River-sea direct new energy ship variable time scale power scheduling method

Abstract

The application discloses a river-sea direct new energy ship variable time scale power scheduling method, comprising the following steps: constructing a full electric propulsion calculation model based on ship resistance, motion state, speed and cargo capacity; constructing a voyage-power generation two-stage coordinated power scheduling model based on the full electric propulsion calculation model, wherein the voyage-power generation two-stage coordinated power scheduling model comprises a voyage scheduling stage and a power scheduling stage; obtaining a ship optimal voyage plan based on the voyage scheduling stage; in the power scheduling stage, a variable time scale control strategy is obtained through an uncertainty power fluctuation quantitative index design based on a stochastic model predictive control method; and obtaining a variable time scale power scheduling optimal scheme based on the ship optimal voyage plan and the variable time scale control strategy. According to the full electric propulsion power fluctuation, the application adjusts the control time domain and time step of the next scheduling period in real time, and further improves the flexibility and reliability of ship navigation.

Description

Technical Field

[0001] This invention relates to the field of new energy vessel scheduling technology, and in particular to a variable time-scale power scheduling method for new energy vessels that can directly navigate between rivers and seas. Background Technology

[0002] New energy vessels that can directly transport goods between rivers and the sea meet both inland waterway and sea navigation standards, enabling direct cargo transport without the need for transshipment or barges, thus greatly reducing transportation costs and pollution emissions.

[0003] Effective power scheduling strategies will improve the energy utilization rate and reduce pollution emissions of river-sea direct new energy vessels. These vessels operate entirely on electric propulsion, with electric propulsion power accounting for over 70% of the total load power. Therefore, the accuracy of the all-electric propulsion calculation model determines the effectiveness of power scheduling. Typically, all-electric propulsion calculation models describe the relationship between propulsion power and speed under ideal conditions using an exponential function. In 2020, Fang introduced the additional resistance caused by wind and waves into the all-electric propulsion calculation model, proposing a robust power scheduling scheme that considers uncertainties. In 2021, Kyaw Hein recalculated the hull resistance coefficient and corrected the functional relationship between propulsion power and speed, reducing vessel operating costs by approximately 10%. Sun, considering the impact of additional resistance on propulsion power, directly optimized the vessel's voyage by minimizing the propulsion load demand over the entire voyage, resulting in a 52.44 kW reduction in average propulsion power over the entire voyage. All of the above studies assume the vessel is in uniform motion and neglect the influence of hull inertia, mass, and motion state on the all-electric propulsion process.

[0004] Ship power scheduling is a collaborative optimization problem involving voyage planning and power generation scheduling. In 2018, Tang transformed the ship power scheduling model into an unconstrained global optimization problem using a penalty function and solved it using an adaptive particle swarm optimization algorithm. Results showed that this method reduced the total cost by approximately 7%. In 2020, Li established a multi-objective mixed-integer linear programming model considering both economic and environmental factors to improve the energy utilization rate of microgrids on multi-energy ships. Results showed that this method reduced operating costs by 7% and gas emissions by 10.55 tons. In the same year, Li continued his research on integrated ship energy systems with flexible thermal loads, proposing a two-stage, multi-timescale robust scheduling strategy that coordinates voyage scheduling and power scheduling. Experiments demonstrated that this method reduced the total ship operating cost by $5475.4 and effectively adapted to flexible adjustments in ship voyage and load. In 2022, Fan proposed a two-stage robust scheduling framework (pre-voyage to in-voyage) with the objective of minimizing the total operating cost of a hydrogen fuel cell ship power system and solved it using a two-level C&CG algorithm. Results showed that this solution strategy can guarantee the economic operation of the system under uncertainty conditions, reducing the total cost by 4.6%. The above studies all treat ship power scheduling as an optimization problem and use different offline optimization strategies to solve it. This invention treats the ship power scheduling problem as a control problem, uses model predictive control technology to obtain ship scheduling schemes, utilizes predictive models to analyze the future state change trends of the system, and achieves real-time ship scheduling through rolling optimization, thereby enhancing the real-time performance and effectiveness of ship power scheduling.

[0005] Scholars have already applied MPC (Model Predictive Control) to ship power scheduling with good results. In 2018, Zhou studied two modes of operation for new energy ships: islanded operation and grid-connected operation. He proposed an optimal scheduling scheme based on model predictive control using rolling solutions. The results showed that the method has good robustness, reducing annual operating costs by $14,300 and fuel consumption and emissions by 3%. Tang studied the economic operation of ships using shore power and proposed an optimal scheduling scheme under different electricity prices and emission limits based on model predictive control. The results showed that the proposed scheduling scheme reduced the total cost of ships during berthing by 2%-5%, and effectively improved scheduling flexibility when electricity prices changed.

[0006] Regarding the power scheduling problem of river-sea direct-roaming new energy vessels, the following technical challenges remain: 1. River-sea direct-roaming new energy vessels operate in both river and sea environments, with various navigation conditions. Propulsion power calculations are affected by multiple variables, including vessel resistance, motion state, speed, and cargo load. Existing all-electric propulsion calculation models simply calculate based on the exponential relationship between propulsion power and speed, which is significantly inconsistent with actual navigation needs, resulting in large calculation errors and difficulty in ensuring the accuracy of power scheduling; 2. Currently, most vessel power scheduling solutions employ a fixed-time-scale global optimization strategy. River-sea direct-roaming new energy vessels are disturbed by environmental factors such as wind, waves, and currents, and the fluctuations in all-electric propulsion power and onboard photovoltaic power generation after disturbances are significant. Fixed-time-scale control causes scheduling commands to lag behind changes in environmental disturbances, decoupling the power scheduling scheme from the vessel's disturbance status, making it difficult to guarantee the real-time performance and reliability of power scheduling. Summary of the Invention

[0007] To achieve economical and low-carbon operation of new energy vessels capable of direct river-sea transport, this invention proposes a variable time-scale power scheduling strategy. First, considering the coupled changes of multiple variables such as vessel resistance, motion state, speed, and cargo load, a novel all-electric propulsion power calculation model is proposed. To solve the scheduling strategy, this invention proposes a two-stage power scheduling framework based on stochastic model predictive control, encompassing both range and power generation. Considering the quantification of uncertain power fluctuations, this invention proposes a novel variable time-scale control strategy.

[0008] This invention provides a variable time-scale power scheduling method for river-sea direct new energy vessels, comprising:

[0009] A calculation model for all-electric propulsion is constructed based on ship resistance, motion state, speed, and cargo capacity;

[0010] Based on the all-electric propulsion calculation model, a two-stage coordinated power scheduling model for range and power generation is constructed, wherein the two-stage coordinated power scheduling model for range and power generation includes a range scheduling stage and a power scheduling stage.

[0011] The optimal voyage plan for the ship is obtained based on the aforementioned voyage scheduling phase;

[0012] During the power scheduling phase, a variable time-scale control strategy is designed based on the stochastic model predictive control method by using an uncertainty power fluctuation quantification index.

[0013] The optimal power scheduling scheme based on the ship's optimal voyage plan and the variable time-scale control strategy is obtained.

[0014] Optionally, the all-electric propulsion calculation model is as follows:

[0015]

[0016] In the formula, V represents the total electric propulsion power during the time interval Δt. t Indicates the speed during time period t. represents the total weight of cargo on the ship during time period t, and represents the speed during time period Δt.

[0017] Optionally, the voyage scheduling phase determines the optimal voyage plan for the ship with the goal of minimizing the ship's operating costs, while ensuring the ship arrives at the port on time. The ship's operating costs include: fuel costs, equipment maintenance costs, pollution gas treatment costs, diesel generator start-up and shutdown costs, and shore power costs.

[0018] Optionally, the mathematical model for the flight scheduling phase is constructed as follows:

[0019]

[0020]

[0021]

[0022]

[0023]

[0024]

[0025]

[0026]

[0027]

[0028]

[0029]

[0030]

[0031]

[0032]

[0033]

[0034]

[0035]

[0036] in, Represents fuel cost; Represents the cost of environmental governance; This represents the start-up cost of a diesel generator set; This represents the cost of shutting down a diesel generator set; Represents the total operating and maintenance cost of all equipment on the ship; Represents the cost of purchasing electricity while the vehicle is parked; Represents the maintenance coefficient; ξ CI Represents electricity price; η represents the inbound / outbound port speed limit factor; η v This represents the maximum permissible speed adjustment factor; v n T represents the rated speed; p Represents the time period for sailing to and from the port; T c T represents the cruise time period; b Represents the duration of berthing; δ D This represents the maximum deviation in sailing distance; Represents the rated voyage distance to the nth port; This represents the sailing distance of a new energy vessel that directly connects the river and the sea during time period t. Dis represents the rated voyage distance to the destination port; nt This represents the distance traveled by new energy vessels directly from river to sea to their final destination.

[0037] Equation (3) represents fuel cost; Equation (4) represents equipment operation and maintenance cost; Equation (5) represents pollution gas treatment cost; Equations (6)-(7) represent diesel generator set start-up and shutdown cost; Equation (8) represents shore power cost; Equations (9)-(11) represent the speed limits that ships must meet when entering and leaving ports, cruising and berthing when sailing in inland waterways and sea areas; Equations (12)-(13) represent navigation distance limits to ensure that ships arrive at port on time; Equation (14) indicates that diesel generator sets are prohibited from being used for power supply in inland waterways to meet the high standards for pollution discharge in inland waterways; Equations (15)-(16) represent the balance of electric and thermal power when ships are sailing; Equations (17)-(18) represent the balance of electric and thermal power when ships are berthing.

[0038] Optionally, the objective function of the mathematical model for the power scheduling phase is:

[0039]

[0040] The constraints on the parameters of the objective function are the same as the constraints on the corresponding parameters in the mathematical model of the flight scheduling phase.

[0041] Optionally, in the power scheduling phase, a variable time-scale control strategy is designed based on the stochastic model predictive control method using an uncertainty power fluctuation quantification index. The mathematical model of the uncertainty power fluctuation quantification index, constructed based on utility theory, is as follows:

[0042]

[0043] In the formula, f u This is the risk adaptation function, and its value is related to the uncertainty power fluctuation quantification index.

[0044] Optionally, based on the aforementioned uncertainty power fluctuation quantification index, the formula for determining the time-scale decision-making mechanism using a stochastic model predictive control method is as follows:

[0045]

[0046] In the formula, N C Indicates the control interval. and These represent the minimum and maximum values ​​that can be adjusted on the time scale, respectively.

[0047] Optionally, the stochastic model predictive control method determines the variable time-scale power scheduling strategy by adjusting two parameters: the control interval and the control time domain.

[0048] The control time domain includes several control intervals.

[0049] Optionally, the variable time-scale power scheduling process based on the stochastic model predictive control method includes:

[0050] S1. Set the initial control time domain and control interval;

[0051] S2. Import the optimal voyage plan for the vessel;

[0052] S3. Based on the ship's optimal voyage plan, perform uncertainty processing in the control time domain and solve the power scheduling strategy in the current control time domain;

[0053] S4. Based on the power scheduling strategy under the current control time domain, roll to the next time domain control, update the control time domain and control interval based on the variable time scale control strategy, and perform uncertainty handling for the next time domain control.

[0054] S5. Repeat S3-S4 to obtain the optimal power scheduling scheme with variable time scale.

[0055] The present invention has the following technical effects:

[0056] This invention proposes a novel all-electric propulsion calculation model that considers the multivariate coupling of ship resistance, motion state, speed, and cargo capacity. It performs multivariate coupling modeling of the dynamic process of all-electric propulsion for new energy ships that can directly reach river and sea, thereby improving the accuracy of propulsion power calculation.

[0057] This invention proposes a two-stage coordinated power scheduling framework for voyage and power generation to ensure economical and green navigation of ships. The voyage scheduling stage acquires and issues the optimal voyage plan; the power generation scheduling stage utilizes stochastic model predictive control technology to achieve real-time power scheduling.

[0058] This invention designs a variable time-scale power scheduling strategy. Based on the SMPC method, a variable time-scale control strategy is proposed, which adjusts the control time domain and time step size of the next scheduling cycle in real time according to the power fluctuation of all-electric propulsion, thereby further improving the flexibility and reliability of ship navigation. Attached Figure Description

[0059] To more clearly illustrate the technical solutions in the embodiments of the present invention or the prior art, the drawings used in the embodiments will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0060] Figure 1 This is a schematic diagram of the integrated energy system for river-sea direct new energy vessels in an embodiment of the present invention;

[0061] Figure 2 This is a schematic diagram of ship propulsion in an embodiment of the present invention;

[0062] Figure 3 This is a two-stage range-power generation scheduling framework in this embodiment of the invention;

[0063] Figure 4 This is a schematic diagram of the variable time scale control strategy in an embodiment of the present invention;

[0064] Figure 5 This is a flowchart of the variable time scale scheduling strategy in an embodiment of the present invention;

[0065] Figure 6 The diagram shows the interval prediction results in the embodiment of the present invention. In the diagram, (a) represents the PV prediction result, (b) represents the wave height prediction result, (c) represents the wind speed prediction result, (d) represents the wind direction prediction result, (e) represents the electrical load prediction result, and (f) represents the heat load prediction result.

[0066] Figure 7 This refers to the speed scheduling result in the embodiments of the present invention;

[0067] Figure 8 This refers to the propulsion power results in the embodiments of the present invention;

[0068] Figure 9The variable time-scale power scheduling strategy proposed in this embodiment of the invention is shown in Figure (a), which represents the electric power scheduling strategy, Figure (b) represents the thermal power scheduling strategy, and Figure (c) represents the scheduling time scale.

[0069] Figure 10 The multi-timescale power scheduling strategy in this embodiment of the invention is shown in Figure (a), which represents the electrical power scheduling strategy, and Figure (b) represents the thermal power scheduling strategy.

[0070] Figure 11 Figure 7 shows the results of Example 7 in the embodiments of the present invention. Figure (a) represents the electric power scheduling strategy, Figure (b) represents the thermal power scheduling strategy, and Figure (c) represents the scheduling time scale. Detailed Implementation

[0071] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0072] Example 1

[0073] This embodiment discloses a variable time-scale power scheduling method for new energy vessels that can directly navigate between rivers and seas, including the following steps:

[0074] 1. Mathematical Modeling

[0075] This invention focuses on the integrated energy system of new energy vessels capable of direct river and sea navigation. For example... Figure 1 As shown, ship loads can be categorized into three types: electrical loads, thermal loads, and propulsion loads. Propulsion loads account for over 70% of the total ship load power. Electrical loads are further divided into service loads, communication loads, and living loads, used to meet the ship's daily operational needs. Thermal loads primarily ensure the operating temperature of cabins and critical equipment. Ship energy is supplied by photovoltaic power generation systems, diesel power generation systems, combined heat and power systems, electrical energy storage, and thermal storage tanks. During the ship's navigation phase, a comprehensive energy system ensures the fulfillment of all load requirements.

[0076] This embodiment establishes mathematical models for each power unit in the ship's integrated energy system. Unlike traditional propulsion power calculation models, this study considers the coupled changes of multiple variables such as ship resistance, motion state, speed, and cargo load during electric propulsion, and proposes a new calculation model for all-electric propulsion of ships.

[0077] A calculation model for all-electric propulsion is constructed based on ship resistance, motion state, speed, and cargo capacity. The specific implementation process includes:

[0078] 1.1 Modeling of the all-electric propulsion process

[0079] Ship navigation includes cruising, berthing, and port entry / exit, requiring planning of voyage and speed. The power of all-electric propulsion is closely related to ship speed. Typically, it is assumed that propulsion power and ship speed have the following typical exponential function relationship:

[0080]

[0081] The power of the ship's all-electric propulsion during time period t; v t Represents the speed during time period t; α PL and β PL This represents the advancement coefficient.

[0082] This model assumes that the electric propulsion power remains constant when the ship speed is constant, without considering the resistance factors of the navigation environment. Moreover, when the ship is accelerating, the ship speed is low but the propulsion power is high. Equation (1) can only calculate the electric propulsion power when the ship is at a constant speed, but it cannot calculate the ship's propulsion power when accelerating or decelerating, nor can it describe the changes in propulsion power caused by changes in the ship's cargo load, resulting in a large deviation in the calculation of the ship's electric load and affecting the reliability of the ship's power scheduling. Therefore, this embodiment proposes a new ship all-electric propulsion power calculation model, which comprehensively considers the multivariate coupling relationship between ship resistance, motion state, speed, cargo load and propulsion power.

[0083] 1.1.1 Ship navigation motion balance

[0084] like Figure 2 As shown, during navigation, the bow direction of a ship is mainly affected by propulsion and resistance. Considering factors such as hydrostatic resistance, wind resistance, wave resistance, and cargo load, the following motion equilibrium model of the ship's propulsion process is established based on Newton's second law of motion:

[0085]

[0086]

[0087] In equations (2)-(3), m represents the total unloaded weight of the ship; This represents the total weight of the ship's cargo during time period t; This represents the magnitude of thrust provided by the propulsion system during time period t; This represents the total resistance experienced by the ship during time period t; This represents the total hydrostatic resistance of the ship during time period t; This represents the wind resistance experienced by the ship during time period t; This represents the increased wave resistance experienced by the ship during time period t.

[0088] 1.1.2 Ship Resistance

[0089] The accuracy of the ship resistance calculation model will determine the reliability of power scheduling. The total ship resistance consists of static water resistance and additional resistance. Since the new energy ship that travels directly from river to sea navigates in both inland waterways and sea areas, its total static water resistance model is as shown in equation (4):

[0090]

[0091] The static water resistance of a vessel navigating in a restricted Class III inland waterway is calculated as shown in equations (5) and (6):

[0092]

[0093]

[0094] In equations (5)-(6), S b F represents the wetted surface area of ​​the ship. rh denoted by Froude number for water depth; h is the channel depth; d is the ship's draft; L is the ship's length; B is the ship's beam; and W is the channel width.

[0095] According to Holtrop's formula, the hydrostatic resistance during navigation at sea is:

[0096]

[0097]

[0098]

[0099]

[0100]

[0101] In equations (7)-(11), For frictional resistance; C f The coefficient of frictional resistance; For attached resistance; Subsidies related to ship models and actual ships; 1+k is the shape factor; ρ is the water density; Re is the Reynolds number; S app Let (1+k2) be the wetted surface area of ​​the appendage. eq It can be a combination of various attached shape factors; C a This is the subsidy coefficient.

[0102] This invention focuses on bulk carriers, whose cargo capacity varies across different voyages. Considering the impact of cargo capacity on the ship's displacement (total weight) and resistance, the ship's wetted surface area is calculated as follows:

[0103]

[0104] Δ represents the displacement; C d is the conversion factor.

[0105] Therefore, the total hydrostatic resistance of a ship has an exponential relationship with its displacement:

[0106]

[0107] Additional resistance to ships is mainly caused by wind and waves. The additional resistance caused by wind is expressed as:

[0108]

[0109] In equation (14), ρ air air density; S air C is the projected area; air This refers to the drag coefficient; For wind speed; θ t For wind direction angle. According to the regulations of the International Ship Model Test Pool Conference, wave drag is calculated as follows:

[0110]

[0111] In the formula, Wave height; C b This is the square coefficient.

[0112] 1.1.3 Multivariable Coupled All-Electric Propulsion Power Model

[0113] When a ship is traveling at a constant speed, the resistance it experiences is equal to the propulsion force it experiences.

[0114]

[0115] The propulsion force provided for the electric propulsion system.

[0116] Let the power that drives a ship to overcome resistance be called the effective power:

[0117]

[0118] Represents the ship's effective power.

[0119] In actual navigation, the resistance experienced by a ship changes constantly, and the ship is often in a state of variable speed motion. As can be seen from equations (4) to (15), the ship's resistance changes exponentially with the speed. Therefore, calculating the effective power of a ship can be regarded as a problem of variable force doing work.

[0120] According to the method of infinitesimal elements, let dS be the distance traveled by the ship in time dt. The work done against resistance can be expressed as:

[0121] dW=Rtotal dS (18)

[0122] In the formula, dS can be expressed as:

[0123] dS=vdt (19)

[0124] The instantaneous acceleration during the dt stage is:

[0125]

[0126] Right now:

[0127]

[0128] Assuming the ship accelerates / decelerates uniformly, i.e., a is a constant, then the work done by the ship against resistance during the scheduling period t-t+Δt can be expressed using a definite integral as follows:

[0129]

[0130] W Δt It represents doing work by overcoming resistance.

[0131] The acceleration during the time interval Δt is replaced by the average acceleration:

[0132]

[0133] Finally, the average effective power of the ship during the time period Δt is:

[0134]

[0135] Simplifying equations (22)-(24) yields:

[0136]

[0137] Considering the transmission losses and fluctuations caused by components such as propellers and mechanical bearings in the propulsion system, the propulsion power can be expressed as:

[0138]

[0139] According to equations (12)-(13) and (22)-(25), when the cargo load of a ship changes, the resistance and the work done to overcome the resistance are different at the same speed, and the propulsion power will change accordingly. Considering the loading and unloading conditions of ships at various ports, the calculation model of the all-electric propulsion power of river-sea direct vessels can be expressed as:

[0140]

[0141] Equation (27) can be derived from the remaining formulas in Section 2.1, and will not be expanded here. The all-electric propulsion calculation model proposed in this invention is applied to minute-level ship power scheduling. Considering the calculation speed of the real-time scheduling scheme, the transient fluctuation process of the propeller and propulsion motor is simply represented by Equation (26).

[0142] 1.2 Diesel Generator System

[0143] The diesel generator is the main power supply equipment on board. The fuel consumption of the diesel generator is as shown in equation (28):

[0144]

[0145] This represents the fuel consumption of a diesel generator; This represents the output power of the diesel generator; This indicates the start / stop status of the diesel generator.

[0146] The power limit of the diesel generator is shown in equation (29), the gradeability limit is shown in equation (30), and the start-stop time constraints are shown in equations (31)-(32):

[0147]

[0148]

[0149]

[0150]

[0151] This represents the lower and upper limits of the diesel generator's output power; This represents the maximum gradeability limit of the diesel generator; Δt represents the dispatching period. These represent the continuous start-up and shutdown times of the diesel generator, respectively. These represent the minimum start-up and shutdown times for the diesel generator, respectively.

[0152] 1.3 Shipborne Photovoltaic Power Generation System

[0153] The power generation capacity of the shipborne photovoltaic system is shown in equation (33):

[0154]

[0155] Represents PV output power; S PV Represents the area covered by photovoltaic panels; Represents solar irradiance; η PV It represents photoelectric conversion power.

[0156] 1.4 Combined Heat and Power

[0157] The combined heat and power (CHP) system mainly consists of a gas turbine and a heat recovery unit. The gas turbine generates electricity by burning natural gas, and the waste heat is recovered and reused to generate thermal energy. Ignoring the influence of the external environment and combustion efficiency, the relationship between the CHP thermal power and electrical power is shown in equation (34):

[0158]

[0159] COP represents the output thermal power of CHP; CHP Represents electrothermal conversion efficiency; This represents the output electrical power of the CHP.

[0160] The fuel consumption cost and gas emissions of gas turbines are shown in equations (35)-(37):

[0161]

[0162] Represents CHP fuel cost; ξ NG Represents the price of natural gas; η CHP This represents power generation efficiency.

[0163]

[0164] Represents CHP gas emissions; ξ GE This represents the gas emission coefficient.

[0165] CHP must meet power constraints:

[0166]

[0167] This represents the lower and upper limits of CHP output.

[0168] 1.5 Energy Storage System

[0169] Ship energy storage systems consist of batteries and thermal storage tanks:

[0170] 1) Storage Batteries: River-sea direct-route new energy vessels are equipped with lithium iron phosphate batteries to smooth system power fluctuations and ensure the reliability of ship power supply. Their capacity and charging / discharging power should meet the following requirements:

[0171]

[0172] Represents battery capacity; τ ES This represents the battery's self-loss rate; These represent the battery's charging and discharging power, respectively; η ESG η ESDThis represents the charge / discharge efficiency.

[0173] Considering the battery's capacity limitations, rated power limitations, and the inability to charge and discharge simultaneously, the following constraints are established:

[0174]

[0175]

[0176]

[0177]

[0178] This represents the rated capacity of the battery; Represents minimum and maximum capacity limits; This indicates the charge / discharge status of the battery. This represents the upper limit of charging / discharging power.

[0179] To ensure that the battery is always available for use, its capacity state is limited to be consistent at the beginning and end of each scheduling cycle:

[0180]

[0181] This represents the initial capacity of the battery; This indicates the battery capacity at the end of the dispatch process.

[0182] 2) Thermal storage tank: Thermal storage tanks store system thermal energy to achieve system thermal energy scheduling and management.

[0183] Similar to a battery, the thermal storage tank model's capacity and thermal power can be expressed as follows:

[0184]

[0185] Represents the capacity of the thermal storage tank; τ represents the heat storage / heat release power of the thermal storage tank. HS Represents the self-loss rate; η HSC η HSD These represent heat storage efficiency and heat release efficiency, respectively.

[0186] Thermal storage tanks also need to meet limitations such as capacity, power, and the inability to charge and release heat simultaneously.

[0187]

[0188]

[0189]

[0190]

[0191] This represents the rated capacity of the thermal storage tank; This represents the minimum and maximum thermal storage tank capacity limits; This represents the heat storage and release state of the thermal storage tank; This represents the maximum power of heat storage and release.

[0192] 1.6 Electric Boiler

[0193] Electric boilers use electricity for heating to meet the operating temperature requirements of the engine room and important equipment.

[0194]

[0195] A power scheduling model for the two-stage coordinated operation of range and power generation is constructed based on the all-electric propulsion calculation model. The power scheduling model for the two-stage coordinated operation of range and power generation includes a range scheduling stage and a power scheduling stage.

[0196] The optimal voyage plan for the ship is obtained based on the voyage scheduling phase;

[0197] During the power scheduling phase, a variable time-scale control strategy is designed based on the stochastic model predictive control method by using an uncertainty power fluctuation quantification index.

[0198] The optimal power scheduling scheme based on the ship's optimal voyage plan and variable time-scale control strategy is obtained. The specific implementation process includes:

[0199] 2 Power Dispatch Strategy

[0200] To address the decoupling of scheduling time scale and ship disturbance conditions, this invention proposes a variable time scale power scheduling strategy based on SMPC. A two-stage power scheduling model coordinating voyage and power generation is established, and then a variable time scale control method is proposed to solve the scheduling scheme.

[0201] 2.1 Problem Statement

[0202] Two-stage range-power dispatch framework as follows Figure 3 As shown. The upper layer is an hourly-level voyage scheduling problem: predicting electrical load, thermal load, solar irradiance, wind speed and direction, and wave height, and using the proposed multivariate coupled all-electric propulsion model to calculate the ship's propulsion power, determining the ship's optimal speed and the start / stop status of the diesel generator set. The lower layer is a minute-level real-time power scheduling problem: based on short-term forecast information, the all-electric propulsion power is continuously corrected, and the optimal power scheduling scheme for diesel generator sets, cogeneration systems, battery banks, thermal storage equipment, and electric boilers is optimized at a minute-level variable time scale with the goals of ship economy and low-carbon navigation.

[0203] 2.1.1 Flight Schedule

[0204] 1) Objective function and constraints

[0205] In existing ship scheduling studies, the voyage scheduling duration is assumed to be the entire voyage cycle. In this embodiment, the voyage cycle is 52 hours, and the scheduling time scale is 1 hour. Upper-level voyage scheduling aims to minimize ship operating costs while ensuring on-time arrival. The operating costs of river-sea direct new energy vessels include fuel consumption and start-up / shutdown costs of the diesel generator system, fuel consumption costs of the combined heat and power unit, equipment wear and maintenance costs, gas emission treatment costs, and berthing and electricity purchase costs.

[0206]

[0207]

[0208]

[0209]

[0210]

[0211]

[0212]

[0213]

[0214]

[0215]

[0216]

[0217]

[0218]

[0219]

[0220]

[0221]

[0222]

[0223]

[0224] In the formula, Represents fuel cost; Represents the cost of environmental governance; This represents the start-up cost of a diesel generator set; This represents the cost of shutting down a diesel generator set; Represents the total operating and maintenance cost of all equipment on the ship; Represents the cost of purchasing electricity while the vehicle is parked; Represents the maintenance coefficient; ξ CI Represents electricity price; η represents the inbound / outbound port speed limit factor; η v This represents the maximum permissible speed adjustment factor; v n T represents the rated speed; p Represents the time period for sailing to and from the port; T c T represents the cruise time period; b Represents the duration of berthing; δ D This represents the maximum deviation in sailing distance; Represents the rated voyage distance to the nth port; This represents the sailing distance of a new energy vessel that directly connects the river and the sea during time period t. Dis represents the rated voyage distance to the destination port; nt This represents the final voyage distance of new energy vessels that can directly reach the river and sea.

[0225] Equation (51) represents fuel cost; Equation (52) represents equipment operation and maintenance cost; Equation (53) represents pollution gas treatment cost; Equations (54)-(55) represent diesel generator set start-up and shutdown cost; Equation (56) represents shore power cost; Equations (57)-(59) represent the speed limits that ships must meet when entering and leaving ports, cruising and berthing when navigating inland waterways and sea areas; Equations (60)-(61) represent navigation distance limits to ensure that ships arrive at port on time; Equation (62) indicates that diesel generator sets are prohibited from being used for power supply in inland waterways to meet the high standards for pollution discharge in inland waterways; Equations (63)-(64) represent the balance of electric and thermal power when ships are sailing; Equations (65)-(66) represent the balance of electric and thermal power when ships are berthing.

[0226] 2) Flight scheduling decision variables

[0227] The main objective of voyage scheduling is to determine the optimal speed of ships and the start / stop status of diesel generator sets for real-time scheduling, and to calculate the all-electric propulsion power. The generalized form of the model can be written as:

[0228]

[0229]

[0230]

[0231]

[0232] In the formula, Representing continuous and binary decision variables assigned to the real-time power scheduling phase, including v t , Dist t , Equations (70) and (71) are the equality and inequality constraints included in the first stage, respectively.

[0233] 2.1.2 Real-time power scheduling

[0234] 1) Objective function and constraints

[0235] Real-time power scheduling is optimized to minimize ship operating costs while satisfying equipment operating constraints. The solution time domain changes from the entire sailing cycle to a variable 1 / 2 / 3 hours, and the scheduling time scale is shortened from 1 hour to a variable 10 / 20 / 30 minutes. The specific decision-making process is detailed in Section 3.2. The optimization objective function and constraint set are as follows:

[0236]

[0237] Eqs.(51)-(53),(56),(62)-(67) (73)

[0238] 2) Power scheduling decision variables

[0239] During the real-time scheduling phase, the decision variable is the power output of each device, and the generalized form of the model can be expressed as:

[0240]

[0241]

[0242]

[0243]

[0244]

[0245] In the formula, The decision variables representing the flight schedule are treated as constants; Control variables representing the real-time scheduling phase include Equations (76) and (77) represent the equality and inequality constraints included in the second stage, respectively.

[0246] 2.1.3 Model Linearization

[0247] The optimization problem model of this invention contains high-order polynomial functions, such as equations (1), (27), and (50), and is a mixed-integer nonlinear programming problem. To reduce the computational burden and improve the solution quality of the variable time-scale scheduling strategy, a piecewise linearization method is used to approximate equations (1), (27), and (50) as linear models:

[0248]

[0249]

[0250]

[0251]

[0252]

[0253] The linearization method for unit start-up and shutdown costs (Equations (54)-(55)) is as shown in Equations (83)-(84):

[0254]

[0255]

[0256] The linearization methods for minimum start-up time and minimum downtime (Equations (31)-(32)) are similar, as shown in Equations (85)-(87):

[0257]

[0258]

[0259]

[0260] Using the linearization method described above, the entire optimization model is transformed into a mixed-integer linear programming problem, which can be solved using mature commercial solvers.

[0261] 2.2 Solution Method

[0262] SMPC is suitable for handling real-time scheduling problems with uncertainties. Therefore, this section proposes a novel variable-time-scale control strategy based on SMPC, which adjusts the control time domain range and time scale in real time according to the power fluctuations of the all-electric propulsion system, such as... Figure 4 As shown, the proposed variable time-scale control strategy can ensure the economical and low-carbon operation of new energy vessels that directly navigate between rivers and seas.

[0263] 2.2.1 Variable Time Scale Control Mechanism

[0264] This section proposes a novel variable timescale control strategy based on the SMPC method to dynamically coordinate the degree of power fluctuation with the scheduling timescale.

[0265] 1) Quantitative indicators of power fluctuation in all-electric propulsion

[0266] Uncertainty power fluctuations are primarily caused by all-electric propulsion systems. This section proposes quantitative indicators for uncertain power fluctuations in new energy vessels that can directly navigate between rivers and seas.

[0267] Utility theory is a theory used to analyze decision-makers' attitudes toward risk, and the value of the utility function is used to represent the degree of inclination

[33] . According to utility theory, the new energy ships that directly navigate between rivers and seas exhibit risk aversion towards fluctuations in all-electric propulsion power, and this is quantified using a risk aversion index utility function:

[0268]

[0269]

[0270]

[0271] In equations (88)-(90), f u This is the risk adaptation function, and its value is related to the fluctuation of the all-electric propulsion power; ΔP t Defined as power fluctuation rate, it reflects the power fluctuation of the all-electric propulsion system during the Δt period; This represents the maximum power fluctuation within the nΔt time period.

[0272] When ΔP t When f = 0, the power fluctuation of all-electric propulsion is minimal. u The value is 1. At this point, the utility is highest, and the value is 1. Substituting the above special value, the quantitative indicator can be rewritten as:

[0273]

[0274] Equation (91) provides a quantitative indicator for the fluctuation of all-electric propulsion power of new energy ships that can directly navigate between rivers and seas. The smaller the quantitative indicator, the greater the impact of the fluctuation of all-electric propulsion power on the system, and the more necessary it is to adjust the scheduling strategy in a timely manner to avoid risks.

[0275] 2) Time-scale decision-making mechanism

[0276] Control interval and control time domain are important parameters of SMPC, with the control interval directly determining the scheduling time scale (in this embodiment, time scale is used to represent the control interval): an excessively large time scale cannot handle short-term power fluctuations in all-electric propulsion in a timely manner, which is detrimental to the system's ability to respond to emergencies; an excessively short time scale leads to frequent communication and increased computational costs. This invention aims to find a dynamically reasonable control time scale to mitigate uncertain power fluctuations at the lowest cost.

[0277] Based on the quantitative index of power fluctuation in all-electric propulsion, the SMPC time-scale decision-making mechanism is defined as follows:

[0278]

[0279]

[0280] In equation (92), and These represent the minimum and maximum values ​​that can be adjusted on the time scale, respectively.

[0281] The control time domain typically consists of M control intervals:

[0282] T C =MN C (93)

[0283] Time-scale decision-making mechanisms satisfy the following properties:

[0284] a. The time scale increases monotonically with the quantitative index: When the power fluctuation of the all-electric propulsion decreases, the quantitative index gradually approaches 1, and the time scale increases accordingly to avoid redundant calculations and frequent scheduling.

[0285] b. The rate of change of the time scale decreases monotonically with the quantitative index: As the quantitative index approaches 0, it indicates that the fluctuation of the all-electric propulsion power gradually increases, the adjustment speed of the time scale accelerates, and the system responds to uncertain disturbances and emergencies in a timely manner.

[0286] Based on the above properties, (0, ) and (1, Substituting these values, the time-scale decision-making mechanism can be rewritten as follows:

[0287]

[0288] Equation (94) is a continuous function, and the optimal time scale can be calculated based on power fluctuations. Considering the limited computational resources and automation level of river-sea direct vessels, N c The calculation result is rounded to 10 / 20 / 30 minutes.

[0289] 2.2.2 Algorithm Flow

[0290] This embodiment presents the flowchart of a variable time-scale power scheduling algorithm based on SMPC. In the real-time power scheduling stage, a modified SMPC rolling optimization method is used to solve for the optimal scheduling scheme. In rolling optimization, the scheduling scheme has multiple time steps, but the integrated energy system executes only one control signal at a time. In the next time step, the control time domain is shifted backward, and the optimization scheduling is repeated. After each rolling optimization process, timely feedback correction must be made to the prediction model. This invention appropriately adds a normal distribution error to the prediction model to reduce the impact of system uncertainty and model bias.

[0291] The flow of the SMPC-based variable time-scale power scheduling algorithm is as follows: Figure 5 As shown:

[0292] Step 1: Set the initial control time domain T C and control interval N C

[0293] Step 2: Import flight schedule results

[0294] Step 3: Update the forecast sequence based on short-term forecast information

[0295] Step 4: Generate and reduce scenarios based on the probability distribution of the predicted information to handle uncertainty.

[0296] Step 5: Solve T using mixed-integer linear programming. C The optimal time-domain scheduling strategy is to issue only the first control command.

[0297] Step 6: Determine the scheduling parameters for the next time step based on the SMPC variable time scale control mechanism.

[0298] Step 7: Prediction Error Feedback Correction

[0299] Step 8: Scroll the time domain backward to return to Step 3.

[0300] Step 9: Obtain the optimal power scheduling solution

[0301] 3 Case Studies

[0302] This embodiment evaluates the operating cost, environmental benefits, energy utilization rate, power fluctuation, and computation time of the proposed method by designing multiple sets of comparative experiments.

[0303] 3.1 Parameter settings and simulation environment

[0304] The parameters of the river-sea direct new energy vessel selected in this embodiment are based on those of the Chinese vessel "Xiaoqinghe" and the "river-sea direct 120 TEU parallel hybrid container ship". The rated power of the two diesel generator sets is 400 kW, with an upper limit of 450 kW and a lower limit of 200 kW, and a gradeability limit of 200 kW. The initial capacity of the battery is 500 kWh, with a capacity range of 0.2-0.8 kWh, a maximum charge and discharge power of 500 kW, and a charge and discharge efficiency of 95%. The initial capacity of the thermal storage tank is 20 kWh, with a capacity range of 0.1-0.8 kWh, and a thermal storage efficiency of 95%. The fuel price is $0.83 / L, and the natural gas price is $0.36 / kWh. Other equipment parameters are shown in Table 2.

[0305] This embodiment uses a fixed shipping route within Shandong Province, China. The inland waterway is the Xiaoqing River, starting in Jinan City, passing through Zibo Port, and ending at the Yangkou estuary in Weifang, with a total length of approximately 169 km; the route then passes through Weifang into the sea, through Weihai Port, and finally reaches Qingdao, with a total length of approximately 700 km. Navigation and meteorological parameters are obtained from the China National Weather Information Center and the solarGIS database, including data acquired by MTSAT and HIMAWARI satellites between January 1, 2015 and December 31, 2015. Detailed data are shown in Table 3.

[0306] This embodiment uses a three-layer spatiotemporal graph attention network to obtain prediction results, assuming that the prediction error follows a normal distribution of N(0,5). Based on the probability distribution of the prediction results, Latin hypercube sampling and synchronous back-substitution reduction methods are used for scene generation and reduction to handle the uncertainty introduced by the prediction. The results are shown in […]. Figure 6 .

[0307] Table 1

[0308]

[0309]

[0310] Table 2

[0311]

[0312] Table 3

[0313]

[0314] The effectiveness of the proposed method was verified by designing two sets of comparative experiments. The first set discussed the accuracy and impact on operating costs of using different all-electric propulsion power models. The second set analyzed the economics, environmental benefits, and power fluctuation performance of different scheduling strategies. All experiments were implemented on a computer equipped with a Windows operating system, an AMD EPYC 770264-Core CPU@2.GHz, 512GB RAM, and an NVIDIA Quadro RTX 6000 GPU. The software platform for the prediction model was Python 3.7.0 and PyTorch 1.8, and the optimization model was solved using the Gurobi solver.

[0315] 3.2 Comparison of All-Electric Propulsion Power Models

[0316] Three case studies were set up to perform global optimization of flight scheduling results.

[0317] Case 1: Using the classic propulsion power calculation model;

[0318] Case 2: The proposed improved propulsion power calculation model is adopted, but the changes in the ship's cargo load are not considered;

[0319] Case 3: A new all-electric propulsion power calculation model considering changes in ship cargo load. The ship unloads 1 / 4 of its cargo at Weifang Port, completes unloading at Weihai Port, and sails empty to Qingdao for a new round of loading.

[0320] Figure 7 It can be seen that when using the classical model, the speed of each part is almost constant. The classical model neglects the coupled changes in ship resistance and motion state, resulting in a significant range error. During the first 20 hours of navigation on inland waterways, where resistance is relatively gentle, there is no significant speed difference between different models. However, when the ship is sailing at sea, the speed will fluctuate using the model proposed in this embodiment. Cases 2 and 3 show that the ship's inertia and uncertain resistance affect its speed, which is more consistent with actual navigation conditions. The results of Case 3 indicate that the ship's speed scheduling is closely related to its cargo load.

[0321] Figure 8The results of Case 1 and Case 3 show that different all-electric propulsion power models exhibit significant deviations during ship entry / exit and acceleration / deceleration. The total propulsion load demands in Case 1 and Case 2 are 28931 kWh and 26952 kWh, respectively, demonstrating that the classical model overestimates propulsion power. The results of Case 3 show that the ship's cargo load has a significant impact on propulsion power. The propulsion power demand in Case 3 is only 13706 kW, a 49% reduction compared to Case 2 and a 52.6% reduction compared to Case 1. Even at the same speed, ships with different cargo loads exhibit significant differences in propulsion load demand, a point completely overlooked in previous studies. Ignoring ship cargo load leads to substantial deviations in subsequent power scheduling.

[0322] Table 4 shows that propulsion power demand determines the operating cost of ship power scheduling. The scheduling cost in Case 2 is reduced by approximately 12% compared to Case 1, and the scheduling cost in Case 3 is reduced by 54.4% compared to Case 1. The results indicate that the classic all-electric propulsion power model has a large estimation error in ship operating costs. The all-electric propulsion power model proposed in this embodiment improves the accuracy of propulsion power and operating cost estimation.

[0323] Table 4

[0324]

[0325] 3.3 Analysis of Variable Time Scale Scheduling Strategies

[0326] This example discusses the performance of different scheduling strategies under four case studies. Case 4 is the classic speed-power global optimization method. Case 5 is a multi-timescale scheduling strategy based on the classic MPC method. Case 6 is the proposed variable-timescale scheduling strategy based on SMPC. Case 7 discusses the performance of this method under adverse navigation environments.

[0327] Cases 4-7 in this section all use the all-electric propulsion power model from Case 3. The harsh navigation environment in Case 7 is generated based on Monte Carlo simulation, with approximately 10% of sudden disturbances superimposed on the load side to further study the effectiveness and robustness of the variable time-scale scheduling strategy. In Case 7, the average propulsion power fluctuation is 300 kW, and the peak fluctuation is 700 kW. The time scale for range scheduling is 1 hour, and the total navigation time is 52 hours. The control time scale for Case 5 is 30 minutes, and the control time domain is 3 hours. In Cases 6 and 7, the variable control time scale is 10 / 20 / 30 minutes, and the control time domain is 1–3 hours.

[0328] Figure 9 and Figure 10 The results for Case 6 and Case 5 are presented respectively. Figure 9 It can be seen that the proposed scheduling strategy can achieve power control with variable time scale.

[0329] like Figure 9 (a) As shown in the rectangle, the electrical load power fluctuations are relatively large during hours 23-25, 27, and 33. The more significant the power fluctuations, the smaller the control timescale. The control timescale adapts to the degree of power fluctuation. Therefore, the total power generation of the DGs is 4526.5 kWh, less than in Case 5. The maximum output fluctuation of the DGs is 400 kW, a reduction of 50 kW compared to the multi-timescale dispatch strategy. The maximum output fluctuation of the CHP units is 112.5 kW, with a standard deviation reduction of 5.1%. The total battery discharge is 456.1 kWh, approximately 75 kWh higher than in Case 5. The total battery charge is approximately 452 kWh, approximately 100 kWh higher than in Case 5, resulting in an overall improvement of approximately 24% in the utilization rate of the energy storage system.

[0330] During the voyage from Weihai to Qingdao (37h–52h), the power demand for all-electric propulsion is reduced due to less unloading and disturbance. During the voyage from 49h to 51h, due to larger power fluctuations, the control timescale is adjusted to 10 minutes. Simultaneously, power fluctuations are mainly mitigated by the batteries and CHP. Using the method proposed in this invention, the maximum output fluctuation of CHP is reduced from 152kW to approximately 17kW, resulting in a 9.8% reduction in the ship's overall power fluctuation. These results demonstrate that the variable timescale control strategy can mitigate power fluctuations while minimizing the output of high-power equipment such as DGs and CHP, thus improving the reliability of the ship's integrated energy system.

[0331] From an economic and emissions perspective, compared to multi-timescale scheduling strategies, this strategy reduces operating costs by approximately 5.1%, and compared to classic scheduling strategies, by approximately 16.5%. Total carbon emissions are reduced by approximately 11.4% compared to Case 5, and by approximately 32% compared to Case 4. The variable timescale scheduling strategy reduces the operating time of high-power equipment, resulting in better economic and environmental benefits. The average scheduling calculation time is 2.8 seconds, far less than the minimum scheduling time for ships, meeting real-time scheduling requirements.

[0332] Table 5

[0333]

[0334] Variable timescale scheduling results under severe sea conditions are as follows Figure 11As shown, to cope with sudden strong power disturbances, the scheduling time scale was mostly adaptively adjusted to 10 minutes. Table 6 shows the output fluctuation of high-power equipment (DGs, CHP) under different scheduling strategies. Comparison with the results of fixed-time-scale scheduling reveals that: the maximum total output fluctuation of the diesel generator system was 288 kW, a decrease of 87 kW, and the standard deviation was 140.5, a decrease of 10.4%; the maximum output fluctuation of the CHP unit was 265 kW, the standard deviation was 59.1, a decrease of 14.2%, and the power fluctuation rate decreased by approximately 8.4%; the overall battery utilization rate increased by approximately 18%.

[0335] Figure 11 The variable time-scale scheduling scheme under harsh navigation conditions is shown. To cope with sudden fluctuations, most control time scales are adaptively adjusted to 10 minutes. As shown in Table 6, the maximum output fluctuation of DGs is 288 kW. Compared with the fixed time-scale method, the maximum output fluctuations of DGs and CHP are reduced by 23.2% and 6.4%, respectively. The output standard deviations of DGs and CHP are reduced by 10.4% and 14.2%, respectively, and the overall battery utilization rate is improved by about 18%.

[0336] Table 6

[0337]

[0338] The results show that the proposed scheduling strategy can effectively mitigate power fluctuations under any circumstances. Compared with other classical scheduling strategies, the variable time-scale scheduling strategy has better robustness and economy.

[0339] 4. Conclusion

[0340] The navigation environment of new energy vessels operating directly on rivers and seas is complex and subject to severe power fluctuations. Traditional propulsion power calculation methods result in significant errors in propulsion load calculation, making it difficult to guarantee scheduling accuracy. Fixed-time-scale power scheduling strategies cannot handle fluctuations in all-electric propulsion power in a timely manner, leading to a mismatch between scheduling commands and time-varying uncertainties. To address these issues, this invention proposes an all-electric propulsion power calculation model that considers the coupled effects of multiple variables such as resistance, motion state, speed, and cargo load. Based on an improved SMPC method, a novel variable-time-scale control strategy considering the quantification of uncertain power is proposed. Experimental results show that the vessel's cargo load has a significant impact on all-electric propulsion power. The proposed all-electric propulsion power model reduces propulsion load demand by 52.6% and improves scheduling economy by 67%. This control strategy reduces operating costs by 16.5% and power fluctuations by approximately 13.3%. In future work, we will study optimization solution methods specifically for variable-time-scale scheduling, introducing logistics scheduling problems into vessel voyage scheduling.

[0341] The foregoing has shown and described the basic principles, main features, and advantages of the present invention. Those skilled in the art should understand that the present invention is not limited to the above embodiments. The embodiments and descriptions in the specification are merely illustrative of the principles of the invention. Various changes and modifications can be made to the invention without departing from its spirit and scope, and all such changes and modifications fall within the scope of the present invention as claimed. The scope of protection of the present invention is defined by the appended claims and their equivalents.

Claims

1. A method for variable time-scale power scheduling of new energy vessels that can directly reach river and sea, characterized in that, Includes the following steps: A calculation model for all-electric propulsion is constructed based on ship resistance, motion state, speed, and cargo capacity; Based on the all-electric propulsion calculation model, a two-stage coordinated power scheduling model for range and power generation is constructed, wherein the two-stage coordinated power scheduling model for range and power generation includes a range scheduling stage and a power scheduling stage. The optimal voyage plan for the ship is obtained based on the aforementioned voyage scheduling phase; During the power scheduling phase, a variable time-scale control strategy is designed based on a stochastic model predictive control method using an uncertainty power fluctuation quantification index. The mathematical model of this uncertainty power fluctuation quantification index, constructed based on utility theory, is as follows: ； In the formula, This is the risk adaptation function, and its value is related to the uncertainty power fluctuation quantification index. , For power fluctuation rate, for Maximum power fluctuation within the time period A quantitative indicator for uncertain power fluctuations; Based on the aforementioned uncertainty power fluctuation quantification index, the formula for determining the time-scale decision-making mechanism using the stochastic model predictive control method is as follows: ； In the formula, Indicates the control interval. and These represent the minimum and maximum values ​​that can be adjusted on the time scale, respectively. Based on the ship's optimal voyage plan and the variable time-scale control strategy, the optimal scheme for variable time-scale power scheduling is obtained; The calculation model for all-electric propulsion is as follows: ； In the formula, express All-electric propulsion power during the period, Indicates the speed during time period t. express Total weight of cargo on the vessel during the specified time period. express Speed ​​during the time period.

2. The variable time-scale power scheduling method for river-sea direct new energy vessels according to claim 1, characterized in that, The voyage scheduling phase determines the optimal voyage plan for a vessel with the goal of minimizing vessel operating costs, while ensuring that the vessel arrives at the port on time. The vessel operating costs include: fuel costs, equipment maintenance costs, pollution gas treatment costs, diesel generator start-up and shutdown costs, and shore power costs.

3. The variable time-scale power scheduling method for river-sea direct new energy vessels according to claim 1, characterized in that, The objective function of the mathematical model for the power scheduling phase is: ； In the formula, Represents fuel cost; Represents the cost of environmental governance; Represents the total operating and maintenance cost of all equipment on the ship; Represents the cost of purchasing electricity while the vehicle is parked; The t represents the berthing time period, M represents the number of control intervals in the control time domain, and t represents the berthing time period. n This refers to the current moment.

4. The variable time-scale power scheduling method for river-sea direct new energy vessels according to claim 1, characterized in that, The stochastic model predictive control method determines a variable time-scale power scheduling strategy by adjusting two parameters: the control interval and the control time domain. The control time domain includes several control intervals.

5. The variable time-scale power scheduling method for river-sea direct new energy vessels according to claim 4, characterized in that, The variable time-scale power scheduling process based on the stochastic model predictive control method includes: S1. Set the initial control time domain and control interval; S2. Import the optimal voyage plan for the vessel; S3. Based on the ship's optimal voyage plan, perform uncertainty processing in the control time domain and solve the power scheduling strategy in the current control time domain; S4. Based on the power scheduling strategy under the current control time domain, roll to the next time domain control, update the control time domain and control interval based on the variable time scale control strategy, and perform uncertainty handling for the next time domain control. S5. Repeat S3-S4 to obtain the optimal power scheduling scheme with variable time scale.

Images