Ship berthing decision-making method and system based on local optimization genetic algorithm

Abstract

The invention provides a ship docking decision-making method and system based on a local optimization genetic algorithm, and relates to the technical field of port scheduling, and the method comprises the steps: obtaining management data in a set management period; generating an initial population according to the number of ships to be berthed; the fitness of each individual is obtained based on the management data, whether the fitness smaller than a set threshold exists or not is judged, if yes, the individual corresponding to the maximum value of the fitness serves as a ship docking scheme in a set management period, and if not, the fitness is ranked from large to small, and then the first N individuals are selected as parent individuals; the priority of each ship needing to be berthed is obtained based on the ship information, local optimization is carried out on the N parent individuals based on each priority, and N optimized parent individuals are obtained; and performing crossover operator processing and mutation operator processing on the optimized N parent individuals to obtain a new population. According to the method, the ship priority is considered, and the problem of environmental protection and efficiency imbalance is solved.

Description

Technical Field

[0001] This invention relates to the field of port scheduling technology, and in particular to a ship berthing decision-making method and system based on a local optimization genetic algorithm. Background Technology

[0002] Port production management primarily focuses on production scheduling. The key challenge is enhancing port service capabilities by improving operational efficiency and resource utilization. Vessel berthing scheduling, or vessel berthing dispatching, is an extremely complex systems engineering project. It requires comprehensive consideration of various factors, including vessel conditions, berth availability, port infrastructure, and tidal conditions. The efficiency and quality of scheduling directly impact the overall efficiency of port operations, making it a hot research topic in port scheduling.

[0003] Existing scheduling decision-making methods are mostly based on static parameters and do not take into account the emergency needs of ships and environmental protection indicators, resulting in an imbalance between environmental protection and efficiency. Summary of the Invention

[0004] The purpose of this invention is to provide a ship berthing decision-making method and system based on a local optimization genetic algorithm, which takes into account ship priority and solves the problem of the imbalance between environmental protection and efficiency.

[0005] A ship berthing decision-making method based on a local optimization genetic algorithm includes:

[0006] S1, acquire management data within a set management period; the management data includes berth status information, weather information, and vessel information of vessels that need to berth;

[0007] S2, generate an initial population based on the number of ships that need to dock; each individual in the initial population represents the berth and docking time of each ship that needs to dock.

[0008] S3, based on the management data, obtain the fitness of each individual, and determine whether there is a fitness less than a set threshold. If there is, execute S6. If not, arrange the fitness values ​​in descending order and select the top N individuals as parent individuals.

[0009] The expression for fitness is:

[0010] R=α×E+β×(1-C)+γ×(1-S)-K;

[0011] Where R is fitness, E is efficiency factor, C is carbon emission factor, S is safety factor, K is matching factor, α is efficiency coefficient, β is carbon emission coefficient, γ is safety factor, α+β+γ=1, E=1-y / Y, y is the average waiting time of B ships that need to berth, Y is the duration of the set management cycle, C is the normalized average of the carbon emissions of B ships that need to berth, the carbon emission is the sum of emissions during waiting time and emissions during travel time, and S is determined based on environmental risk value and trajectory conflict probability; the higher the environmental risk value, the larger S is, and the higher the trajectory conflict probability. The higher the probability, the larger S is. K is the sum of the matching degrees of B ships that need to berth. The matching degree is the sum of the draft matching degree and the required equipment matching degree. When the ship's draft is greater than the maximum draft of the berth, the draft matching degree is the set maximum value. When the ship's draft is less than or equal to the maximum draft of the berth, the draft matching degree is the set minimum value. When the ship's required unloading equipment is the same as the unloading equipment installed at the berth, the required equipment matching degree is the set minimum value. When the ship's required unloading equipment is different from the unloading equipment installed at the berth, the required equipment matching degree is the set maximum value.

[0012] S4. Based on the ship information, obtain the priority of each ship that needs to berth, and perform local optimization on the N parent individuals based on the priority to obtain the optimized N parent individuals.

[0013] S5, perform crossover and mutation operations on the optimized N parent individuals to obtain a new population, and return to S3;

[0014] S6, the individual corresponding to the maximum fitness value is used as the ship berthing scheme within the set management period.

[0015] Optionally, the vessel information includes the vessel's draft, vessel type, vessel cargo attributes, remaining fuel, vessel malfunction status, waiting time, current location, required unloading equipment, and credit score; the meteorological information includes wind speed, wave height, visibility, and precipitation probability; the berth status information includes the berth's maximum draft, the berth's installed unloading equipment, and the remaining berthing time, where the remaining berthing time is the remaining berthing time of the vessel currently berthed at the berth at the start of the set management cycle.

[0016] Optionally, the priority is based on vessel type, vessel cargo attributes, vessel fuel remaining, vessel malfunction status, and vessel credit value.

[0017] Optionally, the environmental risk value is determined based on the meteorological information, as follows:

[0018] Set initial environmental risk values;

[0019] Divide the wind speed by the safe wind speed to get the wind speed ratio.

[0020] Divide the wave height by the safe wave height to obtain the wave height ratio.

[0021] Divide the safe visibility by the visibility to obtain the visibility ratio.

[0022] Divide the probability of precipitation by the probability of safe precipitation to obtain the precipitation probability ratio;

[0023] The environmental risk value is the product of the initial environmental risk value, the wind speed ratio, the wave height ratio, the visibility ratio, and the precipitation probability ratio.

[0024] Optionally, the real-time location of the vessel to be docked at each time point within a set management period is obtained based on the vessel's current location, the berth, and the docking time.

[0025] Based on the remaining berthing time of the vessel, determine the real-time location of the berthed vessel at each time point within the set management cycle;

[0026] If the real-time position of the i-th vessel needing to dock at time k coincides with the real-time position of the j-th vessel needing to dock at time k, or the real-time position of the i-th vessel needing to dock at time k coincides with the real-time position of the a-th already docked vessel at time k, then the trajectory conflict probability is 1; otherwise, the trajectory conflict probability is 0.

[0027] This invention also provides a ship berthing decision system based on a local optimization genetic algorithm, which includes:

[0028] The data acquisition module is used to acquire management data within a set management period; the management data includes berth status information, meteorological information, and vessel information of vessels that need to berth.

[0029] The initial population module is used to generate an initial population based on the number of ships that need to dock; each individual in the initial population represents the berth and docking time of each ship that needs to dock.

[0030] The fitness module is used to obtain the fitness of each individual based on the management data, determine whether there is a fitness value less than a set threshold, if so, execute S6; if not, sort the fitness values ​​in descending order and select the top N individuals as parent individuals.

[0031] The expression for fitness is:

[0032] R=α×E+β×(1-C)+γ×(1-S)-K;

[0033] Where R is fitness, E is efficiency factor, C is carbon emission factor, S is safety factor, K is matching factor, α is efficiency coefficient, β is carbon emission coefficient, γ is safety factor, α+β+γ=1, E=1-y / Y, y is the average waiting time of B ships that need to berth, Y is the duration of the set management cycle, C is the normalized average of the carbon emissions of B ships that need to berth, the carbon emission is the sum of emissions during waiting time and emissions during travel time, and S is determined based on environmental risk value and trajectory conflict probability; the higher the environmental risk value, the larger S is, and the higher the trajectory conflict probability. The higher the probability, the larger S is. K is the sum of the matching degrees of B ships that need to berth. The matching degree is the sum of the draft matching degree and the required equipment matching degree. When the ship's draft is greater than the maximum draft of the berth, the draft matching degree is the set maximum value. When the ship's draft is less than or equal to the maximum draft of the berth, the draft matching degree is the set minimum value. When the ship's required unloading equipment is the same as the unloading equipment installed at the berth, the required equipment matching degree is the set minimum value. When the ship's required unloading equipment is different from the unloading equipment installed at the berth, the required equipment matching degree is the set maximum value.

[0034] The local optimization module is used to obtain the priority of each ship that needs to berth based on the ship information, and to perform local optimization on N parent individuals based on the priority to obtain N optimized parent individuals.

[0035] The crossover and mutation module is used to perform crossover and mutation operations on the optimized N parent individuals to obtain a new population, which is then returned to the fitness module.

[0036] The scheme decision module is used to select the individual corresponding to the maximum fitness value as the ship berthing scheme within the set management period.

[0037] Optionally, the vessel information includes the vessel's draft, vessel type, vessel cargo attributes, remaining fuel, vessel malfunction status, waiting time, current location, required unloading equipment, and credit score; the meteorological information includes wind speed, wave height, visibility, and precipitation probability; the berth status information includes the berth's maximum draft, the berth's installed unloading equipment, and the remaining berthing time, where the remaining berthing time is the remaining berthing time of the vessel currently berthed at the berth at the start of the set management cycle.

[0038] Optionally, the priority is based on vessel type, vessel cargo attributes, vessel fuel remaining, vessel malfunction status, and vessel credit value.

[0039] Optionally, the environmental risk value is determined based on the meteorological information, as follows:

[0040] Set initial environmental risk values;

[0041] Divide the wind speed by the safe wind speed to get the wind speed ratio.

[0042] Divide the wave height by the safe wave height to obtain the wave height ratio.

[0043] Divide the safe visibility by the visibility to obtain the visibility ratio.

[0044] Divide the probability of precipitation by the probability of safe precipitation to obtain the precipitation probability ratio;

[0045] The environmental risk value is the product of the initial environmental risk value, the wind speed ratio, the wave height ratio, the visibility ratio, and the precipitation probability ratio.

[0046] Optionally, the real-time location of the vessel to be docked at each time point within a set management period is obtained based on the vessel's current location, the berth, and the docking time.

[0047] Based on the remaining berthing time of the vessel, determine the real-time location of the berthed vessel at each time point within the set management cycle;

[0048] If the real-time position of the i-th vessel needing to dock at time k coincides with the real-time position of the j-th vessel needing to dock at time k, or the real-time position of the i-th vessel needing to dock at time k coincides with the real-time position of the a-th already docked vessel at time k, then the trajectory conflict probability is 1; otherwise, the trajectory conflict probability is 0.

[0049] The effects of this invention are as follows:

[0050] This invention presents a ship berthing decision-making method based on a local optimization genetic algorithm. When obtaining the fitness of an individual, it considers an efficiency factor, a carbon emission factor, a safety factor, and a matching factor. The efficiency factor is used to improve ship berthing efficiency, the carbon emission factor is used to reduce berthing energy consumption and reduce port carbon emissions, the safety factor is used to avoid conflicts and improve ship berthing safety based on wind speed, wave height, visibility, and precipitation probability, and the matching factor is used to improve scheduling accuracy so that the ship berthing berth meets the load-bearing and unloading conditions. Attached Figure Description

[0051] Figure 1 This is a flowchart of the ship berthing decision method based on local optimization genetic algorithm of the present invention. Detailed Implementation

[0052] Hereinafter, embodiments of the present invention will be described with reference to the accompanying drawings.

[0053] Figure 1 This is a flowchart of the ship berthing decision-making method based on local optimization genetic algorithm of the present invention, as shown below. Figure 1As shown, this invention provides a ship berthing decision-making method based on a local optimization genetic algorithm, which includes:

[0054] S1, acquire management data for the set management period. Management data includes berth status information, meteorological information, and vessel information for vessels requiring berthing. Specifically, vessel information includes vessel draft, vessel type, cargo attributes, remaining fuel, vessel malfunction status, waiting time, current position, required unloading equipment, and credit rating; meteorological information includes wind speed, wave height, visibility, and precipitation probability; berth status information includes maximum berth draft, unloading equipment installed at the berth, and remaining berthing time for vessels, where remaining berthing time is the remaining berthing time for vessels currently berthed at the set management period at the start of the period.

[0055] Preferably, a dual technology of "differential privacy + homomorphic encryption" is adopted. Before data is uploaded, noise is added through differential privacy, and cross-node data calculation is performed using partial homomorphic encryption to achieve "data does not leave the local machine, only the calculation results are shared".

[0056] S2 generates an initial population based on the number of ships that need to berth. Each individual in the initial population represents the berth and berthing time of each ship that needs to berth.

[0057] S3: Obtain the fitness of each individual based on the management data, and determine whether there is a fitness value less than the set threshold. If so, execute S6. If not, sort the fitness values ​​from largest to smallest and select the top N individuals as the parent individuals.

[0058] The expression for fitness is:

[0059] R=α×E+β×(1-C)+γ×(1-S)-K;

[0060] Where R is fitness, E is efficiency factor, C is carbon emission factor, S is safety factor, K is matching factor, α is efficiency coefficient, β is carbon emission coefficient, γ is safety factor, α+β+γ=1, E=1-y / Y, y is the average waiting time of B ships that need to berth, Y is the duration of the set management cycle, C is the normalized average of the carbon emissions of B ships that need to berth, the carbon emission is the sum of emissions during waiting time and emissions during travel time, and S is determined based on environmental risk value and trajectory conflict probability; the higher the environmental risk value, the larger S is, and the higher the trajectory conflict probability. The higher the probability, the larger S is. K is the sum of the matching degrees of B ships that need to berth. The matching degree is the sum of the draft matching degree and the required equipment matching degree. When the ship's draft is greater than the maximum draft of the berth, the draft matching degree is the set maximum value. When the ship's draft is less than or equal to the maximum draft of the berth, the draft matching degree is the set minimum value. When the ship's required unloading equipment is the same as the unloading equipment installed at the berth, the required equipment matching degree is the set minimum value. When the ship's required unloading equipment is different from the unloading equipment installed at the berth, the required equipment matching degree is the set maximum value.

[0061] Preferably, the environmental risk value is determined based on meteorological information, and the process is as follows:

[0062] Set initial environmental risk values.

[0063] Divide the wind speed by the safe wind speed to get the wind speed ratio.

[0064] Divide the wave height by the safe wave height to obtain the wave height ratio.

[0065] The visibility ratio is obtained by dividing the safe visibility by the actual visibility.

[0066] Divide the probability of precipitation by the probability of safe precipitation to obtain the precipitation probability ratio.

[0067] The environmental risk value is the product of the initial environmental risk value, wind speed ratio, wave height ratio, visibility ratio, and precipitation probability ratio.

[0068] Furthermore, the process for obtaining the trajectory conflict probability is as follows:

[0069] Based on the ship's current location, berth, and berthing time, the real-time location of the ships that need to berth within the set management cycle is obtained at each time point.

[0070] The real-time location of vessels already docked within a set management period is determined based on the remaining berthing time.

[0071] If the real-time position of the i-th vessel that needs to dock at time k coincides with the real-time position of the j-th vessel that needs to dock at time k, or the real-time position of the i-th vessel that needs to dock at time k coincides with the real-time position of the a-th already docked vessel at time k, then the probability of trajectory conflict is 1. If none of these exist, then the probability of trajectory conflict is 0.

[0072] S4: Based on the information of each vessel, the priority of each vessel requiring berthing is obtained. Based on each priority, N parent individuals are locally optimized to obtain N optimized parent individuals. The priority is obtained based on the vessel type, vessel cargo attributes, vessel fuel remaining amount, vessel fault status, and vessel credit value.

[0073] Specifically, the information entropy of each indicator is obtained, expressed as:

[0074]

[0075] Among them, e b Let m be the information entropy of the b-th indicator, m be the information factor, m = 1 / lnB, and M be the total number of indicators. x b This represents the weight of the b-th indicator.

[0076] Preferably, for the type of vessel, the specific gravity is 1 when the vessel is an emergency vessel, 0.7 when the vessel is a container ship, and 0.5 when the vessel is a bulk carrier.

[0077] Regarding the cargo attributes of ships, the weight is 1 when the cargo is medical supplies, 0.8 when the cargo is dangerous goods, and 0.5 when the cargo is general goods.

[0078] For the remaining fuel oil of a ship, the specific gravity is 1 when the remaining fuel oil is less than 20%, 0.6 when the remaining fuel oil is greater than or equal to 20% and less than or equal to 50%, and 0.3 when the remaining fuel oil is greater than 50%.

[0079] For the ship's fault status, the weight is 1 when the ship's fault status is serious, 0.6 when the ship's fault status is minor, and 0.3 when the ship's fault status is no fault.

[0080] For ship credit scores, the weighting is 1 when the ship credit score is greater than 90, 0.7 when the ship credit score is less than or equal to 90 but greater than or equal to 70, and 0.3 when the ship credit score is less than 70.

[0081] The weights of each indicator are obtained based on the information entropy of each indicator, expressed as follows:

[0082]

[0083] Among them, w b Let b be the weight of the b-th indicator.

[0084] The priority of ships requiring berthing is determined based on the weight and proportion of each indicator, expressed as:

[0085]

[0086] Among them, P g This is the priority of the g-th vessel that needs to dock.

[0087] Preferably, when the ship's fault status is a serious fault or the ship's cargo is classified as dangerous goods, the priority is multiplied by an additional 1.4 to ensure that emergency needs are met first.

[0088] S5 involves performing crossover and mutation operations on the optimized N parent individuals to obtain a new population, which is then returned to S3.

[0089] S6, the individual corresponding to the highest fitness value is selected as the ship berthing plan within the set management period. This invention also provides a ship berthing decision-making system based on a local optimization genetic algorithm, which includes:

[0090] The data acquisition module is used to acquire management data within a set management period. Management data includes berth status information, meteorological information, and vessel information requiring berthing. Specifically, vessel information includes vessel draft, vessel type, cargo attributes, remaining fuel, vessel malfunction status, waiting time, current position, required unloading equipment, and credit rating. Meteorological information includes wind speed, wave height, visibility, and precipitation probability. Berth status information includes the berth's maximum draft, the presence of unloading equipment, and remaining berthing time for vessels; the remaining berthing time is the remaining berthing time for vessels currently berthed at the set management period at the start of the period.

[0091] The initial population module is used to generate an initial population based on the number of ships that need to dock; each individual in the initial population represents the berth and docking time of each ship that needs to dock.

[0092] The fitness module is used to obtain the fitness of each individual based on management data, and determine whether there are any individuals with fitness values ​​less than a set threshold. If so, S6 is executed; otherwise, the individuals with fitness values ​​are sorted from largest to smallest and the top N individuals are selected as the parent individuals.

[0093] The expression for fitness is:

[0094] R=α×E+β×(1-C)+γ×(1-S)-K;

[0095] Where R is fitness, E is efficiency factor, C is carbon emission factor, S is safety factor, K is matching factor, α is efficiency coefficient, β is carbon emission coefficient, γ is safety factor, α+β+γ=1, E=1-y / Y, y is the average waiting time of B ships that need to berth, Y is the duration of the set management cycle, C is the normalized average of the carbon emissions of B ships that need to berth, the carbon emission is the sum of emissions during waiting time and emissions during travel time, and S is determined based on environmental risk value and trajectory conflict probability; the higher the environmental risk value, the larger S is, and the higher the trajectory conflict probability. The higher the probability, the larger S is. K is the sum of the matching degrees of B ships that need to berth. The matching degree is the sum of the draft matching degree and the required equipment matching degree. When the ship's draft is greater than the maximum draft of the berth, the draft matching degree is the set maximum value. When the ship's draft is less than or equal to the maximum draft of the berth, the draft matching degree is the set minimum value. When the ship's required unloading equipment is the same as the unloading equipment installed at the berth, the required equipment matching degree is the set minimum value. When the ship's required unloading equipment is different from the unloading equipment installed at the berth, the required equipment matching degree is the set maximum value.

[0096] The local optimization module is used to obtain the priority of each vessel requiring berthing based on its information. Based on each priority, it performs local optimization on N parent vessels to obtain optimized N parent vessels. Furthermore, the priorities are derived based on vessel type, cargo attributes, remaining fuel, malfunction status, and credit value.

[0097] The crossover and mutation module is used to perform crossover and mutation operations on the optimized N parent individuals to obtain a new population, which is then returned to the fitness module.

[0098] The scheme decision module is used to select the individual corresponding to the maximum fitness value as the ship berthing scheme within a set management period.

[0099] Preferably, the environmental risk value is determined based on meteorological information, and the process is as follows:

[0100] Set initial environmental risk values.

[0101] Divide the wind speed by the safe wind speed to get the wind speed ratio.

[0102] Divide the wave height by the safe wave height to obtain the wave height ratio.

[0103] The visibility ratio is obtained by dividing the safe visibility by the actual visibility.

[0104] Divide the probability of precipitation by the probability of safe precipitation to obtain the precipitation probability ratio.

[0105] The environmental risk value is the product of the initial environmental risk value, wind speed ratio, wave height ratio, visibility ratio, and precipitation probability ratio.

[0106] Based on the ship's current location, berth, and berthing time, the real-time location of the ships that need to berth within the set management cycle is obtained at each time point.

[0107] The real-time location of vessels already docked within a set management period is determined based on the remaining berthing time.

[0108] If the real-time position of the i-th vessel that needs to dock at time k coincides with the real-time position of the j-th vessel that needs to dock at time k, or the real-time position of the i-th vessel that needs to dock at time k coincides with the real-time position of the a-th already docked vessel at time k, then the probability of trajectory conflict is 1. If none of these exist, then the probability of trajectory conflict is 0.

[0109] The embodiments described above are merely preferred embodiments of the present invention and are not intended to limit the scope of the present invention. Various modifications and improvements made by those skilled in the art to the technical solutions of the present invention without departing from the spirit of the present invention should fall within the protection scope defined by the claims of the present invention.

Claims

1. A method for ship berthing decision based on local optimization genetic algorithm, characterized in that, It comprises: S1, obtaining management data in a set management period; the management data comprises berth state information, weather information and ship information of ships to be docked; S2, generating an initial population according to the number of ships to be docked; each individual in the initial population is the docking berth and docking time of each ship to be docked; S3, obtaining the fitness of each individual based on the management data, judging whether there is a fitness less than a set threshold, if there is, executing S6, if not, arranging the fitness from large to small and selecting the first N individuals as parent individuals; The expression of the fitness is: R = α × E + β × (1-C) + γ × (1-S) - K; Wherein, R is the fitness, E is the efficiency factor, C is the carbon emission factor, S is the safety factor, K is the matching degree factor, α is the efficiency coefficient, β is the carbon emission coefficient, γ is the safety coefficient, α+β+γ=1, E = 1-y / Y, y is the average waiting time of B ships to be docked, Y is the length of the set management period, C is the average value of the normalized value of the carbon emission of B ships to be docked, the carbon emission is the sum of the waiting time emission and the driving time emission, S is determined based on the environmental risk value and the trajectory conflict probability, the greater the environmental risk value, the greater S, the greater the trajectory conflict probability, the greater S, K is the sum of the matching degrees of B ships to be docked, the matching degree is the sum of the water depth matching degree and the demand equipment matching degree, when the ship water depth is greater than the maximum water depth of the berth, the water depth matching degree is a set maximum value, when the ship water depth is less than or equal to the maximum water depth of the berth, the water depth matching degree is a set minimum value, when the ship demand unloading equipment is the same as the berth installed unloading equipment, the demand equipment matching degree is a set minimum value, when the ship demand unloading equipment is different from the berth installed unloading equipment, the demand equipment matching degree is a set maximum value; S4, obtaining the priority of each ship to be docked based on each ship information, locally optimizing N parent individuals based on each priority, obtaining N optimized parent individuals; S5, performing cross operator processing and mutation operator processing on the N optimized parent individuals to obtain a new population, and returning to S3; S6, taking the individual corresponding to the maximum value of the fitness as the ship docking scheme in the set management period.

2. The method for ship berthing decision based on local optimization genetic algorithm according to claim 1, characterized in that, The ship information includes ship water depth, ship type, ship cargo attribute, ship fuel remaining amount, ship fault state, ship waiting time, ship current position, ship demand unloading equipment and ship credit value; the weather information includes wind speed, wave height, visibility and precipitation probability; the berth state information includes maximum water depth of the berth, berth installed unloading equipment and ship remaining docking time, the ship remaining docking is the remaining docking time of the current berth docked ship at the beginning of the set management period.

3. The method of claim 2, wherein, The priority is obtained based on the ship type, ship cargo attribute, ship fuel remaining amount, ship fault state and ship credit value.

4. The method of claim 2, wherein, The environmental risk value is determined based on the weather information, the process is as follows: Set an initial environmental risk value; Divide the wind speed by the safety wind speed to obtain the wind speed ratio; The wave height ratio is obtained by dividing the wave height by the safe wave height; The visibility ratio is obtained by dividing the safe visibility by the visibility; The precipitation probability ratio is obtained by dividing the safe precipitation probability by the precipitation probability; The product of the initial environmental risk value, the wind speed ratio, the wave height ratio, the visibility ratio, and the precipitation probability ratio is taken as the environmental risk value.

5. The method of claim 2, wherein, Real-time positions of the vessels to be docked at each time point in the set management period are obtained based on the current positions of the vessels, the docking berths, and the docking times; Real-time positions of the vessels that have docked at each time point in the set management period are determined based on the remaining docking durations of the vessels; If the real-time position of the ith vessel to be docked at the kth time point coincides with the real-time position of the jth vessel to be docked at the kth time point or coincides with the real-time position of the ath vessel that has docked at the kth time point, the trajectory conflict probability is 1, and if not, the trajectory conflict probability is 0.

6. A ship berthing decision system based on a local optimization genetic algorithm, characterized by, It comprises: A data acquisition module is configured to acquire management data in a set management period; the management data comprises berth state information, weather information, and vessel information of vessels to be docked; An initial population module is configured to generate an initial population according to the number of vessels to be docked; each individual in the initial population is a docking berth and a docking time of each vessel to be docked; An adaptability module is configured to acquire adaptabilities of each individual based on the management data, to determine whether there is an adaptability less than a set threshold, and if so, to perform S6, and if not, to arrange the adaptabilities in descending order and select the first N individuals as parent individuals; The expression of the adaptability is: R = α × E + β × (1-C) + γ × (1-S) - K; wherein, R is the adaptability, E is an efficiency factor, C is a carbon emission factor, S is a safety factor, K is a matching degree factor, α is an efficiency coefficient, β is a carbon emission coefficient, γ is a safety coefficient, α + β + γ = 1, E = 1-y / Y, y is an average waiting time of B vessels to be docked, Y is a duration of the set management period, C is an average value of normalized values of carbon emissions of the B vessels to be docked, the carbon emission is a sum of a waiting time emission and a driving time emission, S is determined based on an environmental risk value and a trajectory conflict probability, the greater the environmental risk value, the greater S, the greater the trajectory conflict probability, the greater S, K is a sum of matching degrees of the B vessels to be docked, the matching degree is a sum of a draft matching degree and a demand equipment matching degree, when the draft of the vessel is greater than the maximum draft of the berth, the draft matching degree is a set maximum value, when the draft of the vessel is less than or equal to the maximum draft of the berth, the draft matching degree is a set minimum value, when the demand unloading equipment of the vessel is the same as the installed unloading equipment of the berth, the demand equipment matching degree is a set minimum value, when the demand unloading equipment of the vessel is different from the installed unloading equipment of the berth, the demand equipment matching degree is a set maximum value; A local optimization module is configured to acquire priorities of each vessel to be docked based on the vessel information, to locally optimize the N parent individuals based on the priorities, and to obtain optimized N parent individuals. a crossover and mutation module, configured to perform crossover and mutation on the N parent individuals after optimization to obtain a new population and return to the fitness module; a scheme decision module, configured to take the individual corresponding to the maximum fitness as the ship berthing scheme in a set management period.

7. The local optimization genetic algorithm based ship berthing decision system according to claim 6, characterized in that, The ship information includes ship draft, ship type, ship cargo attribute, ship fuel remaining amount, ship fault state, ship waiting time, ship current position, ship required unloading equipment and ship credit value; the weather information includes wind speed, wave height, visibility and precipitation probability; the berth state information includes berth maximum draft, berth installed unloading equipment and ship remaining berthing time, the ship remaining berthing time being the remaining berthing time of the ship currently berthing at the berth at the beginning of the set management period.

8. The local optimization genetic algorithm based ship berthing decision system according to claim 7, characterized in that, The priority is obtained based on the ship type, the ship cargo attribute, the ship fuel remaining amount, the ship fault state and the ship credit value.

9. The local optimization genetic algorithm based ship berthing decision system according to claim 7, characterized in that, The environmental risk value is determined based on the weather information, in the following process: an initial environmental risk value is set; a wind speed ratio value is obtained by dividing the wind speed by a safe wind speed; a wave height ratio value is obtained by dividing the wave height by a safe wave height; a visibility ratio value is obtained by dividing the safe visibility by the visibility; a precipitation probability ratio value is obtained by dividing the safe precipitation probability by the precipitation probability; a product of the initial environmental risk value, the wind speed ratio value, the wave height ratio value, the visibility ratio value and the precipitation probability ratio value is taken as the environmental risk value.

10. The local optimization genetic algorithm based ship berthing decision system according to claim 7, characterized in that, a real-time position of a required berthing ship at each time point in the set management period is obtained based on the ship current position, the berthing berth and the berthing time; a real-time position of a berthing ship at each time point in the set management period is determined based on the ship remaining berthing time; if the real-time position of the i-th required berthing ship at the k-th time point coincides with the real-time position of the j-th required berthing ship at the k-th time point or the real-time position of the i-th required berthing ship at the k-th time point coincides with the real-time position of the a-th berthing ship at the k-th time point, the trajectory conflict probability is 1, and if not, the trajectory conflict probability is 0.

Images