Diesel-electric hybrid power ship energy management method and system based on mode division

Abstract

The invention relates to the technical field of ship power system control, in particular to a diesel-electric hybrid power ship energy management method and system based on mode division, and the method comprises the steps: dividing operation modes in a mode division layer by combining working conditions with the state of charge (SOC) of an energy storage system, and adopting an A-PSO pressure drop compensation droop control method in an equipment control layer, according to the method, the quick response of the hybrid power system to the load power requirement is realized, the bus voltage and the energy storage SOC are always in a reasonable working interval, the self-adaptive control of the power battery and the energy storage system of the ship in different operation modes is realized, the durability of the energy storage system is improved, and the replacement cost of the power battery and the operation cost of the hybrid power system are reduced. A PDEHS simulation model is established through Matlab / Simulink software, and the effectiveness and feasibility of the method are verified.

Description

Technical Field

[0001] This invention relates to the field of marine power system control technology, specifically to a mode-based energy management method and system for diesel-electric hybrid ships. Background Technology

[0002] Currently, energy optimization control methods for diesel-electric hybrid ships mainly fall into three categories: rule-based control methods, instantaneous optimization control methods, and global optimization control methods.

[0003] Rule-based control strategies and instantaneous optimization control strategies can allocate torque statically and dynamically, respectively, achieving a certain degree of fuel saving, but it is difficult to guarantee that the total fuel consumption is optimal at all times. For example, the instantaneous optimization control energy method based on the Equivalent Fuel Consumption Minimization Strategy (ECMS) can only optimize each step within a local interval, and the computational load is also large, making it difficult to implement in real time.

[0004] Energy management and control systems pose a significant challenge to the research of diesel-electric hybrid power ships due to their nonlinearity, multivariability, and time-varying characteristics. Instantaneous optimization energy management strategies, based on the idea of ​​optimization, typically use a sampling time interval as the optimization interval, establish an objective function, and employ optimization algorithms to solve it, ultimately obtaining the instantaneous optimal operating point. The goal of instantaneous optimization strategies is to achieve the optimization of instantaneous control objectives, but it cannot guarantee optimality over the entire operating interval, and the calculations are also quite complex.

[0005] Currently, numerous energy management methods have emerged, evolving from the initial traditional filter-based power reference segmentation strategy to model predictive control, multi-objective optimization, and particle swarm optimization algorithms. However, in actual navigation, ships experience frequent load changes, rendering these intelligent optimization algorithms ineffective in ship control systems. Therefore, while meeting the performance requirements of ship hybrid power systems under real-world navigation conditions, adopting relatively simple energy management methods is currently the direction of research for the practical application of diesel-electric hybrid power systems.

[0006] The goal of instantaneous optimization strategies is to achieve the optimal instantaneous control objective, but they cannot guarantee optimality over the entire operating range, and the calculation process is relatively complex. Algorithms such as PMP and ECMS are examples of this type of energy management strategy. Rule-based (Proportion Integration, PI) control methods are relatively simple to implement, but they lack good adaptability to operating conditions. Online identification can obtain the latest controlled object information in a timely manner, but it must be based on the latest model design.

[0007] While centralized energy management methods can fully utilize the advantages of different power sources to rationally allocate ship load power, they have significant drawbacks in practical engineering applications. For example, the large number of sensors required necessitates extensive communication and computation, leading to a decrease in the dynamic response capability of the energy control system's power allocation. Communication point failures can also cause the energy management system to malfunction. Furthermore, the system suffers from poor versatility, requiring redesign for different systems.

[0008] In summary, the existing technology mainly has the following problems: 1. Conventional PI controllers are difficult to adjust, have low sensitivity to parameter changes, and suffer from integral saturation problems.

[0009] 2. In the application of traditional droop control algorithms, the DC bus voltage will generate voltage drop loss, which makes them unsuitable for applications with high requirements for output voltage accuracy or regulation.

[0010] 3. To address the adverse effects of the diesel-electric hybrid power system's operating mode on the lifespan of the power battery, as well as the issues of poor energy storage SOC stability and poor power battery durability. Summary of the Invention

[0011] This invention addresses the technical problems existing in the prior art by providing a method and system for energy management of diesel-electric hybrid ships based on pattern partitioning.

[0012] The technical solution of the present invention to solve the above-mentioned technical problems is as follows: A mode-based approach to energy management for diesel-electric hybrid ships includes: Real-time data collection of ship navigation status and energy storage system state of charge allows for the classification of hybrid power systems into different operating modes. The parameters of the PI controller are optimized online using an adaptive particle swarm optimization algorithm to form an adaptive particle swarm optimized PI controller. The reference voltage compensation amount ΔV of the voltage control loop of the energy storage converter is dynamically adjusted using the adaptive particle swarm optimized PI controller to compensate and correct the DC bus voltage drop, so as to enable adaptive matching of the power battery and energy storage system of the ship under different operating modes.

[0013] Preferably, the operating mode includes: Float charging mode A is activated when the state of charge is within a reasonable range and the vessel is in a stable operating condition. Charging mode B is activated when the state of charge is below a preset lower threshold. Discharge mode C is activated when the state of charge is higher than the preset upper limit threshold or when the ship is in a high power demand condition.

[0014] Preferably, when the hybrid power system is in float charging mode A, the reference voltage of the energy storage converter voltage control loop is the bus reference voltage; When the hybrid system is in charging mode B, the reference voltage of the energy storage converter voltage control loop is the bus reference voltage minus a positive voltage compensation amount ΔV. When the hybrid power system is in discharge mode C, the reference voltage of the energy storage converter voltage control loop is the bus reference voltage minus a negative voltage compensation amount ΔV.

[0015] Preferably, the ship's navigation status includes maximum speed operation, propulsion / reverse operation, constant speed cruise operation, battery failure operation, energy storage failure operation, stop stroke operation, and start operation.

[0016] Preferably, the bus reference voltage is 500V, and the absolute value of the compensation coefficient ΔV is 5V.

[0017] Preferably, the adaptive particle swarm optimization algorithm achieves adaptation in the following way: The inertia weight ω is dynamically adjusted based on the fitness value of the particles, and the population fitness variance D is calculated. When the algorithm gets stuck in a local optimum, a mutation operation is performed on the extreme value of the particle swarm with a preset probability.

[0018] Preferably, the inertial weight ω of the i-th particle i The calculation formula is:

[0019] Where, ω max and ω min These are the maximum and minimum values ​​of the inertia weight, f. i f is the fitness value of the i-th particle. avg f is the average fitness value of the population. best This represents the optimal fitness value for the population.

[0020] Preferably, the inertia weight ω is dynamically adjusted based on the particle's fitness value, and the population fitness variance D is calculated. When the algorithm gets stuck in a local optimum, a preset probability p is used to... m Performing mutation operations on particle swarm extrema specifically includes: (1) Calculate the fitness variance D: ; ; In the formula, N is the total size of the particle swarm; f i Let f be the fitness value of the i-th particle, i = 1, 2, ..., N; avg is the average fitness value of the population; D is the fitness variance of the entire population; f is the normalization factor, used to control the range of fitness variance; (2) If the algorithm gets stuck in a local optimum, then the population extreme value of the current particle is... With a preset probability p m Mutation, p m The calculation is expressed as:

[0021] Where q is a random number in the interval [0, 0.4]. It is a preset threshold that is less than the fitness variance D; (3) Based on the particle swarm extrema Perform the following mutation operations: Generate a random number r that follows a normal distribution n(0,1); Mix the random number r with p m Compare, if r is greater than p m If the particle undergoes a mutation, the specific calculation formula for the mutation operation is as follows: ; This represents the extreme value of the particle swarm after the mutation. The extreme value of the particle swarm before the mutation, if r is less than p m If the mutation does not occur, then no mutation will occur.

[0022] Preferably, the online optimization of the PI controller parameters using the adaptive particle swarm optimization algorithm specifically includes: Collect the actual output voltage values ​​of each module in the energy storage system and calculate the average value; The error between the average value and the desired voltage is input into the integral regulator; The PI controller parameters are optimized in real time using an adaptive particle swarm optimization algorithm.

[0023] This invention also provides a mode-partition-based energy management system for diesel-electric hybrid ships. The management system is used to implement the mode-partition-based energy management method for diesel-electric hybrid ships as described above, specifically including: The mode division module is configured to collect the ship's navigation status and the charge status of the energy storage system in real time, and divide the hybrid power system into different operating modes. The control module is configured to optimize the parameters of the PI controller online using an adaptive particle swarm optimization algorithm to form an adaptive particle swarm optimized PI controller. The adaptive particle swarm optimized PI controller is used to dynamically adjust the reference voltage compensation amount ΔV of the voltage control loop of the energy storage converter to compensate and correct the DC bus voltage drop, so as to enable adaptive matching of the power battery and energy storage system in different operating modes of the ship.

[0024] The beneficial effects of this invention are: 1. By dividing the operating modes in the mode division layer according to the working conditions and the state of charge (SOC) of the energy storage system, the power system can quickly respond to the load power demand and restore the bus voltage and energy storage SOC.

[0025] 2. To address the drawbacks of particle swarm optimization (PSO), such as premature convergence, local optima, and failure to achieve global convergence, an adaptive particle swarm optimization algorithm (A-PSO) is proposed. First, the inertia weight in the velocity formula is improved by replacing it with dynamic weights. Second, an adaptive mutation optimization PSO algorithm is used to mutate certain particles, enabling them to escape local optima and obtain the global optimum.

[0026] 3. Conventional PI controllers are difficult to adjust, have low sensitivity to parameter changes, and suffer from integral saturation. To address this, the particle swarm optimization algorithm (A-PSO) is used to optimize the parameters of the PI controller, forming an A-PSO-PI controller. This improves the flexibility and adaptive characteristics of feedback control, as well as its robustness.

[0027] 4. In the equipment control layer, to address the issue that traditional droop control algorithms cause voltage drop losses on the DC bus, making them unsuitable for applications requiring high output voltage accuracy or regulation, an A-PSO voltage drop compensation droop control method is designed to compensate and correct the DC bus voltage drop, enabling adaptive control of the power battery and energy storage system under different operating modes. Attached Figure Description

[0028] Figure 1 This is a schematic diagram of the A-PSO strategy process provided by the present invention; Figure 2 This is a schematic diagram of the A-PSO-based optimized PI voltage drop compensation droop control method provided by the present invention; Figure 3 A schematic diagram illustrating the different operating modes of the driver provided by the present invention; Figure 4 A schematic diagram of the hierarchical energy management method based on pattern partitioning provided by the present invention. Detailed Implementation

[0029] The technical solutions of the embodiments of this application will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of this application, and not all embodiments. Based on the embodiments of this application, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of this application.

[0030] In the description of this application, the terms "first" and "second" are used for descriptive purposes only and should not be construed as indicating or implying relative importance or implicitly specifying the number of indicated technical features. Thus, a feature defined as "first" or "second" may explicitly or implicitly include one or more of the stated features. In the description of this application, "multiple" means two or more, unless otherwise explicitly specified.

[0031] In the description of this application, the term "for example" is used to mean "used as an example, illustration, or description." Any embodiment described as "for example" in this application is not necessarily to be construed as being more preferred or advantageous than other embodiments. The following description is provided to enable any person skilled in the art to make and use the invention. Details are set forth in the following description for purposes of explanation. It should be understood that those skilled in the art will recognize that the invention can be made without using these specific details. In other instances, well-known structures and processes will not be described in detail to avoid obscuring the description of the invention with unnecessary detail. Therefore, the invention is not intended to be limited to the embodiments shown, but is consistent with the broadest scope of the principles and features disclosed in this application.

[0032] Example 1 Combination Figures 1 to 4 This invention provides an energy management method for diesel-electric hybrid ships based on pattern partitioning, comprising the following steps: Step 1: To facilitate the design and verification of control strategies, a diesel-electric hybrid power ship model is first established.

[0033] (1) Diesel Engine Model. Energy optimization of parallel hybrid power ships requires consideration of engine fuel consumption. Considering the complexity of the engine, a quasi-static model was established. Fuel Rate It can be represented as: (1) In the formula, and These are engine speed and torque, respectively.

[0034] (2) In the formula, Indicates the engine's available power. This indicates the lower heating value of the fuel.

[0035] (2) Motor Model. The motor is a three-phase asynchronous motor, which can be used as a traction motor to provide torque and as a generator to charge the battery. Therefore, the motor power Pm can be expressed as: (3) In the formula: This refers to the motor speed; The efficiency of the motor driving the motor; Tm represents the efficiency of the motor as a generator, and Tm represents the torque of the motor.

[0036] (3) Battery Model. Ignoring the effect of temperature on the battery, a circular model is used to model the battery, and the expressions for the battery SOC and output power are as follows: (4) (5) In equation (5), SOC (State of Charge) is the rate of change of battery SOC; U oc Rb is the battery open-circuit voltage; Q is the battery internal resistance; b P represents the battery's maximum capacity. b Battery output power P bat It is an intermediate variable.

[0037] (4) Transmission system model. This paper uses a dual reduction ratio to transmit power, namely the main reduction ratio and the PTI (Power Take-In) reduction ratio: (6) In the formula, and These are the output torque and speed; and These are the input torque and speed. This refers to the transmission ratio of the transmission system. A parallel gearbox is used as the torque synthesis device. In this case, only the superposition of torques is considered. The modeling is as follows: (7) In the formula, and These are the two input torques of the parallel gearbox; It refers to the output torque.

[0038] (5) Engine-propeller mathematical model: The equations of motion for the ship and propeller are as follows: (8)

[0039] (9)

[0040] In the formula, It is the speed of the ship, in m / s; It is the resistance of the ship, N; It is the torque output by the diesel engine, in N·m; is the resistance torque, N·m; i is the reduction ratio of the gearbox.

[0041] (10)

[0042] In the formula, In the formula, λ is the ship resistance coefficient. It is a constant.

[0043] Step 2: Collect real-time data on the ship's navigation status and the state of charge of the energy storage system to classify the hybrid power system into different operating modes.

[0044] like Figure 3 As shown, ship navigation conditions are divided into seven types (maximum speed, propulsion / reverse, constant speed cruise, battery failure, energy storage failure, stop stroke, and start-up). Based on the SOC status of the energy storage system, three operating modes, A, B, and C, are ultimately determined.

[0045] Float charging mode A is activated when the state of charge (SOC) is within a reasonable range and the vessel is in stable operating conditions. In this mode, the SOC is within the ideal range of 50% to 70%, and the vessel's power demand is stable. The system maintains a stable bus voltage, and the battery is essentially neither charged nor discharged. When the hybrid power system is in float charging mode A, the reference voltage of the energy storage converter voltage control loop is the bus reference voltage. Reference voltage = 500V (ΔV = 0V).

[0046] Charging mode B is activated when the state of charge (SOC) is below a preset lower threshold; in this mode, the SOC is below 40%. The system compensates by appropriately increasing the bus voltage reference value to guide energy flow to charge the battery. When the hybrid system is in charging mode B, the reference voltage of the energy storage converter voltage control loop is the bus reference voltage minus a positive voltage compensation amount ΔV. Reference voltage = 500V - 5V (ΔV = +5V).

[0047] Discharge mode C is activated when the state of charge (SOC) exceeds a preset upper limit threshold or the vessel is operating under high power demand conditions. In this mode, the SOC is above 80% or the vessel is under high power conditions such as rapid acceleration or maximum speed. The system compensates by appropriately reducing the bus voltage reference value to guide battery discharge and assist the diesel engine in meeting power requirements. When the hybrid power system is in discharge mode C, the reference voltage of the energy storage converter voltage control loop is the bus reference voltage minus a negative voltage compensation amount ΔV. Reference voltage = 500V - (-5V) = 505V (ΔV = -5V).

[0048] Step 3: The parameters of the PI controller are optimized online using an adaptive particle swarm optimization algorithm to form an adaptive particle swarm optimized PI controller. This adaptive particle swarm optimized PI controller is then used to dynamically adjust the reference voltage compensation amount ΔV of the energy storage converter voltage control loop to compensate for the DC bus voltage drop, enabling adaptive matching between the power battery and energy storage system under different operating modes of the ship. This includes three steps: 3.1. A-PSO-PI Controller Design: Adaptive Particle Swarm Optimization (A-PSO) algorithm is used to optimize the proportional gain K of the traditional PI controller. p and integral coefficient K i Perform online optimization. The A-PSO algorithm process is as follows: Figure 1 As shown, it achieves adaptive weighting and adaptive mutation through formulas (32) to (38), continuously searching for the optimal controller parameters that minimize the voltage error. The specific process is as follows: 3.1.1. Dynamically adjust the weight ω of each particle in the particle swarm based on the particle's fitness value, where the weight of the i-th particle is ω. i The calculation process for adaptive weights is as follows: (32); In the formula, This represents the maximum weight. This is the minimum weight, usually taken as... 0.9, 0.4; Let be the fitness value of the i-th particle. The average fitness value of the entire particle. This represents the optimal fitness value for the entire particle.

[0049] The velocity and position corresponding to the i-th particle can be updated as follows: (33); (34); In the formula, It is the adaptive weight corresponding to the i-th particle; Let be the velocity of the i-th particle after the k-th iteration; Let be the position of the i-th particle after the k-th iteration; and It is a constant learning factor. and is a random number in the interval [0,1]; k is the number of evolution iterations; m is the population size; d is the spatial dimension of the problem to be solved. 1≤i≤m, 1≤j≤d. It is the single optimal value of the i-th particle in the k-th iteration. It is the global optimal value in the k-th iteration.

[0050] 3.1.2. Adaptive Mutation. The specific calculation of fitness variance is as follows: (35); In the formula, N is the total size; Let be the fitness value of the i-th particle, where i = 1, 2, ..., N; denoted as the average fitness value of the entire population; D is the variance of the fitness of the entire population. This is a normalization factor used to control the range of fitness variance. A specific value is dynamically adjusted based on the current particle's fitness value.

[0051] (36); If the algorithm gets stuck in a local optimum, then the population extremum of the current particle is... With a certain probability mutation, The specific calculation is expressed as follows (37); Where q is a random number in the interval [0, 0.4], and It is much smaller than the maximum value of the fitness variance D, and is the preset threshold. Based on the particle swarm extreme value... Perform the following mutation operation.

[0052] (1) Randomly generate a random number r that follows a normal distribution n (0,1).

[0053] (2) Combine the random number r with Compare them. If r is greater than... If this occurs, the particle undergoes a mutation. The specific calculation formula for the mutation operation is as follows: (38) in, This represents the extreme value of the particle swarm after the mutation. The extreme value of the particle swarm before the mutation, if r is less than If the mutation does not occur, then no mutation will occur.

[0054] 3.2. Voltage Drop Compensation Process: Collect the actual output voltage values ​​of each module in the energy storage system and calculate the average value. That is, collect the actual output voltage values ​​of each DC-DC parallel module and calculate the average value f. avg f avg The error signal is compared with the desired bus voltage (e.g., 500V) and input to a PI controller optimized by A-PSO. The output of this controller is the required voltage compensation amount ΔV.

[0055] In the equipment control layer, the designed A-PSO voltage drop compensation droop control method is adopted to automatically adjust the voltage compensation coefficient of the energy storage system droop control according to the navigation mode, as shown in formula (39).

[0056] (39); In the formula, This is the bus voltage ripple factor, which is usually taken as 0.01.

[0057] When the hybrid system is in mode A, the reference voltage of the energy storage converter voltage control loop is the bus reference voltage V. nom 500V; When the hybrid system is in mode B, the reference voltage of the energy storage converter voltage control loop is the bus reference voltage 500V minus a positive bus voltage compensation coefficient ΔV=5V; When the hybrid system is in mode C, the reference voltage of the energy storage converter voltage control loop is the bus reference voltage minus a negative bus voltage compensation coefficient ΔV=-5V.

[0058] Finally, substituting the voltage compensation coefficient into formula (39), under the premise of satisfying fast response, the energy storage system can simultaneously charge or discharge, thereby keeping the SOC of the power battery within a reasonable operating range. The power battery and energy storage system can then perform adaptive control under different navigation modes. The control flowchart of the mode-based hierarchical energy management method is shown in Figure 4.

[0059] The structural diagram of the A-PSO voltage drop compensation droop control method proposed in this embodiment of the invention is as follows: Figure 2 As shown. A-PSO optimizes PI control, improving the flexibility and adaptive characteristics of feedback control, and enhancing its robustness. The DC-DC parallel module internally employs a traditional droop control dual-closed-loop control system, with an inner current loop and an outer voltage loop. R x For virtual resistance, V ref R is the voltage control command, and I is the output current feedback value. x The product of I and V is used as the control quantity for the voltage setpoint in the droop control algorithm. This part mainly acts as the current sharing control quantity and is the direct cause of the voltage drop loss in the module's output voltage. It can also be seen that V... ref This actually corresponds to the open-circuit voltage in the module's power supply output characteristics. Therefore, it can be adjusted by changing V. ref This is to compensate for the voltage drop loss at the power supply output. The method involves uniformly adjusting the voltage setpoints of all modules within the power supply to compensate for the voltage drop on the DC bus. The method proposed in this patent involves externally adjusting the V... ref A unified adaptive voltage drop compensation correction is performed. At the same time, the actual output voltage values ​​of each module are collected and averaged. The error between the average value and the expected output voltage value of the power supply is input into the integral regulator. Then, the PI controller is optimized in real time through A-PSO. The output of the controller is used as the compensation amount and added to the original voltage setpoint to obtain the new voltage setpoint.

[0060] Through the above process, this invention achieves precise and adaptive control of the DC bus voltage under different navigation conditions and battery statuses, ensuring reasonable power distribution and stable and efficient system operation. Matlab / Simulink simulations verify that this method can effectively maintain the battery SOC within an excellent range of 60% ± 10% and significantly reduce DC bus voltage fluctuations. Furthermore, while meeting the requirement of rapid response, the energy storage system can perform adaptive charging and discharging, keeping the power battery SOC always within a reasonable operating range.

[0061] Example 2 This invention also provides a mode-partition-based diesel-electric hybrid power ship energy management system, which is used to implement the mode-partition-based diesel-electric hybrid power ship energy management method described above, specifically including: The mode division module is configured to collect the ship's navigation status and the charge status of the energy storage system in real time, and divide the hybrid power system into different operating modes. The control module is configured to optimize the parameters of the PI controller online using an adaptive particle swarm optimization algorithm to form an adaptive particle swarm optimized PI controller. The adaptive particle swarm optimized PI controller is used to dynamically adjust the reference voltage compensation amount ΔV of the voltage control loop of the energy storage converter to compensate and correct the DC bus voltage drop, so as to enable adaptive matching of the power battery and energy storage system in different operating modes of the ship.

[0062] The beneficial effects of this invention are: 1. Fast response and stable SOC: Through mode partitioning, a fast response to load power demand is achieved, and the bus voltage and energy storage SOC are effectively maintained within a reasonable operating range, significantly improving system stability.

[0063] 2. Advanced optimization algorithm: The proposed A-PSO algorithm effectively avoids premature convergence through adaptive weights and mutation operations, and can obtain the global optimal solution, providing a reliable guarantee for controller parameter optimization.

[0064] 3. Strong control robustness: The A-PSO-PI controller overcomes the inherent defects of conventional PI controllers and greatly improves the flexibility, adaptability and robustness of feedback control.

[0065] 4. High voltage accuracy: The A-PSO voltage drop compensation droop control method accurately corrects the DC bus voltage drop, solving the problem of poor applicability of traditional droop control in high-precision applications.

[0066] 5. Extended battery life: Through reasonable mode division and adaptive control, the durability of the energy storage system is effectively improved, and the replacement cost of the power battery and the operating cost of the entire hybrid power system are reduced.

[0067] While embodiments or examples of this disclosure have been described with reference to the accompanying drawings, it should be understood that the above embodiments are merely exemplary embodiments or examples, and the scope of the invention is not limited by these embodiments or examples, but only by the granted claims and their equivalents. Various elements in the embodiments or examples may be omitted or replaced by their equivalents. Furthermore, the steps may be performed in a different order than that described in this disclosure. Further, various elements in the embodiments or examples may be combined in various ways. Importantly, as the technology evolves, many elements described herein can be replaced by equivalents that appear after this disclosure.

Claims

1. A method for energy management of a diesel-electric hybrid vessel based on mode partitioning, characterized by, The method comprises the following steps: Real-time acquisition of the ship navigation state and the state of charge of the energy storage system, and division of the hybrid power system to obtain different operating modes; An adaptive particle swarm optimization algorithm is used to online optimize the parameters of the PI controller to form an adaptive particle swarm optimization PI controller, and the adaptive particle swarm optimization PI controller is used to dynamically adjust the reference voltage compensation amount ΔV of the energy storage converter voltage control loop to compensate and correct the DC bus voltage drop, so that the adaptive matching of the power battery and the energy storage system of the ship under different operating modes is realized.

2. The mode partition based diesel-electric hybrid ship energy management method according to claim 1, characterized in that, The operating modes include: Float charging mode A, which is enabled when the state of charge is in a reasonable interval and the ship is in a stable working condition; Charging mode B, which is enabled when the state of charge is lower than a preset lower threshold; Discharging mode C, which is enabled when the state of charge is higher than a preset upper threshold or the ship is in a high power demand working condition.

3. The diesel-electric hybrid ship energy management method based on mode division according to claim 2, wherein: When the hybrid power system is in the float charging mode A state, the reference voltage of the energy storage converter voltage control loop is the bus reference voltage; When the hybrid power system is in the charging mode B, the reference voltage of the energy storage converter voltage control loop is the bus reference voltage minus a positive voltage compensation amount ΔV; When the hybrid power system is in the discharging mode C, the reference voltage of the energy storage converter voltage control loop is the bus reference voltage minus a negative voltage compensation amount ΔV.

4. The mode partition based diesel-electric hybrid ship energy management method according to claim 1, characterized in that, The ship navigation state includes maximum speed working condition, propulsion / reverse working condition, constant speed cruising working condition, battery failure working condition, energy storage failure working condition, stop ship charging working condition and starting working condition.

5. The mode partition based diesel-electric hybrid ship energy management method of claim 1, wherein, The bus reference voltage is 500V, and the absolute value of the compensation coefficient ΔV is 5V.

6. The mode partition based diesel-electric hybrid ship energy management method of claim 1, wherein, The adaptive particle swarm optimization algorithm realizes self-adaption in the following way: The inertia weight ω is dynamically adjusted according to the fitness value of the particle, and the population fitness variance D is calculated, and when the algorithm falls into local optimum, mutation operation is performed on the particle swarm extreme value with a preset probability.

7. The mode partition based diesel-electric hybrid marine vessel energy management method of claim 6, wherein, The inertia weight ω of the ith particle i The calculation formula is: ； where ω max and ω min are the maximum and minimum of the inertia weight, respectively, f i is the fitness value of the ith particle, f avg is the population average fitness value, and f best is the population best fitness value.

8. The mode partition based diesel-electric hybrid ship energy management method according to claim 6, characterized in that, According to the fitness value of the particle, the inertia weight ω is dynamically adjusted, and the population fitness variance D is calculated, and when the algorithm falls into local optimum, a preset probability p is taken m The mutation operation is performed on the particle swarm extreme value, specifically including: (1) Calculate the fitness variance D: ； ； where N is the total population size; f i is the fitness value of the ith particle, i = 1, 2,..., N; f avg is the population average fitness value; D is the fitness variance of the entire population; and f is a normalization factor used to control the range of the fitness variance; (2) If the algorithm falls into a local optimal solution, the population extremum of the current particle with a preset probability p m mutation, p m is calculated as: ； wherein q is a random number in the interval [0, 0.4], is a preset threshold value less than the fitness variance D; (3) According to the particle swarm extremum , the following mutation operation is performed: A random number r following the normal distribution n (0, 1) is randomly generated; The random number r is compared with p m If r is greater than p m Then the particle mutates, and the specific calculation formula of the mutation operation is as follows: ； is the mutation after the particle swarm extremum, is the mutation before the particle swarm extremum, if r is less than p m mutation is not performed.

9. The mode partition based diesel-electric hybrid ship energy management method of claim 1, wherein, The adaptive particle swarm optimization algorithm online optimizes the parameters of the PI controller, which specifically includes: The actual output voltage values of each module in the energy storage system are collected and averaged; The error between the average value and the expected voltage is input into the integral regulator; The PI controller parameters are optimized in real time by the adaptive particle swarm optimization algorithm.

10. A diesel-electric hybrid vessel energy management system based on mode partitioning, characterized by, The management system is used to realize the diesel-electric hybrid ship energy management method based on mode division as claimed in any one of claims 1 to 9, and specifically includes: A mode division module configured to real-time acquisition of the ship navigation state and the state of charge of the energy storage system, and division of the hybrid power system to obtain different operating modes; A control module configured to use an adaptive particle swarm optimization algorithm to online optimize the parameters of the PI controller to form an adaptive particle swarm optimization PI controller, and use the adaptive particle swarm optimization PI controller to dynamically adjust the reference voltage compensation amount ΔV of the energy storage converter voltage control loop to compensate and correct the DC bus voltage drop, so that the adaptive matching of the power battery and the energy storage system of the ship under different operating modes is realized.

Images