Multi-energy electric propulsion ship energy management method and system based on load prediction

By using load prediction models and model prediction control in multi-energy power propulsion ship systems, the power source output is adjusted in real time, and the problem of large amount of calculations and the need to know the working condition data in the existing technology is solved, and efficient energy management and environmental protection are achieved Performance improvement.

CN119010226APending Publication Date: 2024-11-22JIANGSU UNIV OF SCI & TECH +1
View PDF 0 Cites 3 Cited by

Patent Information

Application Number
CN202411045245.6
Authority / Receiving Office
CN · China
Patent Type
Applications(China)
Current Assignee / Owner
Filing Date
2024-08-01
Publication Date
2024-11-22

AI Technical Summary

Technical Problem

The existing multi-energy power propulsion ship energy management system has a large amount of calculation and requires the entire ship's operating condition data in advance, making it difficult to achieve real-time optimization and efficient management.

Method used

Using multi-energy power propulsion ship energy management method based on load prediction, by obtaining the power source output power and load data at the current moment, combining variational modal decomposition and spider wasp algorithm with long and short-term memory neural network to establish a load prediction model, and combining the model Predictive control and weight adjustment MPC controller, adjusting control variables in real time to optimize energy distribution.

Benefits of technology

It improves load prediction accuracy and real-time optimization capabilities of energy management, reduces ship energy consumption, improves system efficiency and environmental performance, and is suitable for actual optimization control of hybrid systems.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN119010226A_ABST
    Figure CN119010226A_ABST
Patent Text Reader

Abstract

The invention discloses a multi-energy electric propulsion ship energy management method and system based on load prediction. The method comprises the steps that 1, the output power of a diesel engine, the active power of a super capacitor, the active power and net power of a storage battery pack, the SOC of the storage battery pack and the SOC of the super capacitor at the current moment are obtained; weight factors of the active power of the storage battery pack, the active power of the super capacitor and the output power of the diesel engine are determined; 2, inputting the information obtained in the step 1 into a prediction controller to obtain the optimized output power of the diesel engine, the active power of the super capacitor and the active power of the storage battery pack at the current moment; 3, the optimized output power of the diesel engine, the active power of the super capacitor and the active power of the storage battery pack at the current moment obtained in the step 2 act on the multi-energy electric propulsion ship; and 4, repeatedly executing the steps 1 to 3.
Need to check novelty before this filing date? Find Prior Art

Description

Technical Field

[0001] The present invention belongs to the technical field of ship energy management, and in particular relates to a method and system for energy management of a multi-energy electric propulsion ship based on load forecasting. Background Art

[0002] Energy management for ship power systems containing multiple energy sources requires solving a large-scale, nonlinear, and constrained optimization problem. It is necessary to comprehensively consider factors such as the charge state of the energy storage system, the optimal power of each power generation unit, and the ship's operating conditions in order to allocate energy. Different energy management strategies have different effects on fuel economy, power source dynamic performance, and service life.

[0003] The multi-energy electric propulsion system includes different types of power sources with different working modes, and these power sources themselves have different dynamic characteristics. Therefore, in order to reasonably manage the multi-power source system, it is necessary to formulate a suitable energy management strategy to distribute the output power or torque of each power source and coordinate the electric propulsion, so as to improve the system efficiency and improve the system energy saving and emission reduction performance on the basis of ensuring the power and safety of the ship.

[0004] Ship energy management strategies are divided into rule-based and optimization-based strategies. The rule-based strategy includes deterministic rules and fuzzy rules, and the optimization-based strategy can be subdivided into global optimization and real-time optimization. Although the rule-based energy management strategy is simple and practical, it requires engineers' rich experience to formulate rules, and it cannot achieve the optimal power distribution of the hybrid power system; the energy management strategy based on global optimization needs to obtain the working condition information of the entire voyage in advance, so the solution is complex and difficult to be applied in the actual optimization control of the hybrid power system; the energy management strategy based on model predictive control can transform global optimization into local optimal control based on real-time identification of ship dynamic parameters, and realize real-time optimization of ship energy efficiency through prediction models and optimization algorithms.

[0005] The existing energy management strategy based on model predictive control is still limited by the large amount of calculation of the dynamic programming algorithm and the need to know the operating condition data of the entire ship in advance. This energy management strategy still has a certain amount of work to be supplemented before it can be applied in practice, such as: improving the operating speed of the controller; in view of the particularity of the fixed sailing section of the ship, it is necessary to use big data to analyze the operating conditions in different seasons, different times and different meteorological conditions, and design corresponding energy management strategies to further improve the practicality of the algorithm and reduce the energy consumption of the system. Summary of the invention

[0006] Purpose of the invention: In order to solve the problems of large amount of calculation and the need to know the operating condition data of the whole ship in advance when performing energy management on the ship power system containing multiple energy sources, the present invention proposes a method and system for energy management of multi-energy electric propulsion ships based on load forecasting.

[0007] Technical solution: A multi-energy electric propulsion ship energy management method based on load forecasting, comprising the following steps:

[0008] Step 1: Get the output power P of the diesel engine at the current moment DG , the active power P of the supercapacitor C , the active power P of the battery pack bat , Net power P net , battery pack SOC and supercapacitor SOC; the net power P net Expressed as: P net =P DG +P pv -P load , P pv is the photovoltaic output power, P load Load prediction value of load unit for multi-energy electric propulsion ship;

[0009] Determine the active power P of the battery bank bat , the active power P of the supercapacitor C , diesel engine output power P DG The weight factors are denoted as λ1, λ2, and λ3;

[0010] Step 2: Input the information obtained in step 1 into the prediction controller to obtain the optimized output power P of the diesel engine at the current moment. DG , the active power P of the supercapacitor C And the active power P of the battery pack bat ;

[0011] Step 3: The optimized output power P of the diesel engine at the current moment obtained in step 2 is DG , the active power P of the supercapacitor C And the active power P of the battery pack bat Applicable to multi-energy electric propulsion ships;

[0012] Step 4: Repeat steps 1 to 3.

[0013] The present invention also discloses a multi-energy electric propulsion ship energy management system based on load forecasting, comprising:

[0014] Data acquisition module, used to obtain the output power P of the diesel engine at the current moment DG , the active power P of the supercapacitorC , the active power P of the battery pack bat , Net power P net , battery pack SOC and supercapacitor SOC; the net power P net Expressed as: P net =P DG +P pv -P load , P pv is the photovoltaic output power, P load Load prediction value of load unit for multi-energy electric propulsion ship;

[0015] The weight factor determination module is used to determine the active power P of the battery pack based on the information obtained by the data acquisition module. bat , the active power P of the supercapacitor C , diesel engine output power P DG The weight factors are denoted as λ1, λ2, and λ3;

[0016] The prediction controller is used to obtain the optimized output power P of the diesel engine at the current moment based on the information obtained by the data acquisition module and the weight factor determined by the weight factor determination module. DG , the active power P of the supercapacitor C And the active power P of the battery pack bat ; The output power P of the diesel engine at the current moment after optimization DG , the active power P of the supercapacitor C And the active power P of the battery pack bat Acts on multi-energy electric propulsion ships; and calls the data acquisition module.

[0017] Beneficial effects: Compared with the prior art, the present invention has the following advantages:

[0018] (1) The method of the present invention reasonably distributes the output power of the supercapacitor, the battery and the diesel generator based on the minimum fuel consumption rate working point of the diesel engine, adopts the MPC controller control mode with weight adjustment, improves the traditional MPC control method, and adjusts the control variables in real time according to the load power conversion, which can effectively reduce the energy consumption of the ship;

[0019] (2) In terms of load forecasting, the method of the present invention solves the problem of uncertainty and nonlinear relationship of multi-energy ship load forecasting data by coupling variational mode decomposition and Spider wasp optimizer (SWO) with long short-term memory neural network (LSTM) to establish a load forecasting model (VMD-SWO-LSTM), which enhances the ability to capture the complex time and frequency spectrum characteristics of power load data, thereby improving the prediction accuracy;

[0020] (3) The method of the present invention adopts a weight-adjusted MPC controller control method to improve the traditional MPC control method and adjust the control variables in real time according to the load power conversion; Model Predictive Control (MPC) can integrate various constraints, such as energy constraints, system state constraints, and operating variable constraints, because it can perform optimization calculations at each time step and can be adjusted according to the current system state and external conditions. MPC allows multiple performance indicators and objective functions to be considered simultaneously, rather than just a single energy efficiency or cost. This allows MPC to improve the efficiency, reliability, and environmental performance of the system in multi-energy ship energy management, thereby providing obvious advantages for ship operation;

[0021] (4) The method of the present invention divides a plurality of operation modes according to the state information of the energy storage unit, formulates a corresponding control method according to the load mutation situation, sets the change rule of the weight factor, and adjusts the weight factor in the MPC controller established by the Matlab / Simulink custom s-function. BRIEF DESCRIPTION OF THE DRAWINGS

[0022] Figure 1 Flow chart for load forecasting;

[0023] Figure 2 A block diagram of a method for energy management of a multi-energy electric propulsion ship based on load forecasting;

[0024] Figure 3 Schematic diagram of the MPC controller weight value operating conditions. DETAILED DESCRIPTION

[0025] The technical solution of the present invention is now further described in conjunction with the accompanying drawings.

[0026] Embodiment 1:

[0027] like Figure 2 As shown, this embodiment discloses a multi-energy ship real-time energy management method based on load forecasting, comprising the following steps

[0028] Step 1: Get the output power P of the diesel engine at the current moment DG , the active power P of the supercapacitor C , the active power P of the battery pack bat , Net power P net , battery pack SOC and supercapacitor SOC; net power P net Expressed as: P net =P DG +P pv -P load , Ppv is the photovoltaic output power, P load Load prediction value of load unit for multi-energy electric propulsion ship;

[0029] Among them, Figure 1 As shown in the figure, the load prediction value P of the load unit of the multi-energy electric propulsion ship load Follow the steps below to get:

[0030] Data preparation and preprocessing: Import historical load data and save it. Set sampling frequency and time series parameters.

[0031] The historical load data is decomposed using variational mode decomposition (VMD) to obtain multiple independent modal load components.

[0032] The Spider wasp optimizer (SWO) was used to optimize the parameters of the Long Short Term Memory (LSTM) neural network, including the learning rate, hidden layer size, and time step.

[0033] The load components of each independent mode decomposed by variational mode decomposition (VMD) are input into an independent long short-term memory neural network (Long Short Term Memory, LSTM) for training and prediction;

[0034] The predicted values ​​of the load components of each independent mode are added together to obtain the load prediction value P of the load unit of the multi-energy electric propulsion ship. load .

[0035] Step 2: Determine the active power P of the battery bank bat , the active power P of the supercapacitor C , diesel engine output power P DG The weight factors are denoted as λ1, λ2, and λ3; Figure 3 As shown, the specific operations include:

[0036] The coordination of battery packs, supercapacitors and diesel engines can effectively improve the economy and environmental protection of multi-energy electric propulsion ship operation. Considering the impact of deep charge and discharge of battery packs and supercapacitors on their cycle times and life, the battery pack SOC and supercapacitor SOC state ranges are divided into four areas.

[0037] When P net >0, the working conditions of the battery pack and supercapacitor under different load conditions and the weight matrix Rm The values ​​are shown in Table 1.

[0038] Table 1 P net >0 working condition division

[0039]

[0040] When P net <0, the working conditions of the battery pack and supercapacitor under different load conditions and the weight matrix R m As shown in Table 2.

[0041] Table 2 P net <0 working condition division

[0042]

[0043]

[0044] Step 3: Input the information obtained in step 1 into the prediction controller to obtain the optimized output power P of the diesel engine at the current moment. DG , the active power P of the supercapacitor C And the active power P of the battery pack bat Specific operations include:

[0045] Customize the model predictive controller based on the s-function. Use the s-function customization function to develop a model predictive controller in Matlab / Simulink according to the following customization content:

[0046] The electric power system of a multi-energy electric propulsion ship includes a power generation unit, an energy storage unit, a power unit and a load unit. This embodiment establishes a simulation model of a multi-energy electric propulsion ship based on the idea of ​​modular modeling. The specific operations include:

[0047] Among them, the power generation unit includes a marine diesel generator and a marine photovoltaic system.

[0048] The mathematical model of the marine diesel generator is expressed as:

[0049] F uel =b e ·P DG

[0050] Where P DG is the output power of the diesel generator; b e is the fuel consumption rate of the diesel generator at different load rates, g / (kW·h); F uel It is the fuel consumed when the diesel generator is running, L.

[0051] The model of the marine photovoltaic system is expressed as:

[0052]

[0053] Where P pv Represents the output power of the photovoltaic unit; I sc is the short-circuit current under illumination; V is the voltage of the photovoltaic unit; V oc is the open circuit voltage; I0 is the reverse saturation current; R s is the series resistance of the photovoltaic cell; R sh is the parallel resistance of the photovoltaic unit; q is the elementary charge; k is the Boltzmann constant; T is the temperature of the photovoltaic unit; β is the reverse saturation current, I0 is the temperature coefficient that varies with temperature; T ref It is the reference temperature, usually 25°C under standard test conditions.

[0054] The energy storage unit consists of a battery pack and a supercapacitor. The electromotive force of the energy storage unit can be expressed as:

[0055]

[0056] Where, E0 is the initial internal electromotive force; K is the polarization voltage constant, which describes the effect of battery polarization on electromotive force; SOC is the state of charge of the battery; i is the charge and discharge current; Q n is the rated capacity of the battery, indicating the rated charge capacity of the battery; A u is the voltage variation coefficient, which describes the rate of change of the battery electromotive force as the state of charge changes; B c is the capacity variation coefficient, which describes the rate of change of battery capacity over time; t is time, which is used to calculate the integral of charge and discharge current; C t is the polarization effect coefficient, which describes the influence of battery polarization effect on electromotive force; T b is the battery temperature, indicating the current temperature of the battery; T ref It is the reference temperature used to compare the difference between the current temperature of the battery and the reference temperature, usually 25°C under standard test conditions.

[0057]

[0058] In the formula, E bat (t+1) and E bat (t) are the current capacities of the battery pack at time t+1 and time t respectively; μ bat and η bat are the charging and discharging efficiency of the battery pack respectively; P bat (t) is the output power of the battery pack at time t, where discharge is a positive value and charge is a negative value; Δt is the time interval.

[0059]

[0060] In the formula, E C (t+1) and E C (t) are the current capacities of the supercapacitor at time t+1 and time t respectively; μ C and η C are the charging and discharging efficiency of the supercapacitor respectively; P C (t) is the output power of the supercapacitor at time t, where discharging is a positive value and charging is a negative value.

[0061] Formulate a multi-energy ship energy management strategy. The purpose of this multi-energy ship energy management control strategy is to reasonably allocate the output power of supercapacitors, batteries and diesel generators according to the current operating conditions of the ship, while ensuring safe, stable and reliable power supply for the entire ship, so as to minimize the fuel consumption of the entire ship. That is, the diesel generator power, battery power and supercapacitor power are used as control variables, and the control variables are adjusted in real time according to the load power conversion.

[0062] The power balance equation of the multi-energy ship energy management system can be expressed as:

[0063] P DG +P pv +P bat +P C =P load

[0064] Where P DG With P pv are the active power of diesel engine and photovoltaic respectively; P C and P bat are the active powers of the supercapacitor and battery respectively. The reference power direction selection rule is: P C and P bat >0 indicates the output power, P C and P bat <0 means absorbed power.

[0065] Define net power P net and the power P generated by the controllable energy gen :

[0066] P net =P DG +P pv -P load

[0067] P gen =P DG +P C +P bat -P pv

[0068] In the formula, the net power P netis the difference between energy and load demand; P net There may be three situations: positive, negative, and zero. net = 0, the electricity generated by the diesel engine and photovoltaics just meets the needs of the load; when the power generation capacity is in excess (P net >0), the energy storage unit absorbs the remaining power. net <0), the energy storage unit supplements the remaining power.

[0069] The optimization goal of the multi-energy ship system is to make the diesel generator operate in the optimal range, improve the efficiency of energy conversion, reduce fuel consumption, and maximize the economy and environmental protection of ship operation.

[0070] Select battery SOC, supercapacitor SOC and P gen As the state variable, select the battery power P bat 、Supercapacitor power P C and diesel engine power P DG As control variables, battery SOC, supercapacitor SOC and P are selected DG Output y as a constraint b , select P gen As the controlled output y c , establish the following linear discrete system state space:

[0071]

[0072] x=[SOC bat ,SOC C ,P gen ]

[0073] u=[P bat ,P C ,P DG ]

[0074] y b =[SOC bat ,SOC C ,P DG ]

[0075] y c =[P gen ]

[0076] In the formula, k is the current moment; k+1 is the next moment; x is the state variable; u is the control variable; y c is the controlled output variable; y b is the constrained output variable; A and B are the system matrix and control matrix respectively; C c is the controlled output matrix; C b is the constraint output matrix.

[0077] By finding the best control method, the predicted controlled output y c and the expected output y ref The closer the better, and the fuel consumption of the diesel generator is B e Minimum, for which the predicted controlled output y c and the expected output y ref The cumulative error between the fuel consumption of the diesel generator and the e The sum is taken as the power objective function J of the multi-energy ship system, which is expressed as:

[0078] In the formula

[0079]

[0080] In the formula, k+j|k represents the prediction of time k for time k+j; y ref is the expected output given; B e is the fuel consumption of the diesel generator; P DG Indicates the active power output of the diesel engine, b e Indicates fuel consumption rate, N p is the prediction time domain; N c To control the time domain; the matrix Q m is the weighting matrix for the output error; the matrix R m is the weighting matrix of the control increment; β is the weight factor of the state variable; λ1, λ2, λ3 represent the weight factors of the control increment, λ1, λ2, λ3 represent the weight factors of the battery power P respectively bat 、Supercapacitor power P C , Diesel generator power P DG Δu(k) represents the change of the control variable. The smaller the expected value, the better. The active power P of the battery group is selected. bat , the active power P of the supercapacitor C and the diesel engine output power P DG as a control variable.

[0081] Select load power P load With P pv The difference between ref (k+j|k), the control goal is to make the diesel generator work in the optimal range, and the total output power can track the given P load , and the fuel consumption of the diesel generator is B e Minimum.

[0082] The controllable energy power constraints are:

[0083]

[0084] The battery SOC and supercapacitor SOC constraints are:

[0085]

[0086] The control increment constraint is:

[0087]

[0088] In the formula, the subscript "max" and the subscript "min" represent the upper and lower limits of the corresponding variables, respectively.

[0089] According to the characteristics of diesel generators, batteries and supercapacitors and various constraints, the power objective function J of the multi-energy ship system is transformed into a quadratic programming problem for solution, and the state parameters are updated to correct the predicted value at the previous moment. Finally, the steps of prediction, optimization and correction are repeated. The solution steps are shown below.

[0090] (1) Set the MPC algorithm sampling time T s , prediction time domain N p , control time domain N c , the weighted matrix Q of the output error m , the weighted matrix R of the control increment m .

[0091] (2) Calculate the coefficient matrix used in the MPC algorithm.

[0092] (3) The functions in the Matlab toolbox are used to solve the optimal solution of the MPC problem and obtain Δu(k) that minimizes the power objective function J of the multi-energy ship system.

[0093] Step 4: The optimized output power P of the diesel engine at the current moment obtained in step 3 is DG , the active power P of the supercapacitor C And the active power P of the battery pack bat Applicable to multi-energy electric propulsion ships;

[0094] Step 5: Repeat steps 1 to 4.

Claims

1. A multi-energy electric propulsion ship energy management method based on load forecasting, characterized in that: The following steps are involved: Step 1: Get the output power P of the diesel engine at the current moment DG , the active power P of the supercapacitor C , the active power P of the battery pack bat , Net power P net , battery pack SOC and supercapacitor SOC; the net power P net Expressed as: P net =P DG +P pv -P load , P pv is the photovoltaic output power, P load Load prediction value of load unit for multi-energy electric propulsion ship; Determine the active power P of the battery bank bat , the active power P of the supercapacitor C , diesel engine output power P DG The weight factors are denoted as λ1, λ2, and λ3; Step 2: Input the information obtained in step 1 into the prediction controller to obtain the optimized output power P of the diesel engine at the current moment. DG , the active power P of the supercapacitor C And the active power P of the battery pack bat ; Step 3: The optimized output power P of the diesel engine at the current moment obtained in step 2 is DG , the active power P of the supercapacitor C And the active power P of the battery pack bat Applicable to multi-energy electric propulsion ships; Step 4: Repeat steps 1 to 3.

2. The method for energy management of a multi-energy electric propulsion ship based on load forecasting according to claim 1 is characterized in that: The load prediction value P of the load unit of the multi-energy electric propulsion ship is obtained by following the steps below: load : The historical load data of the load unit of the multi-energy electric propulsion ship are decomposed by using variational mode decomposition to obtain multiple independent mode load components; According to the Spider-Wasp algorithm, the load components of each independent mode are input into an independent long-short-term memory neural network for training and prediction; The predicted values ​​of the load components of each independent mode are added together to obtain the load prediction value P of the load unit of the multi-energy electric propulsion ship. load .

3. The method for energy management of a multi-energy electric propulsion ship based on load forecasting according to claim 1 is characterized in that: The predictive controller is developed in Matlab / Simulink according to the following customized content using the s-function customization function: Establish the objective function of multi-energy electric propulsion ships; Transform the objective function of multi-energy electric propulsion ships into a quadratic programming problem; Use the information obtained in step 1 to solve the quadratic programming problem and obtain the optimized output power P of the diesel engine at the current moment. DG , the active power P of the supercapacitor C And the active power P of the battery pack bat .

4. The method for energy management of a multi-energy electric propulsion ship based on load forecasting according to claim 3 is characterized in that: Follow the steps below to establish the objective function of a multi-energy electric propulsion ship: Establishing a simulation model of a multi-energy electric propulsion ship, wherein the simulation model of the multi-energy electric propulsion ship includes: a diesel engine, a supercapacitor, a battery pack, a photovoltaic and a load unit; Define the power P generated by controllable energy gen : P gen =P DG +P C +P bat -P pv Where P DG With P pv are the output power of diesel engine and photovoltaic, P C and P bat are the active power of the supercapacitor and the battery pack respectively; P C and P bat >0 indicates the output power, P C and P bat <0 means absorbed power; Select the battery pack SOC, supercapacitor SOC and the power P generated by the controllable energy gen As the state variable, select the active power P of the battery group bat , the active power P of the supercapacitor C and the diesel engine output power P DG As control variables, the battery pack SOC, supercapacitor SOC and diesel engine output power P are selected. DG Output y as a constraint b , select the power P generated by the controllable energy gen As the controlled output y c , establish the following linear discrete system state space: The system state variables, control variables, and output variable matrices are expressed as: x=[SOC bat ,SOC C ,P gen ] u=[P bat, P C, P DG ] y b =[SOC bat ,SOC C ,P DG ] and c =[P gen ] In the formula, k represents the current moment, k+1 represents the next moment, x represents the state variable, u represents the control variable, and y represents the c Represents the controlled output variable, y b represents the constrained output variable; A and B represent the system matrix and control matrix respectively, C c represents the controlled output matrix, C b represents the constraint output matrix; By finding the best control method, the controlled output variable y c and the expected output y ref The closer the better, and the fuel consumption of the diesel generator is B e minimum, for which the controlled output variable y c and the expected output y ref The cumulative error between the fuel consumption of the diesel generator and the e The sum is used as the objective function J of the multi-energy electric propulsion ship, expressed as: Where: In the formula, k+j|k represents the prediction of time k for time k+j, b e Indicates fuel consumption rate, N p represents the prediction time domain, N c represents the control time domain, Q m Represents the weighting matrix of the output error, R m represents the weighted matrix of the control increment, β is the weight factor of the state variable, λ1, λ2, λ3 represent P bat , P C , P DG The weight factor of ; Δu(k) represents the change of the control variable; Select the load prediction value P of the load unit of the multi-energy electric propulsion ship load The output power P of the photovoltaic pv The difference between ref (k+j|k); The constraints of the objective function J include: In the formula, the subscript max and subscript min represent the upper and lower limits of the corresponding variables, respectively.

5. The method for energy management of a multi-energy electric propulsion ship based on load forecasting according to claim 4 is characterized in that: The P bat , P C , P DG The weight factors λ1, λ2, and λ3 are determined according to the following steps: When P net >0, the working conditions of the battery pack and supercapacitor under different load conditions and the weight matrix R m The value includes: When P net <0, the working conditions of the battery pack and supercapacitor under different load conditions and the weight matrix R m Possible values ​​are:

6. A multi-energy electric propulsion ship energy management system based on load forecasting, characterized in that: include: Data acquisition module, used to obtain the output power P of the diesel engine at the current moment DG , the active power P of the supercapacitor C , the active power P of the battery pack bat , Net power P net , battery pack SOC and supercapacitor SOC; the net power P net Expressed as: P net =P DG +P pv -P load , P pv is the photovoltaic output power, P load Load prediction value of load unit for multi-energy electric propulsion ship; The weight factor determination module is used to determine the active power P of the battery pack based on the information obtained by the data acquisition module. bat , the active power P of the supercapacitor C , diesel engine output power P DG The weight factors are denoted as λ1, λ2, and λ3; The prediction controller is used to obtain the optimized output power P of the diesel engine at the current moment based on the information obtained by the data acquisition module and the weight factor determined by the weight factor determination module. DG , the active power P of the supercapacitor C And the active power P of the battery pack bat ; The output power P of the diesel engine at the current moment after optimization DG , the active power P of the supercapacitor C And the active power P of the battery pack bat Acts on multi-energy electric propulsion ships; and calls the data acquisition module.

7. A multi-energy electric propulsion ship energy management system based on load forecasting according to claim 6, characterized in that: The load prediction value P of the load unit of the multi-energy electric propulsion ship is obtained by following the steps below: load : The historical load data of the load unit of the multi-energy electric propulsion ship are decomposed by using variational mode decomposition to obtain multiple independent mode load components; According to the Spider-Wasp algorithm, the load components of each independent mode are input into an independent long-short-term memory neural network for training and prediction; The predicted values ​​of the load components of each independent mode are added together to obtain the load prediction value P of the load unit of the multi-energy electric propulsion ship. load .

8. The multi-energy electric propulsion ship energy management system based on load forecasting according to claim 6 is characterized by: The predictive controller is developed in Matlab / Simulink according to the following customized content using the s-function customization function: Establish the objective function of multi-energy electric propulsion ships; Transform the objective function of multi-energy electric propulsion ships into a quadratic programming problem; Use the information obtained in step 1 to solve the quadratic programming problem and obtain the optimized output power P of the diesel engine at the current moment. DG , the active power P of the supercapacitor C And the active power P of the battery pack bat .

9. A multi-energy electric propulsion ship energy management system based on load forecasting according to claim 8, characterized in that: Follow the steps below to establish the objective function of a multi-energy electric propulsion ship: Establishing a simulation model of a multi-energy electric propulsion ship, wherein the simulation model of the multi-energy electric propulsion ship includes: a diesel engine, a supercapacitor, a battery pack, a photovoltaic and a load unit; Define the power P generated by controllable energy gen : P gen =P DG +P C +P bat -P pv Where P DG With P pv are the output power of diesel engine and photovoltaic, P C and P bat are the active power of the supercapacitor and the battery pack respectively; P C and P bat >0 indicates the output power, P C and P bat <0 means absorbed power; Select the battery pack SOC, supercapacitor SOC and the power P generated by the controllable energy gen As the state variable, select the active power P of the battery group bat , the active power P of the supercapacitor C and the diesel engine output power P DG As control variables, the battery pack SOC, supercapacitor SOC and diesel engine output power P are selected. DG Output y as a constraint b , select the power P generated by the controllable energy gen As the controlled output y c , establish the following linear discrete system state space: x=[SOC bat ,SOC C ,P gen ] u=[P bat, P C ,P DG ] y b =[SOC bat ,SOC C ,P DG ] and c =[P gen ] In the formula, k represents the current moment, k+1 represents the next moment, x represents the state variable, u represents the control variable, and y represents the c Represents the controlled output variable, y b represents the constrained output variable; A and B represent the system matrix and control matrix respectively, C c represents the controlled output matrix, C b represents the constraint output matrix; By finding the best control method, the controlled output variable y c and the expected output y ref The closer the better, and the fuel consumption of the diesel generator is B e minimum, for which the controlled output variable y c and the expected output y ref The cumulative error between the fuel consumption of the diesel generator and the e The sum is used as the objective function J of the multi-energy electric propulsion ship, expressed as: Where: In the formula, k+j|k represents the prediction of time k for time k+j, b e Indicates fuel consumption rate, N p represents the prediction time domain, N c represents the control time domain, Q m Represents the weighting matrix of the output error, R m represents the weighted matrix of the control increment, β is the weight factor of the state variable, λ1, λ2, λ3 represent P bat , P C , P DG The weight factor of ; Δu(k) represents the change of the control variable; Select the load prediction value P of the load unit of the multi-energy electric propulsion ship load The output power P of the photovoltaic pv The difference between ref (k+j|k); The constraints of the objective function J include: In the formula, the subscript max and subscript min represent the upper and lower limits of the corresponding variables, respectively.

10. A multi-energy electric propulsion ship energy management system based on load forecasting according to claim 9, characterized in that: The P bat , P C , P DG The weight factors λ1, λ2, and λ3 are determined according to the following steps: When P net >0, the working conditions of the battery pack and supercapacitor under different load conditions and the weight matrix R m The value includes: When P net <0, the working conditions of the battery pack and supercapacitor under different load conditions and the weight matrix R m Possible values ​​are:

Citation Information

Cited By

  • Pure electric power ship energy storage and conversion management system

    CN120423034A

  • An all-electric marine vessel energy storage and conversion management system

    CN120423034B

  • Control method, system and equipment for green-electricity-driven oil field drilling and production equipment and storage medium

    CN120575836A