A ship power system optimal scheduling method and system based on a genetic algorithm

Abstract

The application discloses a ship power system optimization scheduling method and system based on a genetic algorithm, and relates to the technical field of ship power control. The method comprises the following steps: acquiring real-time operation parameters of a ship power system; constructing an initial population of the genetic algorithm based on the operation parameters; calculating a comprehensive fitness value of each individual based on energy consumption characteristic indexes and emission characteristic indexes; performing selection operation, crossover operation and mutation operation on the initial population to generate a new population; repeatedly performing population iteration optimization until a preset condition is met; outputting a combination of ship speed adjustment parameters, a generator set start-stop decision sequence and a power distribution scheme; and generating a ship power system control instruction set. The ship power system optimization scheduling method and system based on the genetic algorithm improve the operation efficiency of the ship power system, reduce energy consumption and reduce carbon emissions.

Description

Technical Field

[0001] This application relates to the field of ship power control technology, and in particular to a ship power system optimization scheduling method and system based on genetic algorithm. Background Technology

[0002] Currently, most ship power control systems employ traditional controllers and fixed strategies, primarily controlling power output by adjusting single parameters such as fuel quantity. This method has significant drawbacks in hybrid-powered vessels that require simultaneous use of fuel generators and battery power: firstly, it is difficult to balance fuel efficiency and environmental protection, often resulting in a trade-off; secondly, in complex situations such as wind and waves, fixed control rules can lead to unstable power output, easily causing accelerated equipment wear and energy waste.

[0003] Existing optimization technologies have limited ability to comprehensively analyze diverse data such as generator power, battery status, and propeller load. Although some studies have attempted to use genetic algorithms for optimization, they often focus only on a single objective of fuel saving or emission reduction, resulting in fixed optimization strategies. This leads to situations where ships cannot prioritize fuel saving when battery power is insufficient, and cannot quickly turn to reduce emissions when encountering sudden storms, making it difficult to improve overall operational efficiency. Summary of the Invention

[0004] To overcome the shortcomings of the prior art, this application provides a system and method for displaying user interface layout information based on the HarmonyOS system.

[0005] Firstly, this application provides a method for optimizing the scheduling of a ship's power system based on a genetic algorithm, the method comprising:

[0006] The system acquires real-time operating parameters of the ship's power system, including main engine speed data, generator output power data, remaining battery charge data, and propeller load characteristic data.

[0007] The initial population of the genetic algorithm is constructed based on the operating parameters. Each individual in the initial population includes the gene encoding of the speed regulation parameters, generator start-stop decision parameters, and power allocation ratio parameters.

[0008] Based on energy consumption characteristic indicators and emission characteristic indicators, the comprehensive fitness value of each individual is calculated. The weight coefficient corresponding to the energy consumption characteristic indicators is related to the remaining power of the energy storage battery, and the weight coefficient corresponding to the emission characteristic indicators is related to the fluctuation range of the load characteristic data of the thruster.

[0009] Selection, crossover, and mutation operations are performed on the initial population to generate a new population, where the mutation probability adopts an adaptive adjustment strategy based on the number of iterations.

[0010] Repeatedly perform population iteration optimization until the preset conditions are met, output the combination of speed regulation parameters, generator set start-stop decision parameters and power allocation scheme, and generate the ship's power system control command set.

[0011] Preferably, the range of values ​​for the speed adjustment parameter is determined according to the following formula:

[0012]

[0013] Where V_base is the base speed adjustment range, P_load is the real-time load power of the thruster, and P_rated is the rated power of the thruster;

[0014] The generator set start-up and shutdown decision parameters are encoded as binary bits, where 0 represents the shutdown state and 1 represents the running state.

[0015] The coding range for the power allocation ratio is determined based on the rated power parameters of each generator set, wherein the power allocation ratio must meet the following constraints:

[0016]

[0017] Where α_i is the power allocation ratio of the i-th generator set, P_min_i is its minimum allowable output power, and P_total is the total load demand.

[0018] Preferably, the overall fitness value is calculated using the following formula:

[0019]

[0020] Wherein, I1 is the energy consumption characteristic index, I2 is the emission characteristic index, W1 is the weighting coefficient of the energy consumption characteristic index, and W2 is the weighting coefficient of the emission characteristic index.

[0021] Preferably, the fuel consumption is obtained by multiplying the real-time output power value of each generator set by the corresponding fuel consumption coefficient per unit power; the equipment loss is obtained by multiplying the number of start-stop cycles of each generator set by the single start-stop loss coefficient; and the energy consumption characteristic index is obtained by adding the fuel consumption and equipment loss.

[0022] The total carbon emissions are obtained by multiplying the real-time output power of each generator set by the corresponding carbon emission coefficient per unit power; the noise pollution index is obtained by dividing the measured noise value of the propeller area by the maximum noise limit; and the emission characteristic index is obtained by multiplying the total carbon emissions by the noise pollution index.

[0023] Preferably, when the remaining power of the energy storage battery is lower than a preset safety threshold, the weighting coefficient of the energy consumption characteristic index is increased;

[0024] When the load characteristic data of the thruster fluctuates beyond the historical average fluctuation range, the weighting coefficient of the emission characteristic index is increased.

[0025] Preferably, the mutation probability of the current iteration period is calculated using the following formula:

[0026]

[0027] Where P_m represents the mutation probability, P_b is the basic mutation probability parameter, G is the current iteration number, G_max is the maximum iteration number, σ_F is the standard deviation of the population fitness value, and μ_F is the fitness mean.

[0028] Based on the mutation probability, mutation operations are performed on individuals in the population. These mutation operations include applying random perturbations to the speed regulation parameters, reversing the state of the generator start-stop decision parameters, and redistributing the power allocation ratio parameters.

[0029] Preferably, a digital twin simulation model of the ship's power system is established, and the control command set of the ship's power system is input into the digital twin simulation model to perform virtual operation;

[0030] The system monitors the power fluctuation rate and voltage deviation rate during the virtual operation process in real time. When the power fluctuation rate exceeds the first preset value or the voltage deviation rate exceeds the second preset value, the optimization parameter reset mechanism is triggered.

[0031] The process of optimizing the parameter reset mechanism includes regenerating the initial population individuals, adjusting the calculation parameters for mutation probability, and restarting the genetic algorithm optimization process.

[0032] Preferably, a multi-ship collaborative optimization database is established to store various typical navigation environment characteristic parameters and corresponding optimal gene coding schemes;

[0033] The matching degree between the current navigation environment and historical cases is calculated by a feature similarity matching algorithm. When the similarity exceeds the third preset value, the gene code of the corresponding historical case is loaded as the initial population seed.

[0034] The historical gene encoding was adapted using a parameter normalization method to make it conform to the current range of ship power system parameters.

[0035] Secondly, a ship propulsion system optimization scheduling system based on genetic algorithms includes:

[0036] The operating parameter acquisition unit is used to acquire real-time operating parameters of the ship's power system, including main engine speed data, generator output power data, remaining battery charge data, and propeller load characteristic data.

[0037] Gene coding units are used to construct the initial population of the genetic algorithm based on the operating parameters. Each individual in the initial population includes the gene coding of the speed regulation parameters, generator start-stop decision parameters, and power allocation ratio parameters.

[0038] The comprehensive fitness value calculation unit is used to calculate the comprehensive fitness value of each individual based on energy consumption characteristic indicators and emission characteristic indicators. The weight coefficient corresponding to the energy consumption characteristic indicators is related to the remaining power of the energy storage battery, and the weight coefficient corresponding to the emission characteristic indicators is related to the fluctuation range of the load characteristic data of the thruster.

[0039] The population generation unit is used to perform selection, crossover, and mutation operations on the initial population to generate a new population, wherein the mutation probability adopts an adaptive adjustment strategy based on the number of iterations.

[0040] The instruction set generation unit is used to repeatedly perform population iteration optimization until preset conditions are met, outputting combinations of speed regulation parameters, generator set start-stop decision parameters, and power allocation schemes to generate a ship power system control instruction set.

[0041] Compared with the prior art, the present invention has the following characteristics and beneficial effects:

[0042] First, by acquiring multi-source operating parameters of the ship's propulsion system in real time, the system's real-time status is comprehensively perceived, providing accurate data support for optimized scheduling. Then, a multi-dimensional initial population encompassing speed regulation, generator start-up and shutdown, and power allocation is constructed based on a genetic algorithm. This overcomes the rigidity of fixed-rule strategies, improves the search efficiency of the solution space, and avoids getting trapped in local optima. Furthermore, by dynamically weighting and fusing energy consumption and emissions indicators, intelligent balance is achieved in multi-objective conflicts, improving fuel efficiency and reducing carbon emissions. Next, combined with an adaptive mutation probability adjustment strategy, the algorithm effectively balances global search and local convergence, improving iteration speed and optimization stability. This enables the rapid generation of reliable scheduling schemes under complex sea conditions, reducing equipment wear and tear. Attached Figure Description

[0043] Figure 1 This embodiment is a flowchart illustrating the steps of a ship power system optimization scheduling method based on a genetic algorithm.

[0044] Figure 2 This is a structural block diagram of a ship power system optimization scheduling system based on a genetic algorithm, which is the main feature of this embodiment. Detailed Implementation

[0045] The present invention will be further described in detail below with reference to the following embodiments.

[0046] Reference Figure 1A method for optimal scheduling of ship propulsion systems based on genetic algorithms, comprising the following steps:

[0047] S1. Obtain real-time operating parameters of the ship's power system, including main engine speed data, generator output power data, remaining battery charge data, and propeller load characteristic data.

[0048] S2. Construct an initial population for the genetic algorithm based on the operating parameters. Each individual in the initial population includes the gene encoding of the speed regulation parameters, generator start-stop decision parameters, and power allocation ratio parameters.

[0049] S3. Based on energy consumption characteristic indicators and emission characteristic indicators, calculate the comprehensive fitness value of each individual. The weight coefficient corresponding to the energy consumption characteristic indicators is related to the remaining power of the energy storage battery, and the weight coefficient corresponding to the emission characteristic indicators is related to the fluctuation range of the load characteristic data of the thruster.

[0050] S4. Perform selection, crossover, and mutation operations on the initial population to generate a new population, where the mutation probability adopts an adaptive adjustment strategy based on the number of iterations.

[0051] S5. Repeatedly perform population iteration optimization until the preset conditions are met, output the combination of speed regulation parameters, generator set start-stop decision parameters and power allocation scheme, and generate the ship power system control command set.

[0052] First, by acquiring multi-source operating parameters of the ship's propulsion system in real time, the system's real-time status is comprehensively perceived, providing accurate data support for optimized scheduling. Then, a multi-dimensional initial population encompassing speed regulation, generator start-up and shutdown, and power allocation is constructed based on a genetic algorithm. This overcomes the rigidity of fixed-rule strategies, improves the search efficiency of the solution space, and avoids getting trapped in local optima. Furthermore, by dynamically weighting and fusing energy consumption and emissions indicators, intelligent balance is achieved in multi-objective conflicts, improving fuel efficiency and reducing carbon emissions. Next, combined with an adaptive mutation probability adjustment strategy, the algorithm effectively balances global search and local convergence, improving iteration speed and optimization stability. This enables the rapid generation of reliable scheduling schemes under complex sea conditions, reducing equipment wear and tear.

[0053] The specific process of step S1 can be as follows: A speed sensor installed on the main shaft system collects speed signals in real time. These signals are filtered and amplified by a signal conditioning circuit, then converted into digital signals by a data acquisition card and transmitted to the central processing unit. Generator output power data is obtained from the generator controller via a bus, with a sampling frequency of 10Hz. The remaining battery capacity data is obtained through an estimation algorithm of the battery management system (BMS), which combines the ampere-hour integration method and the open-circuit voltage method. The propeller load characteristic data, including torque and vibration signals, are collected by a flange-type torque sensor and an acceleration sensor, respectively, with a sampling frequency of 1kHz. These operating parameters provide comprehensive system status information for subsequent optimized scheduling.

[0054] The specific step S2 includes the following sub-steps:

[0055] The range of values ​​for the speed adjustment parameters is determined using the following formula:

[0056]

[0057] Where V_base is the base speed adjustment range, P_load is the real-time load power of the thruster, and P_rated is the rated power of the thruster;

[0058] The generator set start-up and shutdown decision parameters are encoded as binary bits, where 0 represents the shutdown state and 1 represents the running state.

[0059] The coding range for the power allocation ratio is determined based on the rated power parameters of each generator set, wherein the power allocation ratio must meet the following constraints:

[0060]

[0061] Where α_i is the power allocation ratio of the i-th generator set, P_min_i is its minimum allowable output power, and P_total is the total load demand.

[0062] Specifically, the dynamic adjustment mechanism effectively reflects the real-time matching relationship between the main engine's operating status and load demand. For example, when the main engine speed is close to the rated value, the speed adjustment range tends to stabilize, avoiding the infeasibility of the solution due to speed deviation; while when the propeller load power increases significantly, the upper limit of speed expands accordingly to adapt to the power demand under high load conditions. In addition, the power allocation ratio must satisfy the sum constraint (Σα_i=1) and the lower limit constraint (α_i≥P_min_i / P_total) to ensure that the output power of each generator set meets the total load demand while avoiding falling below the minimum stable operating threshold. The mathematical expression of this constraint directly embeds the physical limitations of the equipment into the encoding process of the genetic algorithm, significantly reducing the probability of generating invalid solutions. The generator set start-stop decision parameters adopt binary encoding and introduce Gray code design. The characteristic that adjacent values ​​differ by only one bit can reduce the mutation amplitude during genetic operations (such as crossover or mutation), thereby enhancing the algorithm's smooth search capability in the solution space. By comprehensively applying dynamic parameter range adjustment, physical constraint embedding, and noise-resistant coding design, the proportion of feasible solutions in the initial population is significantly improved. The convergence speed and stability of the algorithm under complex conditions are optimized, laying a high-quality search starting point for subsequent iterations.

[0063] The specific step S3 includes the following sub-steps:

[0064] The overall fitness value is calculated using the following formula:

[0065]

[0066] Wherein, I1 is the energy consumption characteristic index, I2 is the emission characteristic index, W1 is the weighting coefficient of the energy consumption characteristic index, and W2 is the weighting coefficient of the emission characteristic index.

[0067] The fuel consumption is obtained by multiplying the real-time output power of each generator set by the corresponding fuel consumption coefficient per unit power; the equipment loss is obtained by multiplying the number of start-stop cycles of each generator set by the single start-stop loss coefficient; and the energy consumption characteristic index is obtained by adding the fuel consumption and equipment loss.

[0068] The total carbon emissions are obtained by multiplying the real-time output power of each generator set by the corresponding carbon emission coefficient per unit power; the noise pollution index is obtained by dividing the measured noise value of the propeller area by the maximum noise limit; and the emission characteristic index is obtained by multiplying the total carbon emissions by the noise pollution index.

[0069] When the remaining power of the energy storage battery is lower than the preset safety threshold, the weighting coefficient of the energy consumption characteristic index is increased.

[0070] When the load characteristic data of the thruster fluctuates beyond the historical average fluctuation range, the weighting coefficient of the emission characteristic index is increased.

[0071] Specifically, the fuel consumption is first calculated based on the real-time output power of the generator set and the corresponding fuel consumption coefficient per unit power, where the fuel consumption coefficient per unit power is obtained by looking up the engine characteristic curve in a table. Then, the number of start-stop cycles of the generator set is multiplied by the single start-stop loss coefficient to obtain the equipment loss, which takes into account the unit's thermal fatigue and mechanical wear. Finally, the fuel consumption and equipment loss are added together to obtain the energy consumption characteristic index. The calculation process for the emission characteristic index I2 includes: first, calculating the total carbon emissions based on the real-time output power of the generator set and the corresponding carbon emission coefficient per unit power, where the carbon emission coefficient takes into account fuel type and combustion efficiency; then, obtaining the measured noise value of the propeller area and processing it through A-weighted filtering, dividing it by the maximum allowable noise limit to obtain the noise pollution index; finally, multiplying the total carbon emissions by the noise pollution index to obtain the emission characteristic index. The dynamic adjustment strategy for the weighting coefficients is as follows: when the remaining charge of the energy storage battery is lower than a preset safety threshold, the weighting coefficient W1 of the energy consumption characteristic index is increased linearly; when the fluctuation range of the propeller's load characteristic data exceeds the historical average fluctuation range, the weighting coefficient W2 of the emission characteristic index is smoothly increased using a Sigmoid function.

[0072] The specific step S4 includes the following sub-steps:

[0073] The mutation probability of the current iteration period is calculated using the following formula:

[0074]

[0075] Where P_m represents the mutation probability, P_b is the basic mutation probability parameter, G is the current iteration number, G_max is the maximum iteration number, σ_F is the standard deviation of the population fitness value, and μ_F is the fitness mean.

[0076] Based on the mutation probability, mutation operations are performed on individuals in the population. These mutation operations include applying random perturbations to the speed regulation parameters, reversing the state of the generator start-stop decision parameters, and redistributing the power allocation ratio parameters.

[0077] Specifically, this formula incorporates a dual feedback mechanism: an exponential decay term (exp(-G / G_max)) results in a high initial mutation probability, promoting extensive exploration of the solution space; as iterations progress, the mutation probability gradually decreases to focus on fine-grained local searches. Simultaneously, a fitness statistic compensation term (σ_F / μ_F) dynamically adjusts the mutation intensity based on population diversity. When population fitness tends to be uniform (σ_F is small), the mutation probability is increased to avoid premature convergence; conversely, it is decreased to maintain stability. In practical operation, the speed adjustment parameters employ a Gaussian perturbation strategy, with the perturbation amplitude proportional to the dynamic speed range, ensuring that the perturbation neither deviates excessively from physical limits nor fails to effectively explore potential optimization directions. The mutation probability of generator start-stop decisions is positively correlated with load volatility; when the propeller load characteristics change drastically, the algorithm responds quickly to dynamic conditions by increasing the state reversal probability. The power allocation ratio is redistributed through a differential evolution strategy, randomly adjusting the power of each unit while maintaining the total constant, balancing global optimization and local adjustments. The integrated application of these technical solutions enables the algorithm to intelligently adjust its search strategy based on iterative stages and environmental changes, avoiding getting trapped in local optima while achieving rapid convergence when close to the optimal solution. Through the deep integration of mathematical modeling and physical constraints, this approach significantly improves the efficiency and robustness of optimized scheduling of ship propulsion systems, providing reliable technical support for real-time control under complex sea conditions.

[0078] The specific process of step S5 can be achieved by setting two termination conditions: the first is a convergence condition, which is considered convergent when the fitness value of the optimal individual improves by less than 0.1% over 20 consecutive generations; the second is a maximum iteration count condition, set to 100 generations to ensure the algorithm can complete within a reasonable time. The output includes: a combination of speed regulation parameters, showing the speed change curve over a future scheduling cycle (which could be 15 minutes); generator start-stop decision parameters, determining the start-stop times for each generator unit within the future scheduling cycle; and a power allocation scheme, providing the specific power output values ​​for each generator unit during its operating period. These parameters are transmitted to each actuator through the ship's control system's communication interface.

[0079] In some embodiments, the ship propulsion system optimization scheduling method based on genetic algorithms may further include the following steps:

[0080] Establish a digital twin simulation model of the ship's power system, and input the control command set of the ship's power system into the digital twin simulation model to execute virtual operation;

[0081] The system monitors the power fluctuation rate and voltage deviation rate during the virtual operation process in real time. When the power fluctuation rate exceeds the first preset value or the voltage deviation rate exceeds the second preset value, the optimization parameter reset mechanism is triggered.

[0082] The process of optimizing the parameter reset mechanism includes regenerating the initial population individuals, adjusting the calculation parameters for mutation probability, and restarting the genetic algorithm optimization process.

[0083] Specifically, the digital twin simulation model is built based on multibody dynamics and power system simulation technology, including a main engine dynamics model, a generator set dynamic characteristic model, an equivalent circuit model of the energy storage battery, and a propeller hydrodynamic model. Power fluctuation rate is obtained by calculating the standard deviation of power within a 1-second time window, and voltage deviation rate is obtained by monitoring the percentage deviation of the bus voltage from its rated value. When either indicator exceeds a threshold (power fluctuation rate 5%, voltage deviation rate 3%), the optimization parameter reset mechanism expands the initial population size by 20%, increases the basic mutation probability Pb ​​by 50%, and retains the current best individual to directly enter the next generation of the population.

[0084] In some embodiments, the ship propulsion system optimization scheduling method based on genetic algorithms may further include the following steps:

[0085] Establish a multi-ship collaborative optimization database to store characteristic parameters of various typical navigation environments and their corresponding optimal gene coding schemes;

[0086] The matching degree between the current navigation environment and historical cases is calculated by a feature similarity matching algorithm. When the similarity exceeds the third preset value, the gene code of the corresponding historical case is loaded as the initial population seed.

[0087] The historical gene encoding was adapted using a parameter normalization method to make it conform to the current range of ship power system parameters.

[0088] Specifically, the multi-ship collaborative optimization database uses a time-series database to store historical case data. Each case includes environmental characteristic parameters (such as wind speed, wave height, and current speed), ship state parameters (such as displacement and draft), and the corresponding optimal gene code. The feature similarity matching algorithm uses Euclidean distance to calculate the similarity between the current environmental feature vector and historical cases. When the similarity exceeds 0.8, the historical optimal gene code is used as the initial population seed. Parameter normalization processing includes: speed parameter normalization based on the Froude number similarity criterion; power allocation parameter normalization based on the percentage conversion of the unit's rated power; and start-stop decision parameters directly inherit the start-stop modes of the same type of unit in historical schemes.

[0089] A ship propulsion system optimization scheduling system based on genetic algorithms, employing the aforementioned genetic algorithm-based ship propulsion system optimization scheduling method, includes an operating parameter acquisition unit, a gene encoding unit, a comprehensive fitness value calculation unit, a population generation unit, and an instruction set generation unit, referencing... Figure 2The system employs several mechanisms: a parameter acquisition unit to acquire real-time operating parameters of the ship's propulsion system, including main engine speed data, generator output power data, remaining battery power data, and propeller load characteristic data; a gene encoding unit to construct an initial population based on these operating parameters using a genetic algorithm, where each individual in the initial population contains gene codes for speed regulation parameters, generator start-stop decision parameters, and power allocation ratio parameters; a comprehensive fitness value calculation unit to calculate the comprehensive fitness value of each individual based on energy consumption and emission characteristics, where the weight coefficients corresponding to energy consumption characteristics are related to the remaining battery power, and the weight coefficients corresponding to emission characteristics are related to the fluctuation range of propeller load characteristic data; a population generation unit to perform selection, crossover, and mutation operations on the initial population to generate a new population, where the mutation probability adopts an adaptive adjustment strategy based on the number of iterations; and an instruction set generation unit to repeatedly iterate and optimize the population until preset conditions are met, outputting combinations of speed regulation parameters, generator start-stop decision parameters, and power allocation schemes to generate the ship's propulsion system control instruction set.

[0090] The above are all preferred embodiments of this application and are not intended to limit the scope of protection of this application. Therefore, all equivalent changes made in accordance with the structure, shape and principle of this application should be covered within the scope of protection of this application.

Claims

1. A method for optimal scheduling of ship propulsion systems based on genetic algorithms, characterized in that, Includes the following steps: The real-time operating parameters of the ship's power system are obtained, including main engine speed data, generator output power data, remaining battery power data, and propeller load characteristic data. An initial population for a genetic algorithm is constructed based on the aforementioned operating parameters. Each individual in the initial population includes gene encodings for speed regulation parameters, generator start-stop decision parameters, and power allocation ratio parameters. Based on energy consumption characteristic indicators and emission characteristic indicators, the comprehensive fitness value of each individual is calculated, wherein the weight coefficient corresponding to the energy consumption characteristic indicator is related to the remaining power of the energy storage battery, and the weight coefficient corresponding to the emission characteristic indicator is related to the fluctuation range of the load characteristic data of the thruster. Selection, crossover, and mutation operations are performed on the initial population to generate a new population, wherein the mutation probability adopts an adaptive adjustment strategy based on the number of iterations. Repeatedly perform population iteration optimization until the preset conditions are met, output the combination of speed regulation parameters, generator set start-stop decision parameters and power allocation scheme, and generate the ship power system control command set; The steps for constructing the initial population for the genetic algorithm based on the aforementioned operating parameters are as follows: The range of values ​​for the speed adjustment parameters is determined using the following formula: ； Where V_base is the base speed adjustment range, P_load is the real-time load power of the thruster, and P_rated is the rated power of the thruster; The generator set start-stop decision parameters are encoded as binary bits, where 0 represents the shutdown state and 1 represents the running state. The coding range of the power allocation ratio is determined based on the rated power parameters of each generator set, wherein the power allocation ratio must meet the following constraints: ； Where α_i is the power allocation ratio of the i-th generator set, P_min_i is its minimum allowable output power, and P_total is the total load demand.

2. The method for optimal scheduling of a ship's propulsion system based on a genetic algorithm according to claim 1, characterized in that, The steps for calculating the comprehensive fitness value of each individual based on energy consumption and emission characteristics are as follows: The overall fitness value is calculated using the following formula: ； Wherein, I1 is the energy consumption characteristic index, I2 is the emission characteristic index, W1 is the weighting coefficient of the energy consumption characteristic index, and W2 is the weighting coefficient of the emission characteristic index.

3. The method for optimizing and scheduling a ship's propulsion system based on a genetic algorithm according to claim 2, characterized in that, Based on energy consumption and emission characteristics, the calculation of each individual's overall fitness value also includes: The fuel consumption is obtained by multiplying the real-time output power of each generator set by the corresponding fuel consumption coefficient per unit power; the equipment loss is obtained by multiplying the number of start-stop cycles of each generator set by the single start-stop loss coefficient; and the energy consumption characteristic index is obtained by adding the fuel consumption and equipment loss. The total carbon emissions are obtained by multiplying the real-time output power of each generator set by the corresponding carbon emission coefficient per unit power; the noise pollution index is obtained by dividing the measured noise value of the propeller area by the maximum noise limit; and the emission characteristic index is obtained by multiplying the total carbon emissions by the noise pollution index.

4. The method for optimizing and scheduling a ship's propulsion system based on a genetic algorithm according to claim 3, characterized in that, The method further includes: When the remaining power of the energy storage battery is lower than a preset safety threshold, the weighting coefficient of the energy consumption characteristic index is increased. When the load characteristic data of the thruster fluctuates beyond the historical average fluctuation range, the weighting coefficient of the emission characteristic index is increased.

5. The method for optimizing and scheduling a ship's propulsion system based on a genetic algorithm according to claim 1, characterized in that, The steps of performing selection, crossover, and mutation operations on the initial population to generate a new population are as follows: The mutation probability of the current iteration period is calculated using the following formula: ； Where P_m represents the mutation probability, P_b is the basic mutation probability parameter, G is the current iteration number, G_max is the maximum iteration number, σ_F is the standard deviation of the population fitness value, and μ_F is the fitness mean. Based on the mutation probability, mutation operations are performed on individuals in the population, wherein the mutation operations include applying random perturbations to the speed regulation parameters, reversing the state of the generator start-stop decision parameters, and redistributing the power allocation ratio parameters.

6. The method for optimal scheduling of a ship's propulsion system based on a genetic algorithm according to claim 1, characterized in that, The method further includes: A digital twin simulation model of the ship's power system is established, and the control command set of the ship's power system is input into the digital twin simulation model to execute virtual operation; The system monitors the power fluctuation rate and voltage deviation rate during virtual operation in real time. When the power fluctuation rate exceeds the first preset value or the voltage deviation rate exceeds the second preset value, the optimization parameter reset mechanism is triggered. The execution process of the optimization parameter reset mechanism includes regenerating the initial population individuals, adjusting the calculation parameters of the mutation probability, and restarting the genetic algorithm optimization process.

7. The method for optimal scheduling of a ship's propulsion system based on a genetic algorithm according to claim 1, characterized in that, The method further includes: Establish a multi-ship collaborative optimization database to store characteristic parameters of various typical navigation environments and their corresponding optimal gene coding schemes; The matching degree between the current navigation environment and historical cases is calculated by a feature similarity matching algorithm. When the similarity exceeds the third preset value, the gene code of the corresponding historical case is loaded as the initial population seed. The historical gene encoding was adapted using a parameter normalization method to make it conform to the current range of ship power system parameters.

8. A ship propulsion system optimization scheduling system based on genetic algorithm, characterized in that, The system is used to implement the ship propulsion system optimization scheduling method based on genetic algorithm as described in any one of claims 1-7, including: The operating parameter acquisition unit is used to acquire real-time operating parameters of the ship's power system, including main engine speed data, generator output power data, remaining battery power data, and propeller load characteristic data. A gene encoding unit is used to construct an initial population for a genetic algorithm based on the operating parameters. Each individual in the initial population includes gene encodings for speed regulation parameters, generator start-stop decision parameters, and power allocation ratio parameters. The comprehensive fitness value calculation unit is used to calculate the comprehensive fitness value of each individual based on energy consumption characteristic indicators and emission characteristic indicators. The weight coefficient corresponding to the energy consumption characteristic indicators is related to the remaining power of the energy storage battery, and the weight coefficient corresponding to the emission characteristic indicators is related to the fluctuation range of the load characteristic data of the thruster. A population generation unit is used to perform selection, crossover, and mutation operations on the initial population to generate a new population, wherein the mutation probability adopts an adaptive adjustment strategy based on the number of iterations. The instruction set generation unit is used to repeatedly perform population iteration optimization until preset conditions are met, outputting combinations of speed regulation parameters, generator set start-stop decision parameters, and power allocation schemes to generate a ship power system control instruction set.

Images