents between the crossover points; Repair the sequences that may cause time conflicts and solve the possible time conflict problems.

[0016] Further, the adaptive parameter adjustment includes the mutation rate adjustment, and the mutation rate is dynamically linearly adjusted according to the iteration number; the mutation rate formula is: Mutation rate = initial mutation rate * (1 - current iteration number / maximum iteration number).

[0017] Further, the adaptive parameter adjustment includes the dynamic adjustment of the mutation rate, and the mutation rate is dynamically nonlinearly adjusted according to the iteration number; the mutation rate formula is: Mutation rate = mutation rate at the end of iteration + (initial mutation rate - mutation rate at the end of iteration) * (1 - tanh(X1 * current iteration number / maximum iteration number), X1 being a parameter representing the adjustment decay speed, X1 being a preset value.

[0018] Further, the adaptive parameter adjustment includes population size dynamic adjustment, which evaluates and adjusts the population size after each generation evolution, and the population size dynamic adjustment method includes: Initial population size setting: pop_size=X2*ship berth number; X2 is a preset value; ship berth number: refers to the number of container stacking positions on a container ship, usually calculated in 20-foot standard containers, for example, a 24000TEU ship may have about 24 berths; Diversity index calculation: diversity index=1-(optimal individual fitness value in current population / average fitness value of current population); Dynamic reduction trigger condition: population reduction is triggered when the diversity index is less than X3, and X3 is a preset value; The reduction operation is: new population size=90% of current population size, but not less than 50.

[0019] Further, the method for verifying the digital twin based on the simulation engine includes: Decode the repaired optimal chromosome generated by the genetic algorithm into a job Gantt chart; Import the simulation engine to establish a port operation digital twin model; Input actual parameters for simulation; Collect key performance indicators (KPIs): such as total operation time, equipment utilization rate, ship time in port, etc.; Evaluate whether the scheme meets the constraint conditions and optimization objectives.

[0020] Further, the method for optimizing the genetic algorithm includes: Compare the simulation verification results with the optimization objectives, and calculate the deviation; Analyze the reasons for the deviation and adjust the genetic algorithm parameters (such as population size, crossover rate, mutation rate); Update the fitness function weights to highlight important indicators; Use actual execution data as new input parameters for re-optimization.

[0021] The present application has the following beneficial effects: Efficient solution under dynamic multi-constraints: solves the NP-hard scheduling problem under multi-dimensional constraint conditions such as uncertain ship berthing time, shore crane movement energy consumption limit, operation priority conflict, etc. Compared with traditional methods, the shore crane scheduling scheme generation speed is increased by 80%, and the average ship time in port is shortened by 10%; Balance between real-time and robustness: through the combination of genetic algorithm and digital twin simulation, minute-level scheduling scheme generation and dynamic adjustment are realized, supporting auxiliary operation rescheduling (such as shore crane failure, disassembly of large pieces, weather reasons); Multi-objective collaborative optimization: synchronously optimize key indicators such as quay crane utilization rate, total operation time, energy consumption, etc., improve the overall operation efficiency of the port, reduce the invalid movement of the quay crane through path planning, and reduce energy consumption by 10%-15%.

[0022] Of course, implementing any product of the present application does not necessarily require all the advantages described above to be achieved at the same time. BRIEF DESCRIPTION OF DRAWINGS

[0023] In order to more clearly illustrate the technical solutions of the embodiments of the present application, the drawings needed for the embodiment description will be briefly introduced as follows. Obviously, the drawings in the following description are only some embodiments of the present application, and other drawings can also be obtained by those skilled in the art without creative labor on the basis of these drawings.

[0024] Figure 1 Intelligent generation method of container ship quay crane operation plan based on genetic algorithm of the present application; Figure 2 Method diagram for updating population. DETAILED DESCRIPTION

[0025] In the following description, specific details are set forth in order to provide a thorough understanding of embodiments of the application. However, persons of ordinary skill in the art will readily appreciate that embodiments of the application can be practiced without some or all of the specific details set forth herein. In other instances, well-known structures, devices, circuits, and materials have not been described in detail in order to avoid obscuring the application.

[0026] It should be understood that when used in the specification and the appended claims, the term "comprises" indicates the presence of the described features, integers, steps, operations, elements, and / or components, but does not preclude the presence or addition of one or more other features, integers, steps, operations, elements, components, and / or groups thereof.

[0027] It should also be understood that the term "and / or" as used herein refers to any combination of one or more of the associated listed items and all possible combinations thereof, and includes these combinations.

[0028] As used in the specification and the appended claims herein, the term "if' can be interpreted as meaning "when" or "upon" or "in response to a determination" or "in response to a detection" depending on the context. Similarly, the phrase "if it is determined" or "if [the described condition or event] is detected" can be interpreted as meaning "upon a determination" or "in response to a determination" or "upon a detection [of the described condition or event]" or "in response to a detection [of the described condition or event]" depending on the context.

[0029] In addition, in the description of the present application and the appended claims, the terms "first", "second", "third", etc. are only used to distinguish the description and cannot be understood as indicating or implying relative importance.

[0030] Reference in the specification to "one embodiment" or "some embodiments" means that a particular feature, structure, or characteristic described in connection with the embodiment is included in at least one embodiment of the application. The appearances of the phrases "in one embodiment", "in some embodiments", "in other embodiments", "in additional embodiments", etc. in various places in the specification are not necessarily all referring to the same embodiment, although they can. The terms "comprising", "including", "having" and their variants mean "including but not limited to", unless otherwise expressly specified.

[0031] Embodiment I: Please refer to Figure 1 The application is an intelligent generation method of container ship shore crane operation plan based on genetic algorithm, including the following steps: problem modeling based on input parameters, using a hybrid coding strategy to encode the problem into a chromosome structure that can be processed by genetic algorithm; selecting individuals from the population into the next generation based on the selection algorithm and updating the population, repairing time conflicts through the crossover algorithm, and adjusting the adaptive parameters; verifying the digital twin based on the simulation engine, evaluating whether the repaired chromosome structure meets the constraint conditions and optimization objectives; calculating the deviation of the simulation verification result and the optimization objective, and adjusting the genetic algorithm parameters to optimize the genetic algorithm, solving the NP-hard scheduling problem under the multi-dimensional constraint conditions of uncertain ship berthing time, shore crane movement energy consumption limit, operation priority conflict, etc. Compared with the traditional method, the shore crane scheduling scheme generation speed is improved by 80%, and the average ship time in port is shortened by 10%.

[0032] Embodiment II: The application is an intelligent generation method of container ship shore crane operation plan based on genetic algorithm, including the following steps: Step 1: Problem modeling and encoding: Based on input parameters, model the problem and use a hybrid encoding strategy to encode the problem solution into a chromosome structure that can be processed by genetic algorithms. Step 2: Genetic algorithm operation: Select individuals from the population based on selection algorithms to enter the next generation and update the population, repair time conflicts through crossover algorithms, and adjust adaptive parameters. Step 3: Digital twin verification: Based on the simulation engine, verify the digital twin, and evaluate whether the repaired chromosome structure meets the constraint conditions and optimization objectives. Step 4: Dynamic feedback: Calculate the deviation based on the simulation verification results and optimization objectives, and adjust the genetic algorithm parameters to optimize the genetic algorithm.

[0033] As an embodiment provided by the present application, preferably, the following advantages are possessed: (1) Reduce the time of ships in port through genetic algorithm optimization, and improve port throughput; (2) Enhance the feasibility of the scheme: digital twin verification ensures that the scheme is feasible under real constraints; (3) Reduce operating costs: optimize shore crane allocation and operation time, reduce equipment energy consumption and labor costs; (4) Improve flexibility: dynamic feedback mechanism enables the system to adapt to unexpected situations (such as weather changes, equipment failures); Support decision optimization: Provide quantitative performance indicators to assist management personnel in making more scientific decisions.

[0034] As an embodiment provided by the present application, preferably, the input parameters include shore crane basic information, tidal height and ship draft information, height information, route operation time information, container loading and unloading priority information, and ship import and export message data.

[0035] As an embodiment provided by the present application, preferably, the input parameters include: Shore crane basic information: number, location, work efficiency, load limit, etc. Tidal height and ship draft: affect operation safety and feasibility; Height of the high: affects the efficiency of container lifting operation; Route operation time: standard operation time for different routes; Container loading and unloading priority: determined according to destination, cargo type, etc. Ship import and export message data: contains cargo information, loading and unloading sequence requirements, etc.

[0036] Embodiment three: As an embodiment provided by the present application, preferably, based on embodiment one or two, the method of encoding the problem solution into a chromosome structure that can be processed by genetic algorithms includes: Quay crane allocation part: each gene represents a container, and the value is the allocated quay crane number, which is an integer; Operation time series part: each gene represents the start operation time of the corresponding container (floating point number); Optimization objective: Minimize total operation time + penalize conflicts (quay crane collisions) and priority violations.

[0037] As an embodiment provided by the present invention, preferably, the chromosome structure is:

[0038] Specifically: #Chromosome = [quay crane ID, relative start time, operation mode, priority, coordinated quay crane...] chromosome=[ 4,#quay crane ID 15.3,#Start time (minutes) 1, #operation mode (0: normal, 1: special) ].

[0039] As an embodiment provided by the present invention, preferably, the optimization objective is to optimize the target by combining minimization of total operation time (ship in port time, quay crane idle time) + penalty conflicts (quay crane collision risk, safety distance risk) and priority violations (special operation delay, single hook small cabinet timeout); The target optimization function is: \text{Minimize}F=\underbrace{w_1\cdotT_{\text{total}}_\text{job time}+\underbrace{w_2\cdotP_{\text{collision}}_\text{collision penalty}+\underbrace{w_3\cdotP_{\text{priority}}_\text{priority penalty}; in: T_{\text{total}}=\sum_{i=1}^{N}(t_i^{\text{end}}-t_i^{\text{start}}})$ (total operation time) $P_{\text{collision}}=\sum_{j=1}^{M}\text{CollisionRisk}_j$ (collision risk penalty) $P_{\text{priority}}=\sum_{k\inK}\max(0,\text{ActualTime}_k-\text{Deadline}_k)$ (priority violation penalty).

[0040] As an embodiment provided by the present application, preferably, the optimization step is: Obtain the position of the quay crane and calculate the safety distance of the vessel; Obtain the operation time window and detect the time overlap; Identify the conflict pair and calculate the conflict severity based on the penalty function; Optimize the target based on the optimization function.

[0041] Embodiment four: Based on embodiment one, as an embodiment provided by the present application, preferably, adaptive weight adjustment is also performed: SI1: Evaluate the current population performance every 10 generations; SI2: Calculate the degree of violation of each target: A: Deviation of actual average completion time from theoretical optimal time = actual average time of system executing tasks - theoretically possible shortest time; if the result is positive, it indicates that there is a loss of time efficiency in the system; if it is zero or negative, it reaches or is better than the theoretical optimum; B: Frequency of conflict occurrence = number of tasks with conflict / total number of tasks; the closer the value is to 1, the more frequent the conflict; zero indicates no conflict; C: Proportion of priority tasks delayed = number of priority tasks not completed on time / total number of priority tasks, the larger the value, the worse the system's compliance with priority.

[0042] SI3: Dynamically update the weight: each new weight = (initial weight x current violation value) / total penalty value; If a violation value suddenly increases, the corresponding weight will significantly increase, making the system prioritize optimizing that target.

[0043] If all violation values are zero, the weight will remain in the initial proportion; when a problem (such as a sudden surge in collisions) suddenly becomes serious, its weight is automatically increased to guide resources to prioritize solving that problem.

[0044] As shown in Figure 2 As an embodiment provided by the present application, preferably, the selection algorithm is tournament selection, and the method of selecting individuals from the population into the next generation and updating the population based on the selection algorithm includes: Initialize parameters: Set the tournament size K; Determine the size N of the new generation population (usually the same as the current generation); Randomly select K individuals from the population, K is a preset value; Compare the fitness values of the K individuals, and select the optimal individual into the next generation; Repeat the selection process until the new population size reaches the requirement.

[0045] As an embodiment provided by the present application, preferably, the method for repairing time conflicts by crossover algorithm comprises: Randomly select two crossover points based on the two-point crossover operator for the job sequence part; Exchange the gene segments between the two crossover points of the two parents; Repair the sequence that may cause time conflicts, and repair the possible time conflict problems (such as job order reversal, time overlap, etc.).

[0046] As an embodiment provided by the present application, preferably, the adaptive parameter adjustment includes mutation rate adjustment, the mutation rate is dynamically linearly adjusted according to the iteration number, the mutation rate is higher at the beginning to increase the search range, and the mutation rate is lower at the later stage to improve the precision; the mutation rate formula is: Mutation rate = initial mutation rate * (1-current iteration number / maximum iteration number); based on the mutation rate, small amplitude disturbance is carried out on the job time sequence part, and random reassignment is carried out on the quay crane allocation part.

[0047] Embodiment five: As an embodiment provided by the present application, preferably, the adaptive parameter adjustment includes dynamic adjustment of the mutation rate, the mutation rate is dynamically nonlinearly adjusted according to the iteration number, and the mutation rate formula is: Mutation rate = mutation rate at the end of iteration + (initial mutation rate-mutation rate at the end of iteration) * (1-tanh (X1*current iteration number / maximum iteration number), X1 is a parameter representing the adjustment decay speed, and X1 is a preset value.

[0048] As an embodiment provided by the present application, preferably, the adaptive parameter adjustment includes dynamic adjustment of the population size, the population size is evaluated and adjusted after each generation evolution, and the population size dynamic adjustment method comprises: Initial population size setting: pop_size = X2*ship berth number; X2 is a preset value; Wherein, the ship berth number refers to the number of stacked positions of containers on a container ship (usually calculated in units of 20-foot standard containers, for example, a ship of 8000 TEU may have about 100 berths); the larger the problem size (the more berths), the larger the initial population size is required to cover the solution space; if the ship has 100 berths, the initial population size = 5*100 = 500; Diversity index calculation: diversity = 1-(best_fitness / avg_fitness); where, best_fitness: fitness value of the best individual in the current population (usually the minimum in minimization problem); avg_fitness: average fitness value of the current population; If best_fitness is much smaller than avg_fitness (i.e. the best solution is much better than the average), diversity is close to 1, indicating high population diversity; otherwise, if best_fitness is close to avg_fitness, diversity is close to 0, indicating the population tends to be homogenized; Dynamic reduction trigger condition: population reduction is triggered when diversity index < X3, X3 is a preset value; For example, X3 = 0.2, threshold 0.2: an empirical value, indicating that the population diversity has been significantly reduced (the fitness of the best individual reaches more than 80% of the average level); The reduction operation is: new population size = 90% of the current population size, but not less than 50; pop_size = max(50, pop_size * 0.9) The new population size is 90% of the current population size, but not less than 50 (to avoid premature convergence caused by too small population size), for example, if the current population size is 500, the reduced population size is max(50, 500 * 0.9) = 450; the next time the condition is met, it becomes max(50, 450 * 0.9) = 405, and so on; Execution timing: usually after each generation evolution, evaluation and adjustment are performed; As an embodiment provided by the present application, preferably, adaptive parameter adjustment, initial stage (high diversity): large population size → simulates large-scale reproduction of biological organisms, extensive exploration of solution space; later stage (low diversity): reduced population size → simulates stable population size after natural selection, concentrates computing resources to optimize high-quality solutions; avoids the computational overhead of maintaining a large population throughout the optimization process (the pier scheduling problem has a large amount of evaluation fitness calculation); when diversity > 0.2, the population size is maintained to prevent excessive reduction before a satisfactory solution is found; when the population size is reduced to 50 and still cannot be improved, the optimization process can be terminated in advance (convergence criterion).

[0049] Embodiment six: As an embodiment provided by the present application, the method for digital twin verification based on a simulation engine preferably comprises: Decoding the repaired optimal chromosome generated by the genetic algorithm into a job Gantt chart; Importing a simulation engine such as AnyLogic to establish a digital twin model of port operations; Input actual parameters (such as weather, equipment status, personnel efficiency, etc.) for simulation; Collect key performance indicators (KPIs): such as total job time, equipment utilization, ship in port time, etc. Evaluate whether the solution meets the constraints and optimization objectives.

[0050] As an embodiment provided by the present application, preferably, the method of optimizing genetic algorithm comprises: Compare the simulation verification results with the optimization objectives, and calculate the deviation; Analyze the reasons for the deviation, and adjust the genetic algorithm parameters (such as population size, crossover rate, mutation rate); Update the fitness function weight to highlight important indicators; Use actual execution data as new input parameters for re-optimization.

[0051] A container ship shore crane operation plan intelligent generation method based on genetic algorithm, efficient solution under dynamic multi-constraints: solve the NP-hard scheduling problem under multi-dimensional constraint conditions such as uncertain ship berthing time, shore crane movement energy consumption limit, operation priority conflict, compared with traditional method, the shore crane scheduling scheme generation speed is increased by 80%, the average ship in port time is shortened by 10%; Real-time and robustness balance: through the combination of genetic algorithm and digital twin simulation, realize minute-level scheduling scheme generation and dynamic adjustment, support the rescheduling of auxiliary operations (such as shore crane failure, disassembly of large pieces, weather reasons); Multi-objective collaborative optimization: simultaneously optimize key indicators such as shore crane utilization rate, total operation time, energy consumption, etc., improve the overall operation efficiency of the port, reduce the invalid movement of shore cranes through path planning, and reduce energy consumption by 10%-15%.

[0052] In the description of the present specification, the description of the terms "one embodiment", "example", "specific example" and the like means that the specific features, structures, materials or characteristics described in conjunction with the embodiment or example are included in at least one embodiment or example of the present application. In the present specification, the illustrative description of the above terms does not necessarily refer to the same embodiment or example. Moreover, the specific features, structures, materials or characteristics described can be combined in any one or more embodiments or examples in a suitable manner.

[0053] The preferred embodiments of the application disclosed above are only used to help explain the application. The preferred embodiments do not describe all the details, nor limit the application to the specific embodiments described. Obviously, many modifications and changes can be made according to the content of the present specification. The present specification selects and describes these embodiments in order to better explain the principles and practical applications of the present application, so that those skilled in the art can well understand and utilize the present application. The present application is limited only by the claims and their full scope and equivalents.

Claims

1. A method for intelligently generating container ship quay crane operation plans based on genetic algorithms, characterized in that: It includes the following steps: Based on the input parameters, conduct problem modeling, adopt a hybrid coding strategy, and encode the problem solution into a chromosome structure that can be processed by a genetic algorithm; Based on the selection algorithm, select individuals from the population to enter the next generation and update the population, repair time conflicts through the crossover algorithm, and perform adaptive parameter adjustment; Based on the simulation engine, conduct digital twin verification to evaluate whether the repaired chromosome structure meets the constraint conditions and optimization objectives; Calculate the deviation based on the simulation verification results and the optimization objectives, and adjust the genetic algorithm parameters to optimize the genetic algorithm.

2. The method for intelligently generating container ship quay crane operation plan based on genetic algorithm according to claim 1, characterized in that: The input parameters include quay crane foundation information, tide height and ship draft information, fender height information, route operation working hours information, container loading and unloading priority information, and ship import and export message data.

3. The method for intelligently generating container ship quay crane operation plan based on genetic algorithm according to claim 1, characterized in that: The method of encoding the problem solution into a chromosome structure that can be processed by a genetic algorithm includes: Quay crane allocation part: Each gene represents a container, and the value is the allocated quay crane number; Operation time series part: Each gene represents the start operation time of the corresponding container; Conduct objective optimization based on minimizing the total operation time, penalty for conflicts, and priority violations.

4. The method for intelligently generating container ship quay crane operation plan based on genetic algorithm according to claim 1, characterized in that: The selection algorithm is the tournament selection method. The method of selecting individuals from the population to enter the next generation and updating the population based on the selection algorithm includes: Randomly select K individuals from the population, where K is a preset value; Compare the fitness values of the K individuals and select the optimal individual to enter the next generation; Repeat the selection process until the new population size reaches the requirement.

5. The method for intelligently generating container ship quay crane operation plan based on genetic algorithm according to claim 1, characterized in that: The method of repairing time conflicts through the crossover algorithm includes: Randomly select two crossover points for the operation sequence part based on the two-point crossover operator; Exchange the gene segments between the two parents between the crossover points; Repair the sequences that may generate time conflicts.

6. The method for intelligently generating container ship quay crane operation plan based on genetic algorithm according to claim 1, characterized in that: Adaptive parameter adjustment includes mutation rate adjustment. The mutation rate is dynamically linearly adjusted according to the number of iterations; the mutation rate formula is: Mutation rate = initial mutation rate * (1 - current iteration number / maximum iteration number).

7. The method for intelligently generating container ship quay crane operation plan based on genetic algorithm according to claim 1, characterized in that: Adaptive parameter adjustment includes dynamic mutation rate adjustment. The mutation rate is dynamically non-linearly adjusted according to the number of iterations. The mutation rate formula is: Mutation rate = mutation rate at the end of iteration + (initial mutation rate - mutation rate at the end of iteration) * (1 - tanh(X1 * current iteration number / maximum iteration number)), where X1 is a parameter representing the adjustment decay rate, and X1 is a preset value.

8. The method for intelligently generating container ship quay crane operation plan based on genetic algorithm according to claim 1, characterized in that: Adaptive parameter adjustment includes dynamic population size adjustment. Evaluate and adjust the population size after each generation of evolution. The method of dynamic population size adjustment includes: Initial population size setting: pop_size = X2 * number of ship berths; X2 is a preset value; Diversity index calculation: Diversity index = 1 - (fitness value of the optimal individual in the current population / average fitness value of the current population); Dynamic reduction trigger condition: Trigger population reduction when the diversity index < X3, where X3 is a preset value; Reduction operation: New population size = 90% of the current population size, but not less than 50.

9. The method for intelligently generating container ship quay crane operation plan based on genetic algorithm according to claim 1, characterized in that: The method of conducting digital twin verification based on the simulation engine includes: Decode the repaired optimal chromosome generated by the genetic algorithm into a Gantt chart of operations; Import into the simulation engine to establish a digital twin model of port operations; Input actual parameters for simulation; Collect key performance indicators; Evaluate whether the solution meets the constraints and optimization objectives.

10. The method for intelligently generating container ship quay crane operation plan based on genetic algorithm according to claim 1, characterized in that: Methods for optimizing genetic algorithms include: Compare the simulation verification results with the optimization goals and calculate the deviation; Analyze the causes of deviation and adjust genetic algorithm parameters; Update the fitness function weight; Use the actual execution data as new input parameters and re-optimize.

Images