Particle swarm optimization algorithm and geographic information-based route optimization method and system

Abstract

The invention relates to the field of ship route optimization, discloses a route optimization method and system based on a particle swarm optimization algorithm and geographic information, and is used for providing comprehensive decision support covering route planning, energy scheduling, risk control and carbon emission management for shipping enterprises. Comprising the steps that a commercial decision basic data set integrating geographic information, ship multi-energy system configuration and marine environment forecast data is constructed, and a commercial fractional order parameter set is generated; environment uncertainty is modeled by using a stochastic differential equation, a multi-energy system operation cost function is integrated, and a commercial cost prediction model capable of outputting operation cost probability distribution is formed; and carrying out risk-aware business strategy optimization based on a particle swarm optimization algorithm, and finally generating an executable business decision proposal containing a detailed route trajectory, a speed plan and a multi-energy equipment scheduling strategy. According to the invention, collaborative optimization of route planning and energy management is realized, and economy, safety and environmental protection of ship operation are significantly improved.

Description

Technical Field

[0001] This invention relates to the field of ship route optimization, and in particular to a route optimization method and system based on particle swarm optimization algorithm and geographic information. Background Technology

[0002] In shipping operations, route planning is a crucial step, directly affecting numerous factors such as ship sailing time, fuel consumption, operating costs, and carbon emissions.

[0003] Traditional route planning methods are primarily based on simple geographical information and empirical rules, considering only fixed geographical constraints and basic marine environmental conditions, such as fixed routes and average wind, wave, and current conditions. However, the actual marine environment is complex and variable. Environmental factors such as wind, waves, currents, and ice floes exhibit a high degree of randomness and uncertainty, which significantly impact a ship's navigation performance and energy consumption. Furthermore, modern ships are typically equipped with multiple energy systems, including diesel generators, fuel cells, and battery packs. The configuration and operating strategies of different energy systems also have a significant impact on a ship's operating costs and environmental performance.

[0004] Existing route optimization technologies have many shortcomings in handling these complex problems: The high degree of reliance on personnel decision-making means that route planners and energy managers often work independently and lack collaborative decision-making tools, making it difficult to leverage the combined advantages of multi-energy vessels. The lack of risk management tools means that managers lack quantitative tools to assess and mitigate business risks such as fuel price fluctuations and delays caused by severe weather. The existing system lacks dynamic adjustment capabilities and is unable to dynamically adjust business strategies based on real-time environmental changes and market conditions.

[0005] Therefore, we propose a route optimization method and system based on particle swarm optimization algorithm and geographic information to solve the above problems. Summary of the Invention

[0006] This invention provides a method and system for route optimization based on particle swarm optimization algorithm and geographic information, which provides shipping companies with comprehensive decision support covering route planning, energy scheduling, risk control and carbon emission management.

[0007] The first aspect of this invention provides a route optimization method based on particle swarm optimization (PSO) algorithm and geographic information. This method includes: integrating geographic information data, ship multi-energy system configuration data, and marine environmental forecast data to form a business decision-making dataset; processing the business decision-making dataset using a fractional-order system identification algorithm to establish a historical energy consumption dependency model for the ship's power system, generating a commercial fractional-order parameter set; modeling environmental uncertainties using stochastic differential equations, combining the commercial fractional-order parameter set to establish a propulsion load prediction model, and integrating multi-energy system operating cost functions to form a business cost prediction model; optimizing a risk-aware business strategy based on the PSO algorithm, using the business cost prediction model to perform multi-scenario Monte Carlo simulations on candidate routes, calculating the expected operating cost and conditional value of risk for each option, and iteratively generating the optimal business strategy by minimizing the risk-adjusted total cost objective function, including: centralizing the business decision-making dataset... The process involves generating an initial set of route options from digital nautical charts and randomly assigning speed control parameters to each option, forming an initial commercial strategy population. Each option in the initial commercial strategy population is then input into a commercial cost prediction model, and large-scale stochastic simulations are performed using a set of commercial navigation environment scenarios. The resulting data outputs the operating cost distribution of each option under different environmental scenarios. Based on the operating cost distribution data of each option, the expected operating cost and conditional value of risk (CVaR) are calculated, forming a fitness evaluation function with the objective of minimizing the risk-adjusted total cost, yielding the fitness evaluation results. The spatial path and speed control parameters of each option in the commercial strategy population are updated according to the fitness evaluation results, and the quality of the options is gradually optimized through multiple iterations to generate the commercial strategy population. The option with the minimum risk-adjusted total cost is selected from the commercial strategy population to generate the optimal commercial strategy. Finally, the final commercial decision-making plan is analyzed and output, including detailed route trajectories, speed plans, and supporting multi-energy equipment scheduling strategies, forming a navigation plan and energy management decision recommendation.

[0008] Optionally, in the first implementation of the first aspect of the present invention, the method includes: processing geographic information data, extracting coastline, water depth data, and information on prohibited navigation areas, and generating a digital nautical chart; processing ship multi-energy system configuration data, analyzing the performance parameters and operational constraints of diesel generators, fuel cells, battery packs, thermal management systems, and seawater desalination devices, and generating a ship energy commercial configuration file; processing marine environmental forecast data, using stochastic process modeling methods to perform probability distribution fitting on wind field, wave field, current field, and ice density data, and generating a commercial navigation environment scenario set; and integrating the digital nautical chart, the ship energy commercial configuration file, and the commercial navigation environment scenario set to construct a complete commercial decision-making foundation dataset.

[0009] Optionally, in a second implementation of the first aspect of the present invention, a propulsion load prediction model is established by combining the aforementioned commercial fractional-order parameter set, wherein the load power is : ; Where V is the speed and D is the draft. Let C(t) be the wave height, C(t) be the current velocity, α be the linear speed influence coefficient, β be the nonlinear speed influence coefficient, η and κ be the environmental coupling coefficients, K be the cyclotron exponent, and T be the follower exponent.

[0010] Optionally, in a third implementation of the first aspect of the present invention, the method includes: preprocessing the commercial decision-making dataset, cleaning and standardizing the ship speed sequence, main engine power output sequence and corresponding environmental data sequence to generate a standardized time series dataset; processing the standardized time series dataset using a fractional-order system identification algorithm, and identifying the optimal fractional-order α characterizing the memory effect of the ship's power system by minimizing the model prediction error criterion; fitting the energy consumption response characteristics of the ship's power system based on the optimal fractional-order α to calculate the efficiency parameter κ; and integrating the optimal fractional-order α and the efficiency parameter κ to generate a commercial fractional-order power parameter set.

[0011] Optionally, in the fourth implementation of the first aspect of the present invention, the method includes: establishing stochastic differential equations for each environmental parameter to generate a continuous-time stochastic evolution model of the environmental parameters; combining a commercial fractional-order parameter set, inputting the stochastic evolution model of the environmental parameters into the fractional-order ship dynamics equations to establish a stochastic prediction model of propulsion power demand; integrating a multi-energy system operating cost function to convert the probability distribution of propulsion power demand into a scheduling scheme for multi-energy equipment, and calculating the fuel consumption cost, equipment operating cost, and external energy procurement cost under each scheme; and constructing a commercial cost prediction model based on all cost elements.

[0012] Optionally, in the fifth implementation of the first aspect of the present invention, the method includes: parsing the spatial position sequence of the optimal business strategy scheme and generating a route trajectory file based on the spatial reference system of the digital nautical chart; extracting the time speed parameters of the optimal business strategy scheme and generating a time-synchronized speed plan table in combination with the route trajectory; generating a multi-energy equipment operation plan based on the energy scheduling parameters in the optimal business strategy scheme, including diesel generator load allocation, fuel cell start-up and shutdown sequence, battery charging and discharging strategy, and freshwater production plan, to form an energy management scheme; integrating the route trajectory file, speed plan table, and energy management scheme, adding business decision metadata, and generating a business decision recommendation.

[0013] Optionally, in the sixth implementation of the first aspect of the present invention, the method further includes: generating a multi-dimensional optimization objective weight set containing economic benefits, navigation safety, and environmental protection based on the ship energy commercial configuration files and environmental probability forecast data in the commercial decision-making basic dataset; dynamically adjusting the navigation constraints in the digital nautical chart according to the real-time environmental change data in the commercial navigation environment scenario set to generate a navigation constraint rule set; using a commercial cost prediction model to perform sensitivity analysis on the optimal commercial strategy scheme and generate a robustness assessment report; recording the evolution path and fitness changes of each generation of commercial strategy populations during the optimization process to form a decision-making process traceability archive; comparing and analyzing the actual navigation data with the prediction results, and adjusting the commercial fractional-order power parameter set and the commercial cost prediction model based on the deviation results to form a model parameter update mechanism.

[0014] The second aspect of this invention provides a route optimization system based on particle swarm optimization (PSO) algorithm and geographic information. The system includes: a construction module for integrating geographic information data, ship multi-energy system configuration data, and marine environmental forecast data to form a business decision-making dataset; a quantification module for processing the business decision-making dataset based on a fractional-order system identification algorithm, establishing a historical energy consumption dependency model for the ship's power system, and generating a commercial fractional-order parameter set; a prediction module for modeling environmental uncertainties using stochastic differential equations, combining the commercial fractional-order parameter set to establish a propulsion load prediction model, and integrating a multi-energy system operating cost function to form a business cost prediction model; and a setting module for risk-aware business strategy optimization based on the PSO algorithm, using the business cost prediction model to perform multi-scenario Monte Carlo simulations on candidate routes, calculating the expected operating cost and conditional value of risk for each option, and iteratively generating the optimal business strategy by minimizing the risk-adjusted total cost objective function. The initial set of route options is generated from the digitized nautical charts in the business decision-making dataset, and speed control parameters are randomly assigned to each option to form an initial business strategy population. Each option in the initial business strategy population is input into a business cost prediction model, and large-scale stochastic simulations are performed using a set of business navigation environment scenarios to output the operating cost distribution data of each option under different environmental scenarios. Based on the operating cost distribution data of each option, the expected operating cost and conditional value of risk (CVaR) are calculated to form a fitness evaluation function with the objective of minimizing the risk-adjusted total cost, yielding the fitness evaluation results. The spatial path and speed control parameters of each option in the business strategy population are updated according to the fitness evaluation results, and the quality of the options is gradually optimized through multiple iterations to generate the business strategy population. The option with the minimum risk-adjusted total cost is selected from the business strategy population to generate the optimal business strategy option. The decision module is used to parse and output the final business decision option, including detailed route trajectory, speed plan, and supporting multi-energy equipment scheduling strategies, forming a navigation plan and energy management decision recommendation.

[0015] The mechanism of this invention is as follows: it elevates route optimization from a traditional path search problem to a business decision-making process that comprehensively considers operational economy, navigation safety and environmental adaptability. It achieves multi-objective collaborative optimization through particle swarm optimization and finally outputs a business decision-making proposal containing complete risk quantification indicators and multiple energy scheduling schemes, providing shipping companies with a data-driven intelligent management tool. Beneficial effects: By identifying the historical dependence characteristics of ship power systems through fractional-order systems, the shortcomings of traditional integer-order models in predicting accuracy under varying operating conditions are overcome, providing a more accurate input basis for operating cost calculation and making business decisions more scientific and reliable. By using stochastic differential equations to model environmental uncertainty and combining it with conditional value of risk tools in the financial field, it is possible to quantify the business risks brought about by extreme weather and market price fluctuations, providing shipping companies with a forward-looking risk avoidance tool and supporting the formulation of risk-adjusted optimal business strategies. A cost mapping model for an electricity-heat-hydrogen-freshwater multi-energy system was established, which enabled the coordinated optimization of navigation strategy and energy scheduling, significantly improving energy utilization efficiency and reducing overall operating costs. The output includes a complete decision recommendation document containing route trajectory, speed plan and energy scheduling scheme, which not only meets the automatic execution needs of the ship navigation management system, but also provides managers with human-readable business decision-making basis; Through closed-loop feedback control and adaptive parameter updates, the system can continuously learn and improve model accuracy, ensuring decision-making quality in long-term operation and providing shipping companies with continuously optimized business value. Attached Figure Description

[0016] Figure 1 This is a schematic diagram of an embodiment of the route optimization method based on particle swarm optimization algorithm and geographic information in this invention. Figure 2 This is a schematic diagram of another embodiment of the route optimization method based on particle swarm optimization algorithm and geographic information in this invention; Figure 3 This is a schematic diagram of an embodiment of the flight route optimization system based on particle swarm optimization algorithm and geographic information in this invention. Figure 4 This is a schematic diagram of an embodiment of a flight route optimization device based on particle swarm optimization algorithm and geographic information in this invention. Detailed Implementation

[0017] This invention provides a method and system for route optimization based on particle swarm optimization algorithm and geographic information, offering shipping companies comprehensive decision support covering route planning, energy scheduling, risk control, and carbon emission management. The terms "first," "second," "third," "fourth," etc. (if present) in the specification, claims, and accompanying drawings are used to distinguish similar objects and are not necessarily used to describe a specific order or sequence. It should be understood that such data can be interchanged where appropriate so that the embodiments described herein can be implemented in orders other than those illustrated or described herein. Furthermore, the terms "comprising" or "having" and any variations thereof are intended to cover a non-exclusive inclusion; for example, a process, method, system, product, or apparatus that comprises a series of steps or units is not necessarily limited to those explicitly listed, but may include other steps or units not explicitly listed or inherent to such processes, methods, products, or apparatus.

[0018] For ease of understanding, the specific process of the embodiments of the present invention is described below. Please refer to [link / reference]. Figure 1 One embodiment of the route optimization method based on particle swarm optimization algorithm and geographic information in this invention includes: 101. Construct a multi-dimensional commercial navigation database, integrating geographic information data, ship multi-energy system configuration data, and marine environmental forecast data to form a basic dataset for commercial decision-making that includes navigable area constraints, equipment performance parameters, and environmental probability forecasts; It is understood that the executing entity of this invention can be a route optimization system based on particle swarm optimization algorithm and geographic information, or it can be a terminal or a server; the specific implementation is not limited here. This embodiment of the invention will be described using a server as an example.

[0019] It should be noted that the geographic information data is extracted from maritime charts and satellite remote sensing data, including navigable area constraints such as port coordinates (Shanghai Port: 31.23°N, 121.47°E), channel depth (minimum depth of 12.5 meters in the Yangtze River estuary), prohibited navigation zones (coordinates of military areas), and obstacle locations (reefs, oil platforms). This data is stored in the form of vector layers and raster elevation models.

[0020] Ship multi-energy system configuration data: Equipment performance parameters extracted from the ship management system: Main engine power: 5,000kW, fuel consumption rate (180g / kWh); Battery pack capacity: 1,000kWh, charge / discharge efficiency 95%; Solar panel area: 200m² 2 The average daily power generation is 300 kWh; equipment operation constraints: minimum load rate of main unit is 30%, maximum depth of battery discharge is 80%.

[0021] Marine environmental forecast data: Integrating probabilistic forecast data from meteorological and oceanographic agencies, including wind field (probability distribution of wind speed in the next 72 hours: 70% probability of wind speed ≤10 knots), wave height (50% probability of wave height ≤2 meters), ocean current speed (average current speed in East Asian waters 0.5-1.2 knots), and sea ice coverage (probability of sea ice in Bohai Bay ≥20% in winter).

[0022] Database structure design: Data is organized using a star schema: Fact table: stores core business metrics, navigation event fact table, containing fields: navigation ID, energy consumption (kWh), navigation distance (nautical miles), cost indicators (USD).

[0023] Dimension Tables: Provide contextual information: Geographic Dimension Table: Includes port ID, latitude and longitude, water depth, and area type (navigable / restricted); Ship Equipment Dimension Table: Records equipment ID, power, efficiency, and constraints; Environmental Dimension Table: Includes timestamps, wind, wave, and current probability data, and ice floe risk level.

[0024] By linking the fact table and dimension tables with foreign keys, efficient multi-dimensional query analysis is supported.

[0025] Data Integration and Processing: Through an ETL (Extract, Transform, Load) process: Extraction: Raw data is extracted from APIs (Meteorological Bureau APIs), sensors (ship IoT devices), and databases (port management systems). Transformation: Inconsistent data is cleaned (removing unnavigable points in water depths <10 meters), a unified coordinate system (WGS84) is established, units are standardized (unified units are speed units), and derived indicators (energy efficiency ratio) are calculated. Loading: The processed data is loaded into database tables, and indexes are created to optimize query performance.

[0026] Output base dataset: The final generated structured dataset includes: Navigable area constraints: the set of coordinates of the no-navigation zones marked in the geographic dimension table; Equipment performance parameters: the power and efficiency values ​​of the main engine and batteries in the ship equipment dimension table; Environmental probability forecast: the probability distribution values ​​of wind, waves and currents in the environmental dimension table.

[0027] 102. Based on the fractional-order system identification algorithm, process the basic dataset of business decision-making, establish a historical energy consumption dependency model of ship power system, and generate a commercial fractional-order parameter set characterizing ship power characteristics to quantify energy consumption characteristics under different navigation conditions. It should be noted that a fractional-order system identification algorithm is used to process historical navigation data, constructing a historical energy consumption dependency model for the ship's propulsion system, and generating a commercially viable fractional-order parameter set for quantifying energy consumption characteristics under different navigation conditions. The following explanation uses actual data from a 100,000-ton container ship as an example.

[0028] Historical navigation data for 26 voyages over the past year was collected, including: Navigation status data: speed (knots), draft (meters), heel angle (degrees), main engine speed (RPM), and actual rudder angle (degrees). Energy consumption data: main engine fuel consumption per hour (kg / hour), auxiliary engine power generation (kWh). Environmental data: wind speed (m / s), wave height (meters), and ocean current speed (knots). The data was cleaned to remove sensor anomalies (sudden drops in speed to 0 or negative fuel consumption), and the time frequency was standardized to one sample point per minute.

[0029] Using the nonlinear Nomoto model as the basic structure, its fractional form is represented as: Energy output = K * (α * speed + β * speed) 3 ) / T+γ* Draft correction term; The parameters to be identified are a fractional parameter set [K,T,α,β,γ], where K is the cyclicity exponent, T is the following exponent, α and β are nonlinear coefficients, and γ is the draft influence factor.

[0030] By processing innovation (prediction error) using a sine function, the identification accuracy under small sample data is enhanced. Using 26 sets of data as input, an improved least squares method is used for iterative solution to minimize the error between the model output and the measured fuel consumption.

[0031] After identification, the following dynamic characteristic parameters of the ship were obtained: K=0.86 (characterizing the ship's turning sensitivity); T=216.5s (characterizing heading stability); α=1.24 (linear speed influence coefficient); β=-0.03 (nonlinear speed influence coefficient); γ=0.57 (draft change correction coefficient). This parameter set quantifies energy consumption characteristics under different navigation conditions: High-speed condition (20 knots): β dominates nonlinear energy consumption growth, with fuel consumption per unit distance increasing by 38% compared to 15 knots. Heavy-load condition (12-meter draft): Energy consumption after γ correction increases by 22% compared to unloaded condition (8-meter draft). Environmental interference: Fractional calculus is used to capture the lag effects of wind and waves in historical data during identification; energy consumption fluctuations with a wave height of 2 meters have a delay response of approximately 3 minutes.

[0032] By embedding parameter sets into the ship energy consumption prediction system and inputting real-time speed, draft, and environmental data, a dynamic energy consumption curve can be output. Verification has shown that this model reduces prediction error by 15% compared to traditional integer-order models, making it particularly suitable for scenarios with small sample data. Ship owners can use this to assess the fuel economy of different routes or optimize speed to achieve energy savings (reducing speed by 1 knot can reduce fuel consumption by 5-8%).

[0033] 103. Using stochastic differential equations to model environmental uncertainties, and combining the aforementioned commercial fractional-order parameter set, a propulsion load prediction model considering the random changes of wind, waves, currents, and floating ice is established. Furthermore, the operating cost function of multiple energy systems is integrated to form a commercial cost prediction model capable of outputting the probability distribution of operating costs. It should be noted that stochastic differential equations are used to model environmental uncertainties. Combined with a fractional-order parameter set generated in 102 steps, a propulsion load prediction model considering the random variations of wind, waves, currents, and floating ice is established. Furthermore, the operating cost function of multiple energy systems is integrated to form a commercial cost prediction model that can output the probability distribution of operating costs. The following uses actual ship data from a 100,000-ton container ship as an example to illustrate the specific implementation.

[0034] Stochastic differential equations (SDEs) are used to describe the stochastic variations of wind, waves, currents, and floating ice. Taking wave height as an example, a mean regression model is constructed: ; Where H(t) is the wave height at time t (meters), μ is the long-term average wave height (2.5 meters), θ is the regression rate (0.35), σ is the volatility (0.18), and dW(t) is the Wiener process increment, simulating random disturbances. Similarly, SDE models for wind speed (mean 12 m / s), current velocity (mean 0.8 knots), and ice cover (mean 5%) are established. 1000 sets of environmental scenario time series (1-hour time step, 72-hour forecast duration) are generated using the Euler-Maruyama numerical method to characterize the probability distribution of future environmental conditions.

[0035] Based on the fractional-order parameter set generated in 102 steps (speed energy consumption coefficients α=1.24, β=-0.03), a dynamic response equation for propulsion load is established. Load power. (Unit: kW) Calculated as follows: ; Where V is the speed (knots), D is the draft (meters), H(t) is the wave height, and C(t) is the current velocity; η and κ are environmental coupling coefficients (fitted from historical data as η=85.6, κ=72.3). Inputting the environmental scenario generated by SDE into this model outputs 1000 sets of propulsion load probability distributions. Under the conditions of a speed of 20 knots and a draft of 12 meters, the load baseline is 4200kW, and the load fluctuation range is ±8% at a wave height of 2 meters.

[0036] Operating cost function for integrated marine multi-energy systems (main engine, battery, fuel cell): Main engine cost: Fuel consumption cost (fuel price $650 / ton), calculated as follows: (Unit: USD); Battery cost: Charge and discharge loss cost (Aging cost of USD 0.12 / kWh); Fuel cell cost: Hydrogen consumption cost (hydrogen price $6 / kg). The total cost function is a weighted sum of the costs of each energy source: For each load scenario, the operating cost is calculated, and a cost probability distribution is generated.

[0037] Integrating propulsion load forecasting and cost functions, the probability distribution of operating costs is output: Expected operating cost: $12,500 for a 72-hour flight; Cost risk range: 90% confidence interval of [$11,800, $13,300]; Probability distribution form: Based on Monte Carlo simulation, a cumulative cost distribution function (CDF) is generated, which can be directly used for risk decision-making (conditional value at risk calculation).

[0038] This model captures uncertainties through environmental SDE and combines dynamic characteristic parameters with multiple energy costs to achieve probabilistic prediction of operating costs, providing input for subsequent risk optimization.

[0039] 104. Based on the particle swarm optimization algorithm, optimize the risk perception business strategy, use the business cost prediction model to perform multi-scenario Monte Carlo simulation on the candidate routes, calculate the expected operating cost and conditional risk value of each scheme, and iteratively generate the optimal business strategy scheme by minimizing the risk-adjusted total cost objective function. It should be noted that the risk-aware business strategy optimization is based on the Particle Swarm Optimization (PSO) algorithm. A 103-step business cost prediction model is used to conduct multi-scenario Monte Carlo simulations on candidate routes to calculate the expected operating cost and conditional value at risk (CVaR) of each option. The optimal business strategy is generated iteratively by minimizing the risk-adjusted total cost objective function. The following example illustrates the specific implementation using a 100,000-ton container ship (100,000 deadweight tons, 5,000 kW main engine power, 1,000 kWh battery capacity) on an Asia-Europe route (Shanghai Port to Rotterdam Port).

[0040] Initialize the particle swarm and candidate route schemes: Particle encoding: Each particle represents a candidate route scheme, encoded as a multi-dimensional vector, including the route path point sequence (longitude / latitude coordinates), speed plan (unit: knots), and energy equipment scheduling ratio (output ratio of main engine, battery, and fuel cell): Particle A vector: Path points [122°E / 31°N, 135°E / 35°N, ..., 4°E / 52°N]; Speed ​​[18, 20, 19, ...] knots; Energy ratio [Main engine 70%, battery 20%, fuel cell 10%].

[0041] Population size: Initialize 50 particles (i.e. 50 candidate schemes). The position and velocity of each particle are randomly generated, but must meet the constraints (path points must avoid no-fly zones, and the total energy ratio must be 100%).

[0042] Multi-scenario Monte Carlo Simulation: Utilizing a 103-step business cost prediction model, 1,000 Monte Carlo simulations are performed on each candidate scenario (particle) to simulate random environmental changes such as wind, waves, currents, and floating ice (environmental scenarios generated based on stochastic differential equations). Each simulation outputs an operating cost value (in US dollars).

[0043] Calculate expected operating costs: Calculate the average cost of 1,000 simulations for each scenario: Expected cost for particle A: $12,500 (total cost of the journey). Calculate Conditional Value at Risk (CVaR): Calculate the tail average loss of the loss distribution (i.e., the average cost of the worst-case 5%) at a 95% confidence level: CVaR for particle A: $13,800 (meaning the average cost is $13,800 in the worst-case 5%).

[0044] Risk-adjusted objective function optimization: Objective function design: Minimize the risk-adjusted total cost. Where β is the risk aversion coefficient (β=1.2, set by the shipowner): the objective function value of particle A. Dollar.

[0045] PSO Iterative Optimization: Each particle updates its velocity and position (path, speed, and energy ratio) based on its own historical best position (pBest) and the group's global best position (gBest). An adaptive weighting mechanism is introduced: the inertia weight decreases linearly from 0.9 to 0.4 to balance exploration and development capabilities. Constraint handling: Particles that do not meet navigable area constraints undergo position correction (adjusting their waypoints to the nearest safe waters), and schemes that violate energy constraints are penalized (the objective function value increases by 10%). Termination condition: Convergence after 100 iterations, or the rate of change of the objective function is less than 0.1%.

[0046] The final globally optimal particle (gBest) corresponds to the lowest risk-adjusted cost scheme: Optimal route: Shanghai Port → East China Sea → South of Japan → North Pacific → Southern Bering Sea → Norwegian Sea → Rotterdam Port (total distance approximately 10,200 nautical miles); Speed ​​plan: Average speed 18.5 knots (reduced to 15 knots in high sea states, increased to 20 knots in calm seas); Energy scheduling: Main engine dominant (average 75% power), batteries provide auxiliary power during port entry and exit (15%), fuel cells provide stable power supply in the medium term (10%); Expected cost: US$11,800, CVaR: US$12,900, risk-adjusted total cost. Dollar.

[0047] 105. Analyze and output the final business decision plan, including detailed route trajectory, speed plan and supporting multi-energy equipment scheduling strategy, to form an executable flight plan and energy management decision recommendation.

[0048] It should be noted that the following example uses a 100,000-ton container ship (100,000 deadweight tons, 5,000 kW main engine power, 1,000 kWh battery capacity, and 800 kW fuel cell power) on the Shanghai Port to Rotterdam route, and the final solution is generated based on 104 optimization steps.

[0049] Route coordinate sequence: Starting from Shanghai Port (latitude and longitude: 31.23°N, 121.47°E), the route passes through 6 key waypoints (East China Sea turning point: 29.5°N, 126.8°E; Southern Bering Sea point: 58.5°N, 178.0°E; Norwegian Sea point: 65.0°N, 5.0°E), finally arriving at Rotterdam Port (51.92°N, 4.48°E). The total voyage is approximately 10,200 nautical miles, avoiding restricted areas (military zones, ice zones) and shallow waters (depth < 15 meters) throughout.

[0050] Navigation instructions: Generate navigation instructions based on waypoint sequence, travel 220 nautical miles on a heading of 285° to turn point P1, then travel 1,800 nautical miles on a heading of 350°.

[0051] Segmented speed optimization: Adjusting speed based on marine environment and cost risks. East China Sea to Pacific (0-2,000 nautical miles): Average speed 20 knots (relatively calm, economical speed); North Pacific (2,000-6,000 nautical miles): Reduce speed to 18 knots (to cope with wave heights of 2-3 meters); Bering Sea to Norwegian Sea (6,000-8,500 nautical miles): Increase speed to 19 knots (with currents, saving energy); European coastal area (8,500-10,200 nautical miles): Reduce speed to 16 knots (in areas with heavy traffic, ensuring safety). Dynamic adjustment mechanism: Receives real-time marine forecast data; if wave height exceeds 3.5 meters, automatically triggers a speed reduction to 15 knots.

[0052] Main engine and auxiliary engine configuration: The main engine is responsible for propulsion (average power share of 75%), and the fuel consumption rate is controlled at 180g / kWh; the auxiliary engine starts during port entry and exit periods (power share of 10%) to provide auxiliary power.

[0053] Coordination of new energy equipment: Battery packs: providing power during low-speed maneuvers in port (15% of capacity, 95% charge / discharge efficiency); Fuel cells: supplementing power during cruising (10% of capacity, 6 kg / h hydrogen consumption). Cost control targets: expected fuel consumption of 1,200 tons throughout the trip, with battery cycle count not exceeding 3 times per day.

[0054] Structured Output Documents: Flight Plan Section: Includes route map (vector map with path markings), flight timetable (estimated total flight time 23 days 4 hours), and alternative routes (avoiding typhoon paths). Energy Management Section: Detailed equipment scheduling table (specifying the output power of the main engine, battery, and fuel cell by time interval), Days 1-5: Main engine power 3,800kW, battery charged to 90% capacity. Risk Warning: High-risk areas (times with ≥20% probability of Arctic sea ice) and corresponding countermeasures (early activation of auxiliary engines to increase power).

[0055] Execution format: Generates a navigation manual in PDF format and an energy scheduling timetable in CSV format, which can be directly imported into the ship management system for automated execution.

[0056] In this embodiment of the invention, foreign key associations between the fact table and the dimension table support efficient multi-dimensional query analysis, making data storage and retrieval more rational, improving data processing efficiency, and facilitating the rapid acquisition of required information. By processing historical navigation data, a historical energy consumption dependency model of the ship's power system is constructed, generating a commercial fractional-order parameter set, which can quantify the energy consumption characteristics under different navigation conditions. Stochastic differential equations are used to describe the random changes of wind, waves, currents, and floating ice, and multiple sets of environmental scenario time series are generated through numerical methods to characterize the probability distribution of future environmental conditions, fully considering environmental uncertainties and providing a more comprehensive environmental factor consideration for cost prediction. The candidate routes are subjected to multi-scenario Monte Carlo tree optimization based on the particle swarm optimization algorithm. Carlow simulation calculates the expected operating costs and conditional value of risk for each option. By minimizing the risk-adjusted total cost objective function, it iteratively generates the optimal business strategy. This fully considers risk factors, minimizes risk-adjusted costs, and improves the scientific and rational nature of decision-making. It includes detailed route trajectories, speed plans, and corresponding multi-energy equipment scheduling strategies, forming directly executable navigation plans and energy management decision recommendations. This provides clear guidance for actual ship navigation and improves the operability of decisions. It generates PDF navigation manuals and CSV energy scheduling timelines, which can be directly imported into ship management systems for automated execution, facilitating the application and implementation of decision-making schemes and improving work efficiency.

[0057] Please see Figure 2 Another embodiment of the route optimization method based on particle swarm optimization algorithm and geographic information in this invention includes: 201. Construct a multi-dimensional commercial navigation database, integrating geographic information data, ship multi-energy system configuration data, and marine environmental forecast data to form a basic dataset for commercial decision-making that includes navigable area constraints, equipment performance parameters, and environmental probability forecasts; Specifically, the process involves processing geographic information data, extracting coastline, water depth, and restricted navigation area information, and generating a digital nautical chart with spatial grid coding. This digital nautical chart serves as the geographic constraint basis for subsequent optimization processes. It also involves processing ship multi-energy system configuration data, analyzing the performance parameters and operational constraints of diesel generators, fuel cells, battery packs, thermal management systems, and seawater desalination devices, and generating a standardized ship energy commercial configuration profile, which serves as input parameters for multi-energy cost calculations. Furthermore, it involves processing marine environmental forecast data, using stochastic process modeling to fit probability distributions of wind field, wave field, current field, and ice floe density data, generating a commercial navigation environment scenario set containing multiple possible environmental evolution sequences. This scenario set serves as the input source for subsequent Monte Carlo simulations. Finally, it integrates the aforementioned digital nautical chart, ship energy commercial configuration profile, and commercial navigation environment scenario set to construct a complete commercial decision-making foundation dataset. This dataset provides structured input for subsequent fractional-order parameter identification and cost prediction.

[0058] It should be noted that, in order to achieve route optimization, a multi-dimensional commercial navigation database must first be constructed. This database integrates three core types of data: geographic information data, ship multi-energy system configuration data, and marine environmental forecast data.

[0059] In geographic information processing, taking a container ship sailing in the East China Sea as an example, coastline coordinates, water depth data (gridized water depth, resolution 0.1°×0.1°, water depth range 5-200 meters), and prohibited navigation areas (military zones, ecological protection zones) are extracted from the Electronic Chart System (ECDIS). This data is encoded into a spatial grid (latitude and longitude grid encoding WGS-84) to generate a digital nautical chart, where each grid cell contains a navigability status label (1 for navigable, 0 for prohibited), serving as the basis for geographic constraints.

[0060] Regarding the configuration data of the ship's multi-energy systems, the following components are included: diesel generator (rated power 4000kW, minimum load rate 30%), fuel cell (efficiency 50%, maximum output 500kW), battery pack (capacity 2000kWh, charge / discharge efficiency 95%), thermal management system (cooling power consumption varies with load), and seawater desalination unit (energy consumption ratio 3kWh / m³). 3 Taking this as an example, the performance parameters and operating constraints of these devices (battery SOC needs to be maintained at 20-80%) are analyzed to generate a standardized JSON-formatted energy configuration file, including device number, type, parameter upper and lower limits, and coupling relationships, which serves as input for multi-energy cost calculation.

[0061] In marine environmental forecast data processing, historical forecast data for wind field (wind speed range 0-20 m / s), wave field (wave height 0.5-4 m), current field (current velocity 0.1-2.5 m / s), and sea ice density (0-90%) are collected. Stochastic process modeling (Weibull distribution fitted to wind speed, Gamma distribution fitted to wave height) is employed to generate 100 possible environmental evolution sequences (each sequence containing a 24-hour forecast with a time resolution of 1 hour), forming a commercial navigation environment scenario set. This scenario set is stored in NetCDF format and serves as the input source for Monte Carlo simulations.

[0062] By integrating the aforementioned digital nautical charts, ship energy configuration files, and environmental scenario sets, a unified business decision-making foundation dataset is constructed. This dataset is organized using a spatiotemporal grid index, providing structured input for subsequent optimization.

[0063] 202. Based on the fractional-order system identification algorithm, process the basic dataset of business decision-making, establish the energy consumption history dependence model of the ship power system, and generate a commercial fractional-order parameter set characterizing the ship power characteristics to quantify the energy consumption characteristics under different navigation conditions. Specifically, the business decision-making dataset is preprocessed by cleaning and standardizing the ship speed sequence, main engine power output sequence, and corresponding environmental data sequences to generate a standardized time series dataset suitable for fractional-order system identification. This dataset serves as the input source for parameter identification. The standardized time series dataset is then processed using a fractional-order system identification algorithm. By minimizing the model prediction error criterion, the optimal fractional-order α, characterizing the memory effect of the ship's power system, is identified. This α serves as the core parameter of the ship's power characteristic quantification model. Based on the optimal fractional-order α, the energy consumption response characteristics of the ship's power system are fitted, and the dimensionless efficiency parameter κ is calculated. This parameter κ is used to quantify the energy consumption efficiency of a specific ship under unit power output. Finally, the optimal fractional-order α and the efficiency parameter κ are integrated to generate a ship-specific commercial fractional-order power parameter set. This parameter set serves as the core input for propulsion load calculation in the subsequent business cost prediction model.

[0064] It should be noted that the following is a specific implementation of step 202, using a 7,500-ton container ship sailing in the East China Sea as an example: Speed ​​sequences (18-22 knots), main engine power output sequences (4000-5000kW), and corresponding environmental data sequences (wind speed 6-10m / s, wave height 1-2m) were extracted from the ship's voyage log over the past year. Outliers (fault points where power suddenly drops to zero) were cleaned up and standardized to a uniform dimension, generating a standardized sequence dataset containing 1000 time points.

[0065] By minimizing the model prediction error criterion (root mean square error), the optimal fractional order α characterizing the memory effect of a ship's propulsion system is identified from standardized data. In this example, α = 1.35 is identified, indicating that the system has significant super-diffusion characteristics and can effectively quantify the dependence of historical energy consumption on the current state.

[0066] Based on α=1.35, the energy consumption response characteristics of the ship's propulsion system are fitted. The dimensionless efficiency parameter κ is calculated to quantify the energy consumption efficiency per unit power output. In this example, κ=0.87 (ideal value is 1), indicating that the ship's energy consumption efficiency is 87% under typical operating conditions in the East China Sea, mainly affected by wind and wave resistance and equipment efficiency.

[0067] Integrate α and κ to generate a commercial fractional-order dynamic parameter set. The parameter set is {α=1.35, κ=0.87}, which will serve as the core input for subsequent propulsion load prediction, used to accurately calculate energy consumption characteristics under different navigation conditions (when wind and waves increase).

[0068] 203. Using stochastic differential equations to model environmental uncertainties, and combining the aforementioned commercial fractional-order parameter set, a propulsion load prediction model considering the random changes of wind, waves, currents, and floating ice is established. Furthermore, the operating cost function of multiple energy systems is integrated to form a commercial cost prediction model capable of outputting the probability distribution of operating costs. Specifically, based on a set of commercial navigation environment scenarios, the Ornstein-Uhlenbeck process is used to establish stochastic differential equations for each environmental parameter, generating a continuous-time stochastic evolution model of the environmental parameters. This model serves as the mathematical foundation for quantifying environmental uncertainty. Combining this with a commercial fractional-order parameter set, the stochastic evolution model of the environmental parameters is input into the fractional-order ship dynamics equations to establish a stochastic prediction model for propulsion power demand. This model can output the probability distribution of propulsion power under different environmental conditions. The multi-energy system operating cost function is integrated to convert the probability distribution of propulsion power demand into a scheduling scheme for multiple energy devices, and the fuel consumption cost, equipment operating cost, and external energy procurement cost under each scheme are calculated. Finally, by integrating all cost elements, a complete commercial cost prediction model is constructed. This model can output the complete probability distribution of total operating cost under any navigation strategy, providing a cost assessment basis for subsequent risk perception optimization.

[0069] It should be noted that the following is a specific implementation of step 203, using a 7,500-ton container ship sailing in the East China Sea as an example: Based on a set of commercial navigation environment scenarios, stochastic differential equations for environmental parameters were established using the Ornstein-Uhlenbeck (OU) process. Taking wind field as an example, based on historical East China Sea wind data (average wind speed 10 m / s, wave intensity 2.5 m / s), 100 continuous-time stochastic wind speed sequences (time step 1 hour) were generated using the OU process to characterize the mean reversion characteristics of wind speed (reversion coefficient θ = 0.15, diffusion coefficient σ = 1.8). Similarly, stochastic evolution models were established for wave field (significant wave height 2.5 m ± 0.8 m), current field (current velocity 0.8 m / s ± 0.3 m), and ice density (usually 0 in the East China Sea), forming a joint probability distribution of environmental parameters.

[0070] Combining a commercially available fractional-order parameter set (α=1.35, κ=0.87), the environmental stochastic model is input into the fractional-order ship dynamics equations. Taking the wave field as an example, the random wave height sequence is input into the fractional-order propulsion power equations to calculate the probability distribution of propulsion power demand: within the wave height range of 1.5-3.5m, the power demand of the ship's main diesel engine is distributed as 4200-5800kW (90% confidence interval), with the probability peak occurring near 5000kW.

[0071] An integrated multi-energy system operating cost function is used. Based on the power demand distribution, a multi-energy dispatch scheme is generated: the diesel generator (rated 4000kW) load rate is controlled at 70-95%, with a fuel consumption cost of approximately $0.25 / kWh; the fuel cell (rated 500kW) provides auxiliary power, with an operating cost of approximately $0.35 / kWh; the battery pack discharges during peak power demand (efficiency 95%), with a charging and discharging loss cost of approximately $0.05 / kWh.

[0072] Taking into account fuel costs, equipment wear and tear, and external procurement costs (electricity purchased from shore power if necessary, at a cost of $0.40 / kWh), the complete probability distribution of total operating costs under this navigation strategy is output: the expected cost is $1850 per hour, with a 95% confidence interval of $1600-$2100, providing a quantitative basis for subsequent risk optimization.

[0073] 204. Based on the particle swarm optimization algorithm, optimize the business strategy for risk perception, use the business cost prediction model to perform multi-scenario Monte Carlo simulation on the candidate routes, calculate the expected operating cost and conditional risk value of each scheme, and iteratively generate the optimal business strategy scheme by minimizing the risk-adjusted total cost objective function. Specifically, the process involves initializing a commercial strategy optimization population. An initial set of route options is generated based on digitized nautical charts from the commercial decision-making dataset, and speed control parameters are randomly assigned to each option, forming an initial commercial strategy population encompassing spatial path and time dimensions. Multi-scenario Monte Carlo simulations are then performed, inputting each option from the initial commercial strategy population into a commercial cost prediction model. Large-scale stochastic simulations are conducted using a set of commercial navigation environment scenarios, outputting the operating cost distribution data for each option under different environmental scenarios. A risk-adjusted fitness function is constructed, calculating the expected operating cost and conditional value of risk (CVaR) based on the operating cost distribution data of each option, forming a fitness evaluation function that aims to minimize the risk-adjusted total cost. Particle swarm optimization iterations are implemented, updating the spatial path and speed control parameters of each option in the commercial strategy population based on the fitness evaluation results. Through multiple generations of iterations, the quality of the options is gradually optimized, generating an optimized commercial strategy population. Finally, the optimal commercial strategy option is selected from the final generation of the commercial strategy population, choosing the option with the minimum risk-adjusted total cost, generating the optimal commercial strategy option containing a complete route path and speed plan.

[0074] It should be noted that the following is a specific implementation of step 204, using a 7,500-ton container ship sailing in the East China Sea as an example: Initialize the commercial strategy optimization population: Based on digital nautical charts (spatial grid resolution 0.1°×0.1°), generate 30 initial candidate routes, each containing 20 waypoints (from Shanghai to Ningbo, with latitude and longitude sequences covering navigable waters). Randomly assign speed control parameters (range 16-20 knots) to each route to form the initial population, with each particle representing a route-speed combination scheme.

[0075] Perform multi-scenario Monte Carlo simulations: Input 30 initial scenarios into the commercial cost prediction model and simulate them using a set of 100 environmental sequences in the commercial navigation environment (including random variations in wind speed 6-14 m / s and wave height 1-3 m). Each scenario is run 100 times, and the probability distribution of operating costs is output. The cost distribution of a certain scenario in 100 simulations is: mean $1900 / hour, standard deviation $300, and 95th percentile $2500.

[0076] Construct a risk-adjusted fitness function: Based on the cost distribution of each option, calculate the expected operating cost (E[C]) and conditional value at risk (CVaR, 95% confidence level). The objective is to minimize the fitness function. ; The weights of 0.7 and 0.3 reflect the decision-maker's risk preference (the trade-off between benefits and risks).

[0077] Implement particle swarm optimization iterations: population size 30, 50 generations. Update particle positions based on fitness (course coordinate correction ±0.05°, speed adjustment ±0.5 knots). Retain historical best solutions in each generation (the optimal solution's F-value drops to $1800 in the 20th generation). Ensure the course does not enter restricted areas (water depth <10m or military zones) through constraint processing.

[0078] Selecting the optimal business strategy: The option with the lowest risk adjustment cost (F=$1750 / hour) was selected from the final generation. Its total route length is 220 nautical miles, the average speed is 18.5 knots, the total cost is expected to be 12% lower than the initial option, and the CVaR is controlled within $2000 / hour.

[0079] This process significantly improves the economic efficiency and robustness of routes through PSO iteration, providing decision support for navigation in high-risk waters.

[0080] 205. Analyze and output the final business decision plan, including detailed route trajectory, speed plan and supporting multi-energy equipment scheduling strategy, to form an executable flight plan and energy management decision recommendation; Specifically, the process involves analyzing the spatial position sequence of the optimal business strategy, generating a standardized route trajectory file based on a digital nautical chart spatial reference system. This file contains latitude and longitude coordinate sequences and turning point information, serving as the foundational data for navigation. The time and speed parameters of the optimal business strategy are extracted and combined with the route trajectory to generate a time-synchronized speed plan. This plan details the recommended speed and estimated transit time for each segment, providing guidance for ship maneuvering. Based on the energy scheduling parameters in the optimal business strategy, a multi-energy equipment operation plan is generated, including diesel generator load allocation, fuel cell start-up and shutdown sequence, battery charging and discharging strategies, and freshwater production plans, forming a complete energy management scheme. The route trajectory file, speed plan, and energy management scheme are integrated, and business decision metadata, including estimated total cost, risk value assessment, and carbon emission estimates, is added to generate a final, executable business decision proposal. This business decision proposal is then converted into a standardized machine-readable and human-readable report format, achieving seamless integration with the ship navigation management system and business decision support system.

[0081] It should be noted that the spatial location sequence of the optimal business strategy is analyzed as follows: The latitude and longitude sequence of waypoints is extracted from the optimal solution generated by particle swarm optimization (starting point [122.5°E, 30.8°N] → turning point 1 [123.1°E, 31.2°N] → turning point 2 [123.8°E, 31.5°N] → ending point [124.5°E, 32.0°N]), totaling 12 waypoints. Based on the WGS-84 coordinate system of the digital nautical chart, a standardized route trajectory file (GPX format) is generated, containing the latitude and longitude, turning radius (0.5 nautical miles), and safe water depth (20 meters) information for each point, serving as the basic data for navigation.

[0082] Extracting time and speed parameters and generating a speed plan table: Based on the time and speed parameters in the optimal route and combined with the total 220 nautical mile trajectory, a time-synchronized speed plan table is generated: Segment 1 (0-50 nautical miles) recommended speed 18.5 knots, estimated transit time 2.7 hours; Segment 2 (50-150 nautical miles) reduced to 17.0 knots due to ocean currents, estimated time 5.9 hours; Segment 3 (150-220 nautical miles) restored to 19.0 knots, estimated time 3.7 hours. This plan table details the speed and time points for each segment, serving as a guide for ship maneuvering.

[0083] Generate a multi-energy equipment operation plan: Based on the energy dispatch parameters in the plan, formulate a multi-energy equipment dispatch strategy: Diesel generator: main load 70-85% (output 2800-3400kW), fuel consumption rate 195g / kWh; Fuel cell: activated during low-speed periods in segment 2 (output 400kW), efficiency 52%; Battery pack: discharged during acceleration in segment 3 (SOC drops from 80% to 45%), providing peak power 500kW; Seawater desalination unit: operated during low-load periods at night (energy consumption ratio 3.2kWh / m³), producing 20 tons of fresh water per day.

[0084] Develop a comprehensive energy management plan to ensure a balance between energy supply and demand throughout the entire flight.

[0085] The integrated output delivers actionable business decision recommendations: It integrates route trajectory files, speed plans, and energy management solutions, adding business decision metadata: estimated total cost of $18,500 (expected value), conditional value at risk (CVaR) of $2,100 (95% confidence level), and estimated carbon emissions of 42 tons. The final output is a human-readable PDF report and a machine-readable XML file, enabling seamless integration with the Ship Navigation Management System (ECDIS) and energy management platform.

[0086] 206. Establish a multi-objective collaborative optimization mechanism. Based on the ship energy commercial configuration files and environmental probability forecast data in the commercial decision-making basic dataset, generate a multi-dimensional optimization objective weight set including economic benefits, navigation safety, and environmental protection; implement a dynamic constraint adjustment strategy. Based on real-time environmental change data in the commercial navigation environment scenario set, dynamically adjust the navigation constraints on the digital nautical chart to generate an adaptive navigation constraint rule set; construct a scheme robustness evaluation system. Use the commercial cost prediction model to conduct sensitivity analysis on the optimal commercial strategy scheme and generate a robustness evaluation report including scheme stability indicators and risk tolerance; establish a decision scheme traceability mechanism to record the evolution path and fitness changes of each generation of commercial strategy populations during the optimization process, forming a complete decision process traceability file; implement closed-loop optimization feedback control. Compare and analyze the actual navigation data with the prediction results, and adjust the commercial fractional-order power parameter set and commercial cost prediction model based on the deviation results to form an adaptive model parameter update mechanism.

[0087] It should be noted that a multi-objective collaborative optimization mechanism was established: based on the ship's commercial energy configuration profile (diesel generator efficiency 42%, fuel cell efficiency 50%) and environmental probability forecast data (wind and wave probability distribution), a multi-dimensional optimization objective weight set was generated: economic benefit weight 0.6 (focusing on fuel cost savings), navigation safety weight 0.3 (emphasizing obstacle avoidance and structural stress), and environmental protection weight 0.1 (carbon emission constraints). The weight allocation was verified by the AHP (Analogous Hierarchical Analysis) method, with a consistency ratio of 0.05.

[0088] Based on real-time environmental change data (sudden wind speed increase to 14 m / s) from the commercial navigation environment scenario set, the navigation constraints on the digital nautical chart are dynamically adjusted. The original water depth constraint is increased from 10 meters to 15 meters to cope with the increase in wave height, and the radius of the no-navigation zone is expanded from 500 meters to 800 meters, generating an adaptive navigation constraint rule set (a total of 12 rules).

[0089] Sensitivity analysis was conducted on the optimal business strategy (total cost of $1850 / hour) using a business cost forecasting model. With environmental parameters fluctuating by ±10%, the cost range was $1700-$2000 / hour (stability index 85%), and the risk tolerance threshold was $2200 / hour (based on the historical 95th percentile).

[0090] The evolutionary path of 50 generations of commercial strategy populations during the optimization process was recorded, including changes in population fitness for each generation (from an average of $2,100 in the initial generation to $1,750 in the final generation), the number of spatial path adjustments (a total of 23 major course corrections), and the velocity parameter update trajectory, forming a decision-making process traceability archive in XML format.

[0091] The actual voyage data (the measured fuel consumption was 8% higher than the prediction) was compared with the prediction results. Based on the deviation results, the commercial fractional-order power parameter set was adjusted (α was corrected from 1.35 to 1.38, and κ was adjusted from 0.87 to 0.85) and the environmental response coefficient of the commercial cost prediction model were adjusted to form a monthly updated adaptive model parameter mechanism.

[0092] This mechanism enables the route optimization system to continuously learn. After three months of actual operation, the accuracy of the prediction model improved by 12%, and the stability of the strategy in dynamic environments improved by 18%.

[0093] In this embodiment of the invention, three core data types—geographic information data, ship multi-energy system configuration data, and marine environmental forecast data—are integrated to form a complete commercial decision-making foundation dataset. Geographic information data is processed to generate digital nautical charts, providing a geographical constraint basis for route optimization; ship multi-energy system configuration data is processed to generate standardized energy configuration profiles, providing input for multi-energy cost calculation; marine environmental forecast data is processed to generate a commercial navigation environment scenario set, providing an input source for Monte Carlo simulation; and a historical energy consumption dependency model of the ship's propulsion system is established, generating a commercial fractional-order parameter set. By preprocessing the commercial decision-making foundation dataset, cleaning and standardizing relevant data sequences, the optimal fractional-order α is identified using this algorithm, energy consumption response characteristics are fitted, efficiency parameter κ is calculated, and the parameter set is integrated and generated. Stochastic differential equations are used to model environmental uncertainties, and combined with the commercial fractional-order parameter set, a propulsion load prediction model is established, integrating a multi-energy system operating cost function. Taking a specific ship as an example, based on a stochastic evolution model of environmental parameters and fractional-order ship dynamics equations, the probability distribution of propulsion power demand is output, thereby generating multi-energy scheduling schemes, calculating the cost of each scheme, and constructing a complete commercial cost prediction model that can output the complete probability distribution of total operating costs under any navigation strategy. The system initializes a commercial strategy optimization population, performs multi-scenario Monte Carlo simulations, constructs a risk-adjusted fitness function, implements particle swarm optimization iterations, and selects the optimal commercial strategy scheme, providing effective decision support for navigation in high-risk waters and achieving a balance between economic benefits and risk control in route optimization. The optimal commercial strategy scheme is analyzed, generating standardized route trajectory files, speed plans, and multi-energy equipment operation plans. Commercial decision metadata is integrated and added to generate an executable commercial decision proposal, which is then converted into machine-readable and human-readable formats, achieving seamless integration with the ship navigation management system and commercial decision support system. Simultaneously, a multi-objective collaborative optimization mechanism, a dynamic constraint adjustment strategy, a scheme robustness evaluation system, a decision scheme traceability mechanism, and closed-loop optimization feedback control are established.

[0094] The above describes the route optimization method based on particle swarm optimization algorithm and geographic information in the embodiments of the present invention. The following describes the route optimization system based on particle swarm optimization algorithm and geographic information in the embodiments of the present invention. Please refer to [link / reference]. Figure 3An embodiment of the route optimization system based on particle swarm optimization algorithm and geographic information in this invention includes: a construction module 301, used to integrate geographic information data, ship multi-energy system configuration data, and marine environmental forecast data to form a basic dataset for business decision-making; a quantification module 302, used to process the basic dataset for business decision-making based on a fractional-order system identification algorithm, establish a historical energy consumption dependency model of the ship's power system, and generate a commercial fractional-order parameter set; a prediction module 303, used to model environmental uncertainties using stochastic differential equations, combine the commercial fractional-order parameter set, establish a propulsion load prediction model, and integrate the multi-energy system operating cost function to form a business cost prediction model; a setting module 304, used to optimize business strategies based on risk perception using particle swarm optimization algorithm, use the business cost prediction model to perform multi-scenario Monte Carlo simulations on candidate routes, calculate the expected operating cost and conditional risk value of each scheme, and iteratively generate the optimal business strategy scheme by minimizing the risk-adjusted total cost objective function; and a decision module 305, used to parse and output the final business decision scheme, including detailed route trajectory, speed plan, and supporting multi-energy equipment scheduling strategy, forming a navigation plan and energy management decision recommendation.

[0095] In this embodiment of the invention, multi-source data such as geographic information, ship multi-energy system configuration, and marine environmental forecasts are integrated to provide a comprehensive and accurate foundation for business decision-making, breaking through the limitations of traditional single data sources and making decisions more scientific and forward-looking. A fractional-order system identification algorithm is used to establish a historical dependence model of ship power system energy consumption, which can more accurately capture the complex relationship between energy consumption and historical factors. A propulsion load prediction model is established by combining stochastic differential equations to model environmental uncertainties, effectively addressing the impact of complex and changeable marine environments on predictions and improving prediction accuracy. Based on particle swarm optimization algorithm, risk perception business strategy optimization is performed. Risks are comprehensively assessed through multi-scenario Monte Carlo simulations, minimizing the risk-adjusted total cost objective function to generate the optimal business strategy scheme, achieving cost optimization of route planning and energy management under controllable risks. The decision module not only outputs detailed route trajectories and speed plans, but also provides supporting multi-energy equipment scheduling strategies, forming a complete navigation plan and energy management decision-making recommendations, providing comprehensive guidance for ship operation and improving operational efficiency and effectiveness.

[0096] above Figure 3 The route optimization system based on particle swarm optimization algorithm and geographic information in this embodiment of the invention will be described in detail from the perspective of modular functional entities. The route optimization device based on particle swarm optimization algorithm and geographic information in this embodiment of the invention will be described in detail from the perspective of hardware processing.

[0097] Figure 4This is a schematic diagram of a flight route optimization device based on particle swarm optimization algorithm and geographic information according to an embodiment of the present invention. The flight route optimization device 400 based on particle swarm optimization algorithm and geographic information can vary considerably due to different configurations or performance. The flight route optimization device 400 based on particle swarm optimization algorithm and geographic information includes a transmitter 401, a receiver 402, and a processor 403. The processor 403 can also be a controller. Optionally, the flight route optimization device 400 based on particle swarm optimization algorithm and geographic information may further include a modem processor 405, which may include an encoder 406, a modulator 407, a decoder 408, and a demodulator 409.

[0098] In one example, transmitter 401 modulates (e.g., analog-to-analog conversion, filtering, amplification, and up-conversion, etc.) the output sample and generates an uplink signal, which is transmitted via an antenna to an access network device. On the downlink, the antenna receives the downlink signal transmitted by the access network device. Receiver 402 modulates (e.g., filtering, amplification, down-conversion, and digitization, etc.) the signal received from the antenna and provides an input sample. In modem processor 405, encoder 406 receives traffic data and signaling messages to be transmitted on the uplink and processes (e.g., formatting, encoding, and interleaving) the traffic data and signaling messages. Modulator 407 further processes (e.g., symbol mapping and modulation) the encoded traffic data and signaling messages and provides an output sample. Demodulator 409 processes (e.g., demodulates) the input sample and provides a symbol estimate. Decoder 408 processes (e.g., deinterleaving and decoding) the symbol estimate and provides the decoded data and signaling messages to a flight path optimization device 400 based on particle swarm optimization and geographic information. Encoder 406, modulator 407, demodulator 409, and decoder 408 can be implemented by a combined modem processor 405. These units process data according to the radio access technology used by the radio access network (e.g., LTE and other evolved systems access technologies). It should be noted that when the flight path optimization device 400 based on particle swarm optimization algorithm and geographic information does not include modem processor 405, the aforementioned functions of modem processor 405 can also be performed by processor 403.

[0099] The processor 403 controls and manages the operation of the flight route optimization device 400 based on particle swarm optimization algorithm and geographic information, and is used to execute the processing procedures performed by the flight route optimization device 400 based on particle swarm optimization algorithm and geographic information in the above embodiments of this disclosure. For example, the processor 403 is also used to execute various steps of the transmitting or receiving device in the above method embodiments, and / or other steps of the technical solutions described in the embodiments of this disclosure.

[0100] Furthermore, the flight path optimization device 400 based on particle swarm optimization algorithm and geographic information may also include a memory 404 for storing program code and data for the flight path optimization device 400 based on particle swarm optimization algorithm and geographic information.

[0101] Understandable, Figure 4 This illustration only shows a simplified design of the flight path optimization device 400 based on particle swarm optimization algorithm and geographic information. In practical applications, the flight path optimization device 400 based on particle swarm optimization algorithm and geographic information can include any number of transmitters, receivers, processors, modem processors, memory, etc., and all devices that can implement the embodiments of this disclosure are within the protection scope of the embodiments of this disclosure.

[0102] The present invention also provides a flight route optimization device based on particle swarm optimization algorithm and geographic information. The flight route optimization device based on particle swarm optimization algorithm and geographic information includes a memory and a processor. The memory stores computer-readable instructions. When the computer-readable instructions are executed by the processor, the processor performs the steps of the flight route optimization method based on particle swarm optimization algorithm and geographic information in the above embodiments.

[0103] The present invention also provides a computer-readable storage medium, which can be a non-volatile computer-readable storage medium or a volatile computer-readable storage medium, wherein the computer-readable storage medium stores instructions that, when the instructions are executed on a computer, cause the computer to perform the steps of the route optimization method based on particle swarm optimization algorithm and geographic information.

[0104] Those skilled in the art will clearly understand that, for the sake of convenience and brevity, the specific working processes of the systems, devices, and units described above can be referred to the corresponding processes in the foregoing method embodiments, and will not be repeated here.

[0105] If the integrated unit is implemented as a software functional unit and sold or used as an independent product, it can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of the present invention, in essence, or the part that contributes to the prior art, or all or part of the technical solution, can be embodied in the form of a software product. This computer software product is stored in a storage medium and includes several instructions to cause a computer device (which may be a personal computer, server, or network device, etc.) to execute all or part of the steps of the methods described in the various embodiments of the present invention. The aforementioned storage medium includes various media capable of storing program code, such as USB flash drives, portable hard drives, read-only memory (ROM), random access memory (RAM), magnetic disks, or optical disks.

[0106] The above-described embodiments are only used to illustrate the technical solutions of the present invention, and are not intended to limit it. Although the present invention has been described in detail with reference to the foregoing embodiments, those skilled in the art should understand that modifications can still be made to the technical solutions described in the foregoing embodiments, or equivalent substitutions can be made to some of the technical features. Such modifications or substitutions do not cause the essence of the corresponding technical solutions to deviate from the spirit and scope of the technical solutions of the embodiments of the present invention.

Claims

1. A route optimization method based on particle swarm optimization algorithm and geographic information, characterized in that, The route optimization method based on particle swarm optimization algorithm and geographic information includes: Integrate geographic information data, ship multi-energy system configuration data, and marine environmental forecast data to form a basic dataset for business decision-making; Based on the fractional-order system identification algorithm, a business decision-making dataset is processed to establish a historical energy consumption dependency model for ship power systems and generate a commercial fractional-order parameter set. Stochastic differential equations are used to model environmental uncertainties. Combined with the commercial fractional parameter set, a propulsion load prediction model is established, and multi-energy system operating cost functions are integrated to form a commercial cost prediction model. Business strategy optimization based on particle swarm optimization algorithm for risk perception, using the aforementioned business cost prediction model to conduct multi-scenario Monte Carlo simulations of candidate routes, calculating the expected operating cost and conditional value of risk for each option, and iteratively generating the optimal business strategy by minimizing the risk-adjusted total cost objective function, including: An initial set of route options is generated based on the digital nautical charts in the business decision-making dataset, and speed control parameters are randomly assigned to each option to form an initial business strategy population. Input each scheme in the initial business strategy population into the business cost prediction model, use the commercial navigation environment scenario set to conduct large-scale stochastic simulation, and output the operating cost distribution data of each scheme under different environmental scenarios. Based on the operating cost distribution data of each option, the expected operating cost and conditional value of risk (CVaR) are calculated, and a fitness assessment function is formed with the goal of minimizing the risk-adjusted total cost, thus obtaining the fitness assessment results. The spatial path and speed control parameters of each scheme in the commercial strategy population are updated based on the fitness assessment results. The quality of the schemes is gradually optimized through multiple generations of iteration to generate a commercial strategy population. The optimal business strategy is generated by selecting the solution with the lowest risk-adjusted total cost from the population of business strategies. The system analyzes and outputs the final business decision-making solution, including detailed route trajectory, speed plan, and supporting multi-energy equipment scheduling strategy, forming a navigation plan and energy management decision-making recommendation.

2. The route optimization method based on particle swarm optimization algorithm and geographic information according to claim 1, characterized in that, include: Process geographic information data, extract coastline, water depth data and information on prohibited navigation areas, and generate digital nautical charts; Process ship multi-energy system configuration data, analyze the performance parameters and operating constraints of diesel generators, fuel cells, battery packs, thermal management systems and seawater desalination devices, and generate ship energy commercial configuration files; To process marine environmental forecast data, a stochastic process modeling method is used to fit the probability distribution of wind field, wave field, current field and ice density data to generate a set of commercial navigation environment scenarios. By integrating the aforementioned digital nautical charts, ship energy commercial configuration files, and commercial navigation environment scenario sets, a complete dataset for commercial decision-making is constructed.

3. The route optimization method based on particle swarm optimization algorithm and geographic information according to claim 1, characterized in that, Based on the aforementioned commercial fractional-order parameter set, a propulsion load prediction model is established, where the load power is... : ； Where V is the speed and D is the draft. Let C(t) be the wave height, C(t) be the current velocity, α be the linear speed influence coefficient, β be the nonlinear speed influence coefficient, η and κ be the environmental coupling coefficients, K be the cyclotron exponent, and T be the follower exponent.

4. The route optimization method based on particle swarm optimization algorithm and geographic information according to claim 2, characterized in that, include: Preprocess the basic dataset for business decision-making, clean and standardize the ship speed sequence, main engine power output sequence and corresponding environmental data sequence to generate a standardized time series dataset; The standardized time series dataset is processed using a fractional-order system identification algorithm. By minimizing the model prediction error criterion, the optimal fractional-order α that characterizes the memory effect of the ship's power system is identified. Based on the optimal fractional order α, the energy consumption response characteristics of the ship's power system are fitted, and the efficiency parameter κ is calculated. By integrating the optimal fractional order α and the efficiency parameter κ, a commercial fractional dynamic parameter set is generated.

5. The route optimization method based on particle swarm optimization algorithm and geographic information according to claim 4, characterized in that, include: Establish stochastic differential equations for each environmental parameter and generate a continuous-time stochastic evolution model for the environmental parameters; By combining commercial fractional-order parameter sets, a stochastic evolution model of environmental parameters is input into the fractional-order ship dynamics equations to establish a stochastic prediction model for propulsion power demand. The integrated multi-energy system operating cost function transforms the probability distribution of propulsion power demand into a multi-energy equipment scheduling scheme, and calculates the fuel consumption cost, equipment operating cost, and external energy procurement cost under each scheme. A business cost forecasting model is constructed based on all cost elements.

6. The route optimization method based on particle swarm optimization algorithm and geographic information according to claim 5, characterized in that, include: Analyze the spatial location sequence of the optimal business strategy and generate route trajectory files based on the spatial reference system of digital nautical charts; Extract the time and speed parameters of the optimal business strategy and generate a time-synchronized flight speed plan table by combining it with the flight path. Based on the energy dispatch parameters in the optimal business strategy, a multi-energy equipment operation plan is generated, including diesel generator load allocation, fuel cell start-up and shutdown sequence, battery charging and discharging strategy, and freshwater production plan, forming an energy management scheme. Integrate route trajectory files, speed plan tables, and energy management solutions, add business decision metadata, and generate business decision recommendations.

7. The route optimization method based on particle swarm optimization algorithm and geographic information according to claim 6, characterized in that, Also includes: Based on the ship energy business configuration files and environmental probability forecast data in the business decision-making basic dataset, a multi-dimensional set of optimization objective weights including economic benefits, navigation safety and environmental protection is generated. Based on real-time environmental change data from commercial navigation environment scenarios, the navigation constraints on digital nautical charts are dynamically adjusted to generate a set of navigation constraint rules. Sensitivity analysis of the optimal business strategy is conducted using a business cost forecasting model, and a robustness assessment report is generated. Record the evolutionary paths and fitness changes of each generation of business strategy populations during the optimization process to form a decision-making process traceability archive; By comparing and analyzing actual navigation data with prediction results, and adjusting the commercial fractional-order dynamic parameter set and commercial cost prediction model based on the deviation results, a model parameter update mechanism is formed.

8. A route optimization system based on particle swarm optimization algorithm and geographic information, characterized in that, The flight route optimization system based on particle swarm optimization algorithm and geographic information includes: The module is used to integrate geographic information data, ship multi-energy system configuration data, and marine environmental forecast data to form a basic dataset for business decision-making. The quantization module is used to process the basic dataset for business decision-making based on the fractional-order system identification algorithm, establish the energy consumption history dependency model of the ship power system, and generate commercial fractional-order parameter sets. The prediction module is used to model environmental uncertainties using stochastic differential equations, combine the commercial fractional-order parameter set, establish a propulsion load prediction model, and integrate the multi-energy system operating cost function to form a commercial cost prediction model. The configuration module is used for risk-aware business strategy optimization based on the particle swarm optimization algorithm. It utilizes the business cost prediction model to perform multi-scenario Monte Carlo simulations on candidate routes, calculates the expected operating cost and conditional value of risk for each option, and iteratively generates the optimal business strategy by minimizing the risk-adjusted total cost objective function. This includes: An initial set of route options is generated based on the digital nautical charts in the business decision-making dataset, and speed control parameters are randomly assigned to each option to form an initial business strategy population. Input each scheme in the initial business strategy population into the business cost prediction model, use the commercial navigation environment scenario set to conduct large-scale stochastic simulation, and output the operating cost distribution data of each scheme under different environmental scenarios. Based on the operating cost distribution data of each option, the expected operating cost and conditional value of risk (CVaR) are calculated, and a fitness assessment function is formed with the goal of minimizing the risk-adjusted total cost, thus obtaining the fitness assessment results. The spatial path and speed control parameters of each scheme in the commercial strategy population are updated based on the fitness assessment results. The quality of the schemes is gradually optimized through multiple generations of iteration to generate a commercial strategy population. The optimal business strategy is generated by selecting the solution with the lowest risk-adjusted total cost from the population of business strategies. The decision-making module is used to analyze and output the final business decision-making solution, including detailed route trajectory, speed plan and supporting multi-energy equipment scheduling strategy, forming a navigation plan and energy management decision-making recommendation.

Images