A method and system for optimizing ship speed based on standard routes

Abstract

This invention provides a method and system for optimizing ship speed based on standard routes. First, historical navigation parameters and operating cost parameters of the ship are obtained. Then, a machine learning algorithm is used to model the relationship between ship speed and fuel consumption. Next, specific calculation methods are used to calculate the fuel cost, expected total revenue, and total voyage time of the standard route, thereby establishing an equivalent time charter rate calculation model. Under the condition of not considering light fuel consumption and a fixed speed at full load, the equivalent time charter rate calculation model is simplified to obtain a speed optimization model. Finally, based on the linear relationship equation between speed and fuel consumption established using linear regression, the speed optimization model is linearized to obtain a processed speed optimization model. The optimal speed of the processed speed optimization model is calculated using a speed optimization algorithm to maximize the equivalent time charter rate, thus optimizing ship speed, improving shipping efficiency, and reducing operating costs.

Description

Technical Field

[0001] This invention relates to the field of digital transformation and intelligent development technology in shipping, specifically to a method and system for optimizing ship speed based on standard routes. Background Technology

[0002] Speed ​​optimization is a key step in improving shipping economics and is also significant for reducing ship carbon emissions and promoting environmental protection. Traditional speed selection often relies on the personal experience of the captain or dispatcher. This method is subjective and unstable, difficult to adapt to the complex and ever-changing shipping environment, and cannot provide a scientific and reasonable basis for decision-making.

[0003] By quantifying various influencing factors and reference indicators related to speed during the voyage, such as voyage revenue, fuel consumption cost, voyage time, and carbon emissions, the speed optimization algorithm aims at voyage benefits and voyage time, while taking business logic as constraints, to find the optimal speed to achieve the optimal goal. It also provides the most effective target value and the values ​​of various auxiliary variables to assist business personnel in making scientific and reasonable decisions.

[0004] In recent years, the development of operations research algorithms has provided solutions for speed optimization. These algorithms include, but are not limited to, genetic algorithms, particle swarm optimization, simulated annealing, and ant colony optimization. They can handle complex optimization problems and have been successfully applied in multiple fields. However, applying operations research algorithms to the field of speed optimization presents the following difficulties and challenges: 1) High computational complexity: Shipping operations involve numerous factors, such as voyage rates, fuel consumption, tallying fees, and port charges, and are related to specific routes, vessels, cargo, and vessel status. The abstraction and characterization of the optimization problem requires deep integration with business logic and comprehensive consideration of relevant factors; 2) Data processing and transformation: The relationship between shipping-related factors and vessel speed needs to be established. For example, historical data stores vessel fuel consumption data as fuel consumption values ​​corresponding to various speeds, rather than explicit relational expressions. Therefore, before establishing a speed optimization model, it is necessary to first quantify the relationship between various relevant factors and speed to create a good data foundation for algorithm optimization.

[0005] Therefore, there is an urgent need for a solution that can deeply study speed optimization technology to improve the scientific level of speed decision-making, thereby improving shipping efficiency and environmental protection. Summary of the Invention

[0006] To address the problems of high computational complexity and lack of consideration for the relationship between shipping-related factors and ship speed in existing target detection algorithms, this invention provides a ship speed optimization method based on standard routes. This method can more accurately describe the ship speed optimization problem, achieve precise assessment of transportation costs and revenues at different speeds, significantly improve shipping efficiency, and reduce operating costs. This invention also relates to a ship speed optimization system based on standard routes.

[0007] The technical solution of the present invention is as follows:

[0008] A method for optimizing ship speed based on a standard route, characterized by the following steps:

[0009] Parameter acquisition steps: Based on the standard route basic data table provided by the business side, obtain the ship's historical navigation parameters and operating cost parameters. The ship's historical navigation parameters include empty speed, full-load speed, heavy fuel oil consumption and light fuel oil consumption at empty speed, heavy fuel oil consumption and light fuel oil consumption at full-load speed, route empty heavy fuel oil and light fuel oil distance, route full-load heavy fuel oil and light fuel oil distance, route total cargo capacity, route waiting time for loading, route loading time, and route unloading time. The operating cost parameters include heavy fuel oil price, light fuel oil price, route waiting fuel cost, route loading fuel cost, route unloading fuel cost, route port usage fees, basic rates, and commission rates.

[0010] Steps for constructing the speed and fuel consumption model: Based on the empty speed and the heavy fuel consumption and light fuel consumption at the empty speed, machine learning algorithms are used to construct the empty speed and heavy fuel consumption model and the empty speed and light fuel consumption model, respectively; Based on the full-load speed and the heavy fuel consumption and light fuel consumption at the full-load speed, machine learning algorithms are used to construct the full-load speed and heavy fuel consumption model and the full-load speed and light fuel consumption model, respectively.

[0011] The standard route fuel cost calculation steps are as follows: First, calculate the estimated empty-load fuel cost based on the empty-load speed and heavy fuel consumption model, the empty-load speed and light fuel consumption model, the empty heavy fuel distance of the route, the empty light fuel distance of the route, the heavy fuel price, the light fuel price, and the empty-load speed. Second, calculate the estimated full-load fuel cost based on the full-load speed and heavy fuel consumption model, the full-load speed and light fuel consumption model, the full-load heavy fuel distance of the route, the full-load light fuel distance of the route, the heavy fuel price, the light fuel price, and the full-load speed. Finally, combine the estimated empty-load fuel cost, the estimated full-load fuel cost, the fuel cost when the route is loaded, the fuel cost when the route is loaded, and the fuel cost when the route is unloaded to obtain the standard route fuel cost.

[0012] The steps for establishing an equivalent time charter fee calculation model are as follows: Calculate the estimated total revenue of the route based on the total cargo volume, base rates, and commission rates; calculate the total voyage time based on the route's full-load heavy fuel oil distance, full-load light fuel oil distance, empty-load heavy fuel oil distance, empty-load light fuel oil distance, full-load speed, empty-load speed, loading and unloading time, and unloading time; and establish an equivalent time charter fee calculation model based on standard route fuel costs, estimated total revenue, total voyage time, and port usage fees.

[0013] Steps for constructing the speed optimization model: Under the condition of not considering light fuel consumption and fixed full-load speed, the equivalent time charter fee calculation model is simplified to obtain a speed optimization model with empty speed as the core independent variable, heavy fuel consumption corresponding to empty speed as the auxiliary independent variable, and equivalent time charter fee as the dependent variable.

[0014] The linear processing and optimal speed calculation steps are as follows: Based on the linear relationship equation between speed and fuel consumption established by the linear regression method, the speed optimization model is linearized to obtain the processed speed optimization model; and the optimal speed of the processed speed optimization model is calculated by the speed optimization algorithm to maximize the equivalent time charter rate, thus completing the optimization of the ship's speed.

[0015] Preferably, in the step of constructing the speed and fuel consumption model, the machine learning algorithm adopts a linear regression model, and the regression coefficients in the linear regression model are estimated by the least squares method to ensure the accuracy of the linear regression model.

[0016] Preferably, in the parameter acquisition step, the ship navigation parameters further include light fuel consumption per unit time under load, heavy fuel consumption per unit time under load, heavy fuel consumption per unit time during loading, light fuel consumption per unit time during loading, heavy fuel consumption per unit time during unloading, and light fuel consumption per unit time during unloading.

[0017] Preferably, in the parameter acquisition step, the fuel cost for the route is calculated based on the route loading time, light fuel consumption per unit time, heavy fuel consumption per unit time, heavy fuel price, and light fuel price; the loading fuel cost is calculated based on the route loading time, heavy fuel consumption per unit time, light fuel consumption per unit time, heavy fuel price, and light fuel price; and the unloading fuel cost is calculated based on the route unloading time, heavy fuel consumption per unit time, light fuel consumption per unit time, heavy fuel price, and light fuel price.

[0018] In the standard route fuel cost calculation model establishment step, the calculated expected empty-load fuel cost, expected full-load fuel cost, fuel cost when the route is loaded, fuel cost when the route is loaded, and fuel cost when the route is unloaded are combined and used as the standard route fuel cost.

[0019] Preferably, in the step of constructing the speed optimization model, simplifying the equivalent time charter rate calculation model means: for large ships, retaining heavy fuel oil consumption and ignoring light fuel oil consumption, and setting the full-load speed as a constant to simplify the equivalent time charter rate calculation model, thereby obtaining a speed optimization model with unloaded speed as the core independent variable, heavy fuel oil consumption corresponding to unloaded speed as the auxiliary independent variable, and equivalent time charter rate as the dependent variable.

[0020] Preferably, in the linear processing and optimal speed calculation steps, the linearization of the speed optimization model is performed by replacing the auxiliary independent variable heavy fuel consumption with a linear mapping based on the linear relationship between the empty speed and heavy fuel consumption in the linear equation, thus obtaining the linearized speed optimization model.

[0021] The speed optimization algorithm includes:

[0022] The derivative curve of the dependent variable of the optimized equivalent time charter rate is obtained by taking the derivative of the processed speed optimization model.

[0023] The derivative curve is analyzed to determine local maxima, including identifying extreme points and evaluating the stability and robustness of the processed speed optimization model;

[0024] Output the speed results corresponding to the top three local maximum values ​​of the equivalent time charter rate dependent variable.

[0025] A ship speed optimization system based on a standard route is characterized by comprising, in sequence, a parameter acquisition module, a speed and fuel consumption model construction module, a standard route fuel cost calculation module, an equivalent time charter rate calculation model establishment module, a speed optimization model construction module, and a linear processing and optimal speed calculation module.

[0026] The parameter acquisition module obtains ship navigation parameters and operating cost parameters based on the standard route basic data table provided by the business terminal. The ship navigation parameters include empty speed, full-load speed, heavy fuel oil consumption and light fuel oil consumption at empty speed, heavy fuel oil consumption and light fuel oil consumption at full-load speed, route empty heavy fuel oil and light fuel oil distance, route full-load heavy fuel oil and light fuel oil distance, route total cargo capacity, route waiting time for loading, route loading time, and route unloading time. The operating cost parameters include heavy fuel oil price, light fuel oil price, route waiting fuel cost, route loading fuel cost, route unloading fuel cost, route port usage fees, basic rates, and commission rates.

[0027] The speed and fuel consumption model building module constructs models based on empty speed and heavy fuel consumption and light fuel consumption at empty speed using machine learning algorithms; and constructs models based on full-load speed and heavy fuel consumption and light fuel consumption using machine learning algorithms.

[0028] The standard route fuel cost calculation module calculates the estimated empty-load fuel cost of a route based on empty-load speed and heavy fuel consumption models, empty-load speed and light fuel consumption models, the route's empty heavy fuel distance, the route's empty light fuel distance, heavy fuel price, light fuel price, and empty-load speed. It also calculates the estimated full-load fuel cost of a route based on full-load speed and heavy fuel consumption models, full-load speed and light fuel consumption models, the route's full-load heavy fuel distance, the route's full-load light fuel distance, heavy fuel price, light fuel price, and full-load speed. Finally, it combines the estimated empty-load fuel cost, the estimated full-load fuel cost, the route's fuel cost when loaded, the route's fuel cost when loaded, and the route's fuel cost when unloaded to obtain the standard route fuel cost.

[0029] The equivalent time charter fee calculation model module calculates the expected total revenue of the route based on the total cargo volume, basic rates, and commission rates; it calculates the total voyage time based on the route's full-load heavy fuel oil distance, full-load light fuel oil distance, empty-load heavy fuel oil distance, empty-load light fuel oil distance, full-load speed, empty-load speed, loading and unloading time, and unloading time; and it establishes an equivalent time charter fee calculation model based on standard route fuel costs, expected total revenue, total voyage time, and port usage fees.

[0030] The speed optimization model construction module simplifies the equivalent time charter rate calculation model under the condition of not considering light fuel consumption and fixed full-load speed, and obtains a speed optimization model with empty speed as the core independent variable, heavy fuel consumption corresponding to empty speed as the auxiliary independent variable, and equivalent time charter rate as the dependent variable.

[0031] The linear processing and optimal speed calculation module linearizes the speed optimization model based on the linear relationship equation between speed and fuel consumption established by the linear regression method, resulting in a processed speed optimization model. The module then uses a speed optimization algorithm to calculate the optimal speed of the processed speed optimization model to maximize the equivalent time charter rate, thus completing the optimization of the ship's speed.

[0032] Preferably, in the speed and fuel consumption model construction module, the machine learning algorithm adopts a linear regression model, and the regression coefficients in the linear regression model are estimated by the least squares method to ensure the accuracy of the linear regression model.

[0033] Preferably, in the parameter acquisition module, the ship navigation parameters further include light fuel consumption per unit time under load, heavy fuel consumption per unit time under load, heavy fuel consumption per unit time during loading, light fuel consumption per unit time during loading, heavy fuel consumption per unit time during unloading, and light fuel consumption per unit time during unloading.

[0034] Preferably, in the parameter acquisition module, the fuel cost for the route is calculated based on the route loading time, light fuel consumption per unit time, heavy fuel consumption per unit time, heavy fuel price, and light fuel price; the loading fuel cost is calculated based on the route loading time, heavy fuel consumption per unit time, light fuel consumption per unit time, heavy fuel price, and light fuel price; and the unloading fuel cost is calculated based on the route unloading time, heavy fuel consumption per unit time, light fuel consumption per unit time, heavy fuel price, and light fuel price.

[0035] In the standard route fuel cost calculation model establishment module, the calculated expected empty-load fuel cost, expected full-load fuel cost, fuel cost when the route is loaded, fuel cost when the route is loaded, and fuel cost when the route is unloaded are combined to form the standard route fuel cost.

[0036] The beneficial effects of this invention are as follows:

[0037] This invention provides a method for optimizing ship speed based on standard routes. First, based on a standard route data table provided by the business side, specific historical ship navigation parameters and operating cost parameters are obtained, covering key indicators such as speed, fuel consumption, distance, time, and cost. This provides a comprehensive and reliable data foundation for subsequent model construction, reducing errors caused by missing data and integrating historical data with real-time parameters (such as oil prices and the WS index) to support dynamic optimization. Based on historical ship navigation parameters and using machine learning algorithms, the relationship between ship speed and fuel consumption is modeled. Separate models are constructed for empty-load speed versus heavy fuel consumption, empty-load speed versus light fuel consumption, fully loaded speed versus heavy fuel consumption, and fully loaded speed versus light fuel consumption. This achieves a formulaic expression of discrete data, ensuring that the model fitting results are both accurate and reliable. Independent models for empty / fully loaded speed and heavy / light fuel consumption are established, supporting differentiated scenario analysis and providing quantitative basis for fuel cost calculation, avoiding reliance on experience-based estimations. Then, specific calculation methods are used to calculate the estimated empty and full-load fuel costs of the route. The estimated empty fuel cost, estimated full-load fuel cost, fuel cost under load, fuel cost at the loading point, and fuel cost at the unloading point are combined as the standard route fuel cost. Fuel costs are then calculated for different scenarios (empty / full-load) to refine the cost structure. Combined with parameters such as flight distance, fuel price, and speed, fuel consumption is precisely quantified, providing core cost variables for the TCE model and supporting economic benefit assessment. Next, the estimated total revenue and total voyage time of the route are calculated. Based on the standard route fuel cost, estimated total revenue, total voyage time, and port usage fees, an equivalent time charter rate calculation model (TCE model) is established. This model integrates multi-dimensional indicators such as revenue, cost, and time to construct a comprehensive economic benefit assessment system. The TCE model standardizes complex shipping scenarios, facilitating cross-route comparisons and providing a clear objective function for speed optimization (maximizing TCE). Under the condition of not considering light fuel consumption and fixed full-load speed, the equivalent time charter rate calculation model is simplified to obtain a speed optimization model with empty speed as the core independent variable, heavy fuel consumption at empty speed as the auxiliary independent variable, and equivalent time charter rate as the dependent variable. By integrating key information such as speed, fuel consumption data, and fuel price from historical navigation parameters and operating cost parameters, and with the equivalent time charter rate (TCE) as the objective and empty speed and heavy fuel consumption function at empty speed as independent variables, the model achieves accurate assessment of transportation costs and revenues at different empty speeds. It can accurately capture the relationship between empty speed and TCE, and by identifying various influencing factors and auxiliary variables, it can more accurately describe the ship speed optimization problem and provide a foundation for subsequent modeling and optimization algorithm implementation.Finally, based on the linear relationship equation between speed and fuel consumption established using linear regression, the speed optimization model is linearized to obtain a processed speed optimization model. This simplifies the model complexity, effectively improves computational efficiency, reduces the difficulty of solving the model, and enhances the reliability of the optimization results. Furthermore, the linearized model is a convex optimization problem with a globally optimal solution, making the solution more stable and efficient. Therefore, it can find the optimal speed more quickly, improving ship operating efficiency. The speed optimization algorithm is then used to calculate the optimal speed of the processed speed optimization model to maximize the equivalent time charter rate, thus completing the ship speed optimization. This more accurately describes the ship speed optimization problem, enabling precise assessment of transportation costs and revenues at different speeds, significantly improving shipping efficiency and reducing operating costs. This invention achieves intelligent decision-making for ship speed through a closed-loop process of data integration → model construction → optimization solution. Each step works in synergy, ultimately achieving multiple goals: improved economic benefits, reduced operating costs, and optimized environmental performance, providing an innovative solution for the sustainable development of the shipping industry.

[0038] This invention also relates to a ship speed optimization system based on standard routes. This system corresponds to the aforementioned ship speed optimization method based on standard routes and can be understood as a system that implements the aforementioned ship speed optimization method based on standard routes. It includes a parameter acquisition module, a speed and fuel consumption model construction module, a standard route fuel cost calculation module, an equivalent time charter rate calculation model establishment module, a speed optimization model construction module, and a linear processing and optimal speed calculation module, all connected sequentially. These modules work collaboratively to model the relationship between ship speed and fuel consumption based on historical navigation parameters and using machine learning algorithms. This achieves a formulaic expression of discrete data, ensuring the model... The fitting results of the model are both accurate and reliable. By integrating key information such as speed, fuel consumption data and fuel price from historical navigation parameters and operating cost parameters, and taking the equivalent time charter rate (TCE) as the objective and the empty speed and heavy fuel consumption function at empty speed as independent variables, the model achieves an accurate assessment of transportation costs and revenues at different empty speeds. It can accurately capture the relationship between speed and TCE. Finally, based on the linear relationship equation between speed and fuel consumption established by the linear regression method, the speed optimization model is linearized to obtain the processed speed optimization model, which simplifies the model complexity, effectively improves the computational efficiency, and can find the optimal speed more quickly, thereby improving the efficiency of ship operation. This invention shifts from relying on human experience to data-driven quantitative analysis, reducing subjective errors; it balances revenue and costs through the TCE model, maximizing profit per unit time, reducing fuel consumption, operating costs, and carbon emissions, and optimizing economic benefits; model simplification and linear processing improve computational efficiency, support real-time optimization, and multi-scenario models (empty / full load, heavy fuel / light fuel) adapt to complex shipping environments, enhancing system robustness; it integrates shipping data, algorithms, and business logic to promote digital transformation, providing standardized tools for ship scheduling and route planning, and improving overall industry efficiency. Attached Figure Description

[0039] Figure 1 This is a flowchart of the ship speed optimization method based on standard routes according to the present invention.

[0040] Figure 2 This is a schematic diagram showing the calculation results of the optimal speed for each ship in this invention.

[0041] Figure 3 This is a schematic diagram showing the specific optimal speed of a certain vessel according to the present invention. Detailed Implementation

[0042] The present invention will now be described with reference to the accompanying drawings.

[0043] This invention relates to a method for optimizing ship speed based on standard routes. This method uses big data analysis and intelligent optimization algorithms to calculate the equivalent time charter rate (also known as ship operating cost) for optimal speed selection. It aims to improve shipping efficiency and reduce operating costs through accurate simulation and optimization strategies. First, this invention proposes a speed optimization model based on standard routes. This model integrates key information such as basic standard route data, ship speed and fuel consumption data, WS index, and oil price index. Using the equivalent time charter rate (TCE) as the objective and the empty speed and heavy fuel consumption function at empty speed as independent variables, it achieves accurate assessment of transportation costs and revenues at different speeds. Through algorithmic optimization, it determines the optimal speed and provides the top three speeds with the highest TCE rankings for decision-making reference. The flowchart of this method is shown below. Figure 1 As shown, the steps are as follows:

[0044] I. Parameter Acquisition Steps: Based on the standard route basic data table provided by the business side, acquire the ship's historical navigation parameters and operating cost parameters. The historical navigation parameters include empty speed, full-load speed, heavy fuel oil consumption and light fuel oil consumption at empty speed, heavy fuel oil consumption and light fuel oil consumption at full-load speed, route distance with empty heavy fuel oil and light fuel oil, route distance with full load heavy fuel oil and light fuel oil, total cargo capacity of the route, route loading time, route loading time, and route unloading time. The operating cost parameters include heavy fuel oil price, light fuel oil price, route loading fuel cost, route loading fuel cost, route unloading fuel cost, route port usage fees, basic rates, and commission rates. Preferably, the ship navigation parameters also include light fuel oil consumption per unit time under loading, heavy fuel oil consumption per unit time under loading, heavy fuel oil consumption per unit time during loading, light fuel oil consumption per unit time during loading, heavy fuel oil consumption per unit time during unloading, and light fuel oil consumption per unit time during unloading.

[0045] Specifically, we first collected and organized historical navigation parameters and operating cost parameters of ships provided by the business side, including multiple key indicators such as speed, fuel consumption, route port usage fees, sailing time, and loading and unloading efficiency, to ensure that the input of the model can accurately reflect all aspects of operating costs and revenue. Furthermore, the basic data tables include standard route basic data tables, vessel speed and fuel consumption basic data tables, WS index data tables, and oil price index data tables. The standard route basic data tables mainly include fixed parameters and costs related to the route, such as the route's empty heavy and light fuel distance, the route's fully loaded heavy and light fuel distance, the route's total cargo capacity, the route's loading and unloading time, the route's loading and unloading time, the route's port usage fees, basic rates, and commission rates. The vessel speed and fuel consumption basic data tables are used to analyze the relationship between speed and fuel consumption, mainly including vessel speed and fuel consumption parameters, such as empty speed, fully loaded speed, heavy and light fuel consumption at empty speed, and heavy and light fuel consumption at fully loaded speed. The WS index data tables provide shipping market-related indices, such as the WS index, used to assess the impact of market conditions on operating costs. The oil price index data tables provide fuel price parameters, such as heavy fuel oil prices and light fuel oil prices, used to calculate fuel costs. The fuel costs for loading, unloading, and transporting along the route are calculated based on a combination of ship speed and fuel consumption data tables and oil price index data tables. The parameters used in this invention are represented and explained in Table 1.

[0046] Table 1

[0047]

[0048]

[0049] II. Steps for constructing the speed and fuel consumption model: Based on the empty speed and the heavy fuel consumption and light fuel consumption at the empty speed, machine learning algorithms are used to construct the empty speed and heavy fuel consumption model and the empty speed and light fuel consumption model, respectively; Based on the full-load speed and the heavy fuel consumption and light fuel consumption at the full-load speed, machine learning algorithms are used to construct the full-load speed and heavy fuel consumption model and the full-load speed and light fuel consumption model, respectively.

[0050] This step, based on real-world historical voyage data, uses machine learning algorithms to model the speed-fuel consumption relationship of ships. This model provides a formulaic expression of discrete data and is applied to the construction of a speed optimization model. This process involves verifying model assumptions and evaluating the fitting effect to ensure the model's fitting results are both accurate and reliable. Specifically, analyzing the speed-fuel consumption data of ships in both empty and fully loaded states reveals a generally linear relationship. Therefore, a linear regression model from machine learning algorithms is used to model the speed-fuel consumption relationship of each ship, based on the empty speed V. bdAnd the heavy fuel consumption F at the ship's unloaded speed bafo And light oil fuel consumption F bago Heavy fuel consumption F at full load speed lafo And light oil fuel consumption F lago And a linear regression model was used to construct models F for unloaded speed and heavy fuel consumption, respectively. bafo (V bd And the model F of no-load speed and light fuel consumption bago (V bd Based on full-load speed V la The study also investigated the heavy fuel consumption and light fuel consumption at full load speed, and constructed linear regression models F for full load speed and heavy fuel consumption, respectively. lafo (V la ) and full-load speed and light fuel consumption model F lago (V la ).

[0051] III. Standard Route Fuel Cost Calculation Steps: Calculate the estimated empty-load fuel cost of the route based on the empty-load speed and heavy fuel consumption model, the empty-load speed and light fuel consumption model, the route's empty heavy fuel distance, the route's empty light fuel distance, heavy fuel price, light fuel price, and empty-load speed. Calculate the estimated full-load fuel cost of the route based on the full-load speed and heavy fuel consumption model, the full-load speed and light fuel consumption model, the route's full-load heavy fuel distance, the route's full-load light fuel distance, heavy fuel price, light fuel price, and full-load speed. Combine the estimated empty-load fuel cost, the estimated full-load fuel cost, the route's fuel cost when loaded, the route's fuel cost when loaded, and the route's fuel cost when unloaded to obtain the standard route fuel cost.

[0052] Specifically, the estimated empty fuel cost (BFC) of a route is first calculated based on the empty speed and heavy fuel consumption model, the empty speed and light fuel consumption model, the empty heavy fuel distance of the route, the empty light fuel distance of the route, the heavy fuel price, the light fuel price, and the empty speed, according to the following formula:

[0053]

[0054] Then, based on the full-load speed and heavy fuel consumption model, the full-load speed and light fuel consumption model, the route's full-load heavy fuel distance, the route's full-load light fuel distance, heavy fuel price, light fuel price, and full-load speed, the estimated full-load fuel cost (LFC) for the route is calculated using the following formula:

[0055]

[0056] Then, based on the route's load time, light fuel consumption per unit time, heavy fuel consumption per unit time, heavy fuel price, and light fuel price, the route's load fuel cost (WFC) is calculated using the following formula:

[0057] WFC=D w F wafo P fo +D w F wago P go (3)

[0058] The fuel cost (LF) for loading the route is calculated based on the loading time, heavy fuel oil consumption per unit time, light fuel oil consumption per unit time, heavy fuel oil price, and light fuel oil price, using the following formula:

[0059] LF=D l F lofo P fo +D l F logo P go (4)

[0060] Then, based on the unloading time of the route, the fuel consumption of heavy fuel oil per unit time of unloading, the fuel consumption of light fuel oil per unit time of unloading, the price of heavy fuel oil, and the price of light fuel oil, the unloading fuel cost DF of the route is calculated according to the following formula:

[0061] DF=D d F difo P fo +D d F digo P go (5)

[0062] Finally, the calculated estimated empty fuel cost, estimated full-load fuel cost, fuel cost when the route is loaded, fuel cost when the route is loaded, and fuel cost when the route is unloaded are combined to form the standard route fuel cost (BFC+LFC+WFC+LF+DF).

[0063] IV. Steps for establishing the equivalent time charter fee calculation model: Calculate the expected total revenue of the route based on the total cargo volume, basic rates, and commission rates; calculate the total voyage time based on the route's full-load heavy fuel oil distance, full-load light fuel oil distance, empty-load heavy fuel oil distance, empty-load light fuel oil distance, full-load speed, empty-load speed, loading and unloading time, and unloading time; and establish the equivalent time charter fee calculation model, i.e., the TCE model, based on the standard route fuel cost, the expected total revenue of the route, the total voyage time, and the route port usage fees.

[0064] Specifically, the expected total revenue Rev for the route is first calculated based on the route's total cargo volume, base rates, and commission rates, using the following formula:

[0065] Rev=WS×FR×Q×(1-comm) (6)

[0066] In the above formula, WS represents the benchmark rate index, FR represents the benchmark rate, Q represents the total cargo volume of the route, and comm represents the commission rate.

[0067] Then, based on the route's distance with full load of heavy fuel oil, the route's distance with full load of light fuel oil, the route's distance with empty load of heavy fuel oil, the route's distance with empty load of light fuel oil, the speed with full load, the speed with empty load, the route's loading and unloading time, the total voyage time VD is calculated using the following formula:

[0068]

[0069] Based on standard route fuel costs, estimated total revenue (Rev), total voyage time (VD), and port usage fees (PC), an equivalent time charter rate calculation model (TCE) is established. TCE measures the vessel's equivalent revenue (i.e., average daily revenue) over a specific time period and is calculated using the following formula:

[0070]

[0071] V. Speed ​​Optimization Model Construction Steps: Under the condition of not considering light fuel consumption and fixed full-load speed, the equivalent time charter fee calculation model is simplified to obtain a speed optimization model with empty speed as the core independent variable, heavy fuel consumption corresponding to empty speed as the auxiliary independent variable, and equivalent time charter fee as the dependent variable.

[0072] Specifically, the equivalent time lease rental calculation model TCE is expanded, that is, equation (8) is expanded to obtain:

[0073]

[0074] Under the condition of not considering light fuel consumption and a fixed full-load speed (since the full-load speed, although fluctuating, has a small error range and is affected by ocean currents, engine aging, etc., and therefore does not have a fixed fluctuation range, it is not a primary concern); similarly, light fuel is generally only used in small cargo ships and passenger ships. However, this invention focuses on the results for large ships under macroscopic conditions. Therefore, heavy fuel consumption is retained while light fuel consumption is ignored, and the full-load speed is set as a constant. The equivalent time charter rate calculation model is simplified to obtain the result based on the empty speed V. bd The core independent variable is the heavy fuel consumption F(V) corresponding to the no-load speed. bd The speed optimization model, with ) as the auxiliary independent variable and the equivalent time charter rate (TCE) as the dependent variable, that is, without considering light fuel and with a fixed speed at full load, can be transformed into:

[0075]

[0076] in:

[0077]

[0078] Equation (10) is the speed optimization model per unit time of the present invention, where a>0, b>0, c>0, d>0.

[0079] VI. Linear Processing and Optimal Speed ​​Calculation Steps: Based on the linear relationship equation between speed and fuel consumption established using the linear regression method, the speed optimization model is linearized to obtain the processed speed optimization model; and the optimal speed of the processed speed optimization model is calculated using the speed optimization algorithm to maximize the equivalent time charter rate, thus completing the optimization of the ship's speed.

[0080] Specifically, the optimal speed in equation (10) can be expressed as:

[0081]

[0082] Where x is the empty speed, y is the heavy fuel consumption corresponding to the empty speed, d is the route-related parameters, and a, b, and c are also affected by user input parameters.

[0083] Then, based on the linear relationship equation between speed and fuel consumption established by the linear regression method, the speed optimization model is linearized to obtain the processed speed optimization model. That is, by linearly mapping the auxiliary independent variable to Equation (11), the model complexity is reduced. According to the linear relationship equation between speed and fuel consumption y=kx-i,k>0,i>0 established by the linear regression method, substituting it into Equation (11) yields:

[0084]

[0085] Then, the optimal speed of the processed speed optimization model is calculated using a speed optimization algorithm to maximize the equivalent time charter rate. That is, the maximum value of equation (12) is obtained by solving the maximum value of equation (12) and differentiating the function f(x) in equation (12):

[0086]

[0087] Based on the current results, the speed is positively monotonic, i.e., ad-bkd-cbi > 0. Therefore, the three highest speeds with increasing TCE are chosen as the three optimal TCE values, and the calculation is completed. By differentiating the processed speed optimization model, the speed optimization algorithm analyzes the derivative curve of the TCE function to determine local maxima. This process involves not only identifying function extrema but also in-depth analysis of model stability and robustness, ensuring that the selected speeds maintain optimal economic performance under different operating conditions. Finally, the algorithm outputs the top three speeds corresponding to the highest TCE (including parameter information and calculation process) for decision support. Figure 2The system displays the calculation results of the ship's historical best speed. Users can filter by update time, route name, and vessel name. The calculation results include the top three largest TCEs, namely Best TCE, Top 2 TCE, and Top 3 TCE, as well as the corresponding empty speed BallastSpeed ​​and full-load speed Laden Speed. Figure 3 It displays the specific optimal speed details for a certain ship, showing the empty and full load speeds corresponding to the three maximum TCEs, as well as the corresponding data such as oil amount, ballast speed, and voyage days.

[0088] This invention also relates to a ship speed optimization system based on standard routes. This system corresponds to the aforementioned ship speed optimization method based on standard routes and can be understood as a system that implements the above method. It includes, in sequence, a parameter acquisition module, a speed and fuel consumption model construction module, a standard route fuel cost calculation module, an equivalent time charter rate calculation model establishment module, a speed optimization model construction module, and a linear processing and optimal speed calculation module. Specifically,

[0089] The parameter acquisition module obtains ship navigation parameters and operating cost parameters based on the standard route basic data table provided by the business terminal. The ship navigation parameters include empty speed, full-load speed, heavy fuel oil consumption and light fuel oil consumption at empty speed, heavy fuel oil consumption and light fuel oil consumption at full-load speed, route empty heavy fuel oil and light fuel oil distance, route full-load heavy fuel oil and light fuel oil distance, route total cargo capacity, route waiting time for loading, route loading time, and route unloading time. The operating cost parameters include heavy fuel oil price, light fuel oil price, route waiting fuel cost, route loading fuel cost, route unloading fuel cost, route port usage fees, basic rates, and commission rates.

[0090] The speed and fuel consumption model building module constructs models based on empty speed and heavy fuel consumption and light fuel consumption at empty speed using machine learning algorithms; and constructs models based on full-load speed and heavy fuel consumption and light fuel consumption using machine learning algorithms.

[0091] The standard route fuel cost calculation module calculates the estimated empty-load fuel cost of a route based on empty-load speed and heavy fuel consumption models, empty-load speed and light fuel consumption models, the route's empty heavy fuel distance, the route's empty light fuel distance, heavy fuel price, light fuel price, and empty-load speed. It also calculates the estimated full-load fuel cost of a route based on full-load speed and heavy fuel consumption models, full-load speed and light fuel consumption models, the route's full-load heavy fuel distance, the route's full-load light fuel distance, heavy fuel price, light fuel price, and full-load speed. Finally, it combines the estimated empty-load fuel cost, the estimated full-load fuel cost, the route's fuel cost when loaded, the route's fuel cost when loaded, and the route's fuel cost when unloaded to obtain the standard route fuel cost.

[0092] The equivalent time charter fee calculation model module calculates the expected total revenue of the route based on the total cargo volume, basic rates, and commission rates; it calculates the total voyage time based on the route's full-load heavy fuel oil distance, full-load light fuel oil distance, empty-load heavy fuel oil distance, empty-load light fuel oil distance, full-load speed, empty-load speed, loading and unloading time, and unloading time; and it establishes an equivalent time charter fee calculation model based on standard route fuel costs, expected total revenue, total voyage time, and port usage fees.

[0093] The speed optimization model construction module simplifies the equivalent time charter rate calculation model under the condition of not considering light fuel consumption and fixed full-load speed, and obtains a speed optimization model with empty speed as the core independent variable, heavy fuel consumption corresponding to empty speed as the auxiliary independent variable, and equivalent time charter rate as the dependent variable.

[0094] The linear processing and optimal speed calculation module linearizes the speed optimization model based on the linear relationship equation between speed and fuel consumption established by the linear regression method, resulting in a processed speed optimization model. The module then uses a speed optimization algorithm to calculate the optimal speed of the processed speed optimization model to maximize the equivalent time charter rate, thus completing the optimization of the ship's speed.

[0095] Preferably, in the speed and fuel consumption model construction module, the machine learning algorithm adopts a linear regression model, and the regression coefficients in the linear regression model are estimated by the least squares method to ensure the accuracy of the linear regression model.

[0096] Preferably, in the parameter acquisition module, the ship navigation parameters also include light fuel consumption per unit time under load, heavy fuel consumption per unit time under load, heavy fuel consumption per unit time during loading, light fuel consumption per unit time during loading, heavy fuel consumption per unit time during unloading, and light fuel consumption per unit time during unloading.

[0097] Preferably, in the parameter acquisition module, the fuel cost for the route is calculated based on the route loading time, light fuel consumption per unit time, heavy fuel consumption per unit time, heavy fuel price, and light fuel price; the loading fuel cost is calculated based on the route loading time, heavy fuel consumption per unit time, light fuel consumption per unit time, heavy fuel price, and light fuel price; and the unloading fuel cost is calculated based on the route unloading time, heavy fuel consumption per unit time, light fuel consumption per unit time, heavy fuel price, and light fuel price.

[0098] In the standard route fuel cost calculation model establishment module, the calculated expected empty-load fuel cost, expected full-load fuel cost, fuel cost when the route is loaded, fuel cost when the route is loaded, and fuel cost when the route is unloaded are combined to form the standard route fuel cost.

[0099] This invention provides an objective and scientific method and system for detecting container damage. By modeling the relationship between ship speed and fuel consumption based on historical navigation parameters and employing machine learning algorithms, it achieves a formulaic expression of discrete data, ensuring that the model's fitting results are both accurate and reliable. Furthermore, by integrating key information such as speed, fuel consumption, and fuel price from historical navigation parameters and operating cost parameters, and using the equivalent time charter rate (TCE) as the objective and the empty speed and heavy fuel consumption function at empty speed as independent variables, it achieves a precise assessment of transportation costs and revenues at different speeds. It can accurately capture the relationship between speed and TCE, more accurately describe the ship speed optimization problem, and achieve a precise assessment of transportation costs and revenues at different speeds, greatly improving shipping efficiency and reducing operating costs.

[0100] It should be noted that the specific embodiments described above enable those skilled in the art to more fully understand the present invention, but do not limit the present invention in any way. Therefore, although the present invention has been described in detail with reference to the accompanying drawings and embodiments, those skilled in the art should understand that modifications or equivalent substitutions can still be made to the present invention. In short, all technical solutions and improvements that do not depart from the spirit and scope of the present invention should be covered within the protection scope of the present invention patent.

Claims

1. A method for optimizing ship speed based on a standard route, characterized in that, Includes the following steps: Parameter acquisition steps: Based on the standard route basic data table provided by the business side, obtain the ship's historical navigation parameters and operating cost parameters. The ship's historical navigation parameters include empty speed, full-load speed, heavy fuel oil consumption and light fuel oil consumption at empty speed, heavy fuel oil consumption and light fuel oil consumption at full-load speed, route empty heavy fuel oil and light fuel oil distance, route full-load heavy fuel oil and light fuel oil distance, route total cargo capacity, route waiting time for loading, route loading time, and route unloading time. The operating cost parameters include heavy fuel oil price, light fuel oil price, route waiting fuel cost, route loading fuel cost, route unloading fuel cost, route port usage fees, basic rates, and commission rates. Steps for constructing the speed and fuel consumption model: Based on the empty speed and the heavy fuel consumption and light fuel consumption at the empty speed, machine learning algorithms are used to construct the empty speed and heavy fuel consumption model and the empty speed and light fuel consumption model, respectively; Based on the full-load speed and the heavy fuel consumption and light fuel consumption at the full-load speed, machine learning algorithms are used to construct the full-load speed and heavy fuel consumption model and the full-load speed and light fuel consumption model, respectively. The standard route fuel cost calculation steps are as follows: First, calculate the estimated empty-load fuel cost based on the empty-load speed and heavy fuel consumption model, the empty-load speed and light fuel consumption model, the empty heavy fuel distance of the route, the empty light fuel distance of the route, the heavy fuel price, the light fuel price, and the empty-load speed. Second, calculate the estimated full-load fuel cost based on the full-load speed and heavy fuel consumption model, the full-load speed and light fuel consumption model, the full-load heavy fuel distance of the route, the full-load light fuel distance of the route, the heavy fuel price, the light fuel price, and the full-load speed. Finally, combine the estimated empty-load fuel cost, the estimated full-load fuel cost, the fuel cost when the route is loaded, the fuel cost when the route is loaded, and the fuel cost when the route is unloaded to obtain the standard route fuel cost. The steps for establishing an equivalent time charter fee calculation model are as follows: Calculate the estimated total revenue of the route based on the total cargo volume, base rates, and commission rates; calculate the total voyage time based on the route's full-load heavy fuel oil distance, full-load light fuel oil distance, empty-load heavy fuel oil distance, empty-load light fuel oil distance, full-load speed, empty-load speed, loading and unloading time, and unloading time; and establish an equivalent time charter fee calculation model based on standard route fuel costs, estimated total revenue, total voyage time, and port usage fees. Steps for constructing the speed optimization model: Under the condition of not considering light fuel consumption and fixed full-load speed, the equivalent time charter fee calculation model is simplified to obtain a speed optimization model with empty speed as the core independent variable, heavy fuel consumption corresponding to empty speed as the auxiliary independent variable, and equivalent time charter fee as the dependent variable. The linear processing and optimal speed calculation steps are as follows: Based on the linear relationship equation between speed and fuel consumption established by the linear regression method, the speed optimization model is linearized to obtain the processed speed optimization model; and the optimal speed of the processed speed optimization model is calculated by the speed optimization algorithm to maximize the equivalent time charter rate, thus completing the optimization of the ship's speed.

2. The method for optimizing ship speed based on a standard route according to claim 1, characterized in that, In the process of constructing the speed and fuel consumption model, the machine learning algorithm uses a linear regression model and estimates the regression coefficients in the linear regression model using the least squares method to ensure the accuracy of the linear regression model.

3. The method for optimizing ship speed based on a standard route according to claim 1, characterized in that, In the parameter acquisition step, the ship navigation parameters also include light fuel consumption per unit time under load, heavy fuel consumption per unit time under load, heavy fuel consumption per unit time during loading, light fuel consumption per unit time during loading, heavy fuel consumption per unit time during unloading, and light fuel consumption per unit time during unloading.

4. The method for optimizing ship speed based on a standard route according to claim 3, characterized in that, In the parameter acquisition step, the fuel cost for the route is calculated based on the route loading time, light fuel consumption per unit time, heavy fuel consumption per unit time, heavy fuel price, and light fuel price; the loading fuel cost is calculated based on the route loading time, heavy fuel consumption per unit time, light fuel consumption per unit time, heavy fuel price, and light fuel price; and the unloading fuel cost is calculated based on the route unloading time, heavy fuel consumption per unit time, light fuel consumption per unit time, heavy fuel price, and light fuel price. In the standard route fuel cost calculation model establishment step, the calculated expected empty-load fuel cost, expected full-load fuel cost, fuel cost when the route is loaded, fuel cost when the route is loaded, and fuel cost when the route is unloaded are combined and used as the standard route fuel cost.

5. The method for optimizing ship speed based on a standard route according to any one of claims 1 to 4, characterized in that, In the process of constructing the speed optimization model, the simplification of the equivalent time charter rate calculation model refers to: for large ships, retaining heavy fuel oil consumption and ignoring light fuel oil consumption, and setting the full-load speed as a constant to simplify the equivalent time charter rate calculation model, thereby obtaining a speed optimization model with the empty-load speed as the core independent variable, the heavy fuel oil consumption corresponding to the empty-load speed as the auxiliary independent variable, and the equivalent time charter rate as the dependent variable.

6. The method for optimizing ship speed based on a standard route according to claim 5, characterized in that, In the linear processing and optimal speed calculation steps, the linearization of the speed optimization model is based on the linear relationship between the empty speed and heavy fuel consumption in the linear equation. The nonlinear speed optimization model is linearly mapped to replace the auxiliary independent variable heavy fuel consumption, resulting in the linearized speed optimization model. The speed optimization algorithm includes: The derivative curve of the dependent variable of the optimized equivalent time charter rate is obtained by taking the derivative of the processed speed optimization model. The derivative curve is analyzed to determine local maxima, including identifying extreme points and evaluating the stability and robustness of the processed speed optimization model; Output the speed results corresponding to the top three local maximum values ​​of the equivalent time charter rate dependent variable.

7. A ship speed optimization system based on a standard route, characterized in that, The system includes, in sequence, a parameter acquisition module, a speed and fuel consumption model construction module, a standard route fuel cost calculation module, an equivalent time charter rate calculation model establishment module, a speed optimization model construction module, and a linear processing and optimal speed calculation module. The parameter acquisition module obtains ship navigation parameters and operating cost parameters based on the standard route basic data table provided by the business terminal. The ship navigation parameters include empty speed, full-load speed, heavy fuel oil consumption and light fuel oil consumption at empty speed, heavy fuel oil consumption and light fuel oil consumption at full-load speed, route empty heavy fuel oil and light fuel oil distance, route full-load heavy fuel oil and light fuel oil distance, route total cargo capacity, route waiting time for loading, route loading time, and route unloading time. The operating cost parameters include heavy fuel oil price, light fuel oil price, route waiting fuel cost, route loading fuel cost, route unloading fuel cost, route port usage fees, basic rates, and commission rates. The speed and fuel consumption model building module constructs models based on empty speed and heavy fuel consumption and light fuel consumption at empty speed using machine learning algorithms; and constructs models based on full-load speed and heavy fuel consumption and light fuel consumption using machine learning algorithms. The standard route fuel cost calculation module calculates the estimated empty-load fuel cost of a route based on empty-load speed and heavy fuel consumption models, empty-load speed and light fuel consumption models, the route's empty heavy fuel distance, the route's empty light fuel distance, heavy fuel price, light fuel price, and empty-load speed. It also calculates the estimated full-load fuel cost of a route based on full-load speed and heavy fuel consumption models, full-load speed and light fuel consumption models, the route's full-load heavy fuel distance, the route's full-load light fuel distance, heavy fuel price, light fuel price, and full-load speed. Finally, it combines the estimated empty-load fuel cost, the estimated full-load fuel cost, the route's fuel cost when loaded, the route's fuel cost when loaded, and the route's fuel cost when unloaded to obtain the standard route fuel cost. The equivalent time charter fee calculation model module calculates the expected total revenue of the route based on the total cargo volume, basic rates, and commission rates; it calculates the total voyage time based on the route's full-load heavy fuel oil distance, full-load light fuel oil distance, empty-load heavy fuel oil distance, empty-load light fuel oil distance, full-load speed, empty-load speed, loading and unloading time, and unloading time; and it establishes an equivalent time charter fee calculation model based on standard route fuel costs, expected total revenue, total voyage time, and port usage fees. The speed optimization model construction module simplifies the equivalent time charter rate calculation model under the condition of not considering light fuel consumption and fixed full-load speed, and obtains a speed optimization model with empty speed as the core independent variable, heavy fuel consumption corresponding to empty speed as the auxiliary independent variable, and equivalent time charter rate as the dependent variable. The linear processing and optimal speed calculation module linearizes the speed optimization model based on the linear relationship equation between speed and fuel consumption established by the linear regression method, resulting in a processed speed optimization model. The module then uses a speed optimization algorithm to calculate the optimal speed of the processed speed optimization model to maximize the equivalent time charter rate, thus completing the optimization of the ship's speed.

8. The ship speed optimization system based on standard routes according to claim 7, characterized in that, In the speed and fuel consumption model construction module, the machine learning algorithm adopts a linear regression model, and the regression coefficients in the linear regression model are estimated by the least squares method to ensure the accuracy of the linear regression model.

9. The ship speed optimization system based on standard routes according to claim 7, characterized in that, In the parameter acquisition module, the ship navigation parameters also include light fuel consumption per unit time under load, heavy fuel consumption per unit time under load, heavy fuel consumption per unit time during loading, light fuel consumption per unit time during loading, heavy fuel consumption per unit time during unloading, and light fuel consumption per unit time during unloading.

10. The ship speed optimization system based on standard routes according to claim 9, characterized in that, The parameter acquisition module calculates the fuel cost for the route based on the loading time, light fuel consumption per unit time, heavy fuel consumption per unit time, heavy fuel price, and light fuel price; it also calculates the loading fuel cost based on the loading time, heavy fuel consumption per unit time, light fuel consumption per unit time, heavy fuel price, and light fuel price; and finally, it calculates the unloading fuel cost based on the unloading time, heavy fuel consumption per unit time, light fuel consumption per unit time, heavy fuel price, and light fuel price. In the standard route fuel cost calculation model establishment module, the calculated expected empty-load fuel cost, expected full-load fuel cost, fuel cost when the route is loaded, fuel cost when the route is loaded, and fuel cost when the route is unloaded are combined to form the standard route fuel cost.

Images