CN115906619A - PSO-BP optimization algorithm-based energy management method for hybrid double electric ships - Google Patents
PSO-BP optimization algorithm-based energy management method for hybrid double electric ships Download PDFInfo
- Publication number
- CN115906619A CN115906619A CN202211373290.5A CN202211373290A CN115906619A CN 115906619 A CN115906619 A CN 115906619A CN 202211373290 A CN202211373290 A CN 202211373290A CN 115906619 A CN115906619 A CN 115906619A
- Authority
- CN
- China
- Prior art keywords
- data
- ship
- optimal
- pso
- optimization algorithm
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Pending
Links
Images
Landscapes
- Feedback Control In General (AREA)
Abstract
The invention discloses a PSO-BP optimization algorithm-based energy management method for a hybrid double-electric-power ship, which comprises the following steps of: s1: modeling the hybrid power double electric ships to obtain basic parameters of the ships; s2: initializing BP nerves, designing a neural network, and determining the topological structure of the BP network; s3: determining an adaptive function, initializing the position and the speed of the particles by the initial normalized data in the S1 through a particle swarm optimization algorithm, and continuously updating and changing the fitness; error calculation is carried out by using a BP network, data are continuously updated, data accuracy is improved, and optimal data are finally obtained; s4: and substituting the processed data into the model for simulation analysis. The invention can prolong the service life of the battery, greatly improves the economic performance, has obvious advantages in energy saving and utilization, is cleaner in energy discharge, can respond immediately under variable working conditions, and can deal with more freely in the face of complex working conditions.
Description
Technical Field
The invention relates to the field of ship energy management, in particular to a hybrid power double-electric-ship energy management method based on a PSO-BP optimization algorithm.
Background
In recent years, international organizations have attracted more and more attention to the pollution problem caused by ship navigation, and the development of new energy ships is urgent. At present, novel clean energy sources such as solar energy, wind energy, LNG (liquefied natural gas), fuel cells and the like have substantial application in the field of ships, environmental protection and efficient sustainable utilization of energy are remarkably improved, and the development demand of new energy sources is higher and higher under the large environmental trend of high efficiency, energy conservation and environmental protection for future development. Solar energy in new energy is greatly influenced by weather and cannot be used as a stable propulsion system of a ship, wind energy is mainly applied to sailing ships, however, the influence is too large and is not stable enough under the working conditions of water supply and sewerage, LNG is used as a clean stable power source, and the safety of LNG still needs to be further improved. Fuel cells have been used for most new energy ships as the best of them.
Chinese patent application CN113705094A discloses a ship fuel oil pipeline fault prediction method based on PSOGRU, which comprises the steps of processing ship information, dividing a training set and a test set after normalization processing, constructing a GRU model, and optimizing the GRU model through a particle swarm optimization algorithm so as to obtain an optimal value. However, a single particle swarm optimization algorithm is not an optimal choice for processing a model, and the problem of local convergence is the common fault of the algorithm.
At present, a hybrid fuel cell, a storage battery and a supercapacitor composite propulsion dual electric ship are applied, but inherent problems of prematurity and local convergence existing in a particle swarm optimization algorithm are not solved, especially on the basis of describing a basic structure of a BP (back propagation) neural network, the optimal weight and threshold obtained by the particle swarm optimization algorithm are applied to the BP neural network for training by combining the advantages of the BP neural network and the particle swarm optimization algorithm, so that the model accuracy is improved, the energy distribution of a ship power source is optimized, and finally the use efficiency of the ship is improved, and no literature report is found.
Disclosure of Invention
The invention aims to provide a ship energy management scheme based on PSO-BP, which can effectively solve the problems of energy distribution, power distribution optimization and energy management of the current ship, carry out simulation, provide practical data for actual operation and improve the use efficiency of the ship.
In order to achieve the purpose, the ship energy management scheme based on the PSO-BP comprises the steps of constructing a ship Simulink simulation model and optimizing the energy distribution of a power source.
The specific technical scheme is as follows:
a simulation model is built for the hybrid electric double-ship, an efficient and practicable ship management method is selected under the condition of being based on a PSO-BP optimization algorithm, parameters of an energy management strategy are mainly combined with parameter optimization of a composite power supply, and combined optimization is carried out. The method comprises the following steps:
s1: the system modeling of the hybrid power double electric ship is carried out, and the system modeling specifically comprises a fuel cell module, a storage battery module, a super capacitor module and a bidirectional DC/DC converter module; and acquiring basic parameters of the ship, including the capacities of a storage battery, a super capacitor and a fuel cell, the running cost of the ship and other limiting conditions, and normalizing the data.
S2: initializing BP nerves, designing a neural network, determining the topological structure of the BP network, and determining the number of nodes in a hidden layer.
S3: and (3) determining an adaptive function according to the initial normalized data in the S1, initializing the position and the speed of the particle through a PSO algorithm, continuously updating and changing the fitness, and obtaining an individual optimal extreme value and a group optimal extreme value of the particle in each iteration, so that the individual optimal extreme value and the group optimal extreme value are transmitted to the optimal weight and the threshold of the neural network. And performing error calculation by using a BP network, continuously updating data, improving data accuracy and finally obtaining optimal data.
S4: and substituting the processed data into a model for simulation analysis, wherein the simulation analysis mainly comprises power comparison curves of a battery, a super capacitor and a fuel cell under the premise of different energy control, front-back comparison of ship operation cost, direct-current bus voltage comparison and the like, and data basis is provided for actual ship operation.
Preferably, the Simulink is adopted to build a hybrid ship model. The system comprises a fuel cell module, a storage battery module, a super capacitor module and a bidirectional DC/DC converter module. Obtaining initialization parameters, wherein the initialization parameters comprise basic parameters of a ship, running cost of the ship and the like. The basic parameters of the ship comprise the capacities of a storage battery, a super capacitor and a fuel cell, and the limiting conditions are as follows:
P bat +P sc +P fc ＝P N
C＝N batref ·M bat +N scref ·M sc +P HESS ·M dcdc
wherein P is bat Is the output power of the battery, P sc Is the output power, P, of the super-capacitor fc Is the output power of the fuel cell.
Maximum value of super capacitance, i bat (t) is the battery current output, i sc (t) is the super-capacitor current output, i fc (t) fuel cell current output, C is ship operating cost, N batref Setting the number of batteries, M bat As the battery price, N scref Number of super capacitors, M sc Unit price P of super capacitor HESS For the power requirements of the hybrid power supply, M dcdc Is the price of the dc converter.
And normalizing the obtained data. The formula is as follows:
preferably, a BP nerve is initialized, and a topological structure of a BP network is determined, specifically, when a neural network is designed, the number of nodes of an input layer and an output layer is generally fixed, and a hidden layer can be freely designated; the topology and the arrows in the neural network structure diagram represent the flow direction of data in the prediction process, and certain difference exists between the topology and the arrows and the data flow in the training process; the BP neural network comprises circles (representing "neurons") and connecting lines (representing connections between "neurons"). Each connecting line obtained through training corresponds to a different weight (the value of which is called a weight).
Determining the topological structure of the BP network, and determining the number of nodes in a hidden layer, wherein the formula is as follows:
wherein n is the number of hidden layer nodes; i is the number of input layer nodes (input layer variables); o is the number of output layer nodes (output layer variables); a is a constant between 0 and 10.
Preferably, the optimal weight and threshold are obtained through a particle swarm algorithm, a BP neural network is introduced, errors are calculated, an optimal solution is obtained, specifically, initial normalized data in S1 are initialized for the position and the speed of the particle through a PSO algorithm, fitness is continuously updated and changed, the individual optimal extreme value and the group optimal extreme value of the particle in each iteration are obtained, and therefore the individual optimal extreme value and the group optimal extreme value are transmitted to the neural network optimal weight and threshold.
The iterative formula for particle velocity and position is as follows:
the expression of the weight is as follows, the optimization speed can be improved by selecting a larger value in the early optimization, and the optimization accuracy can be improved by selecting a smaller value in the later optimization.
In the formula, ω max 、ω min The maximum and minimum values of the weights. k denotes the current number of iterations, k max Indicating the maximum number of iterations.
And (4) performing error calculation by using the BP network, continuously updating data, and finally obtaining optimal data. The update formula that satisfies the accuracy is as follows:
w ij ＝w ij +ΔW ij
wherein eta is learning rate, 0< eta <1.
Preferably, the processed data is substituted into the model for simulation analysis, specifically, the data is provided for actual ship operation for power comparison curves of the battery, the super capacitor and the fuel cell, comparison before and after ship operation cost, comparison of direct current bus voltage and the like on the premise of different energy control.
The invention adopts PSO-BP optimization algorithm to optimize the hybrid power double electric ship energy management method, and has three advantages compared with the prior art:
1. compared with the PSO-BP optimization algorithm, when the ship which manages the energy by adopting the PSO-BP optimization algorithm runs, the energy distribution of the power system is particularly important under the condition of complex working condition, the fuel cell is more suitable for stabilizing the power, and when the working condition needs to be changed emergently, the composite power supply is mainly used for bearing, the response speed is higher, and the sudden sailing condition can be better handled.
2. For the ship which does not adopt the PSO-BP optimization algorithm, the energy proportioning and distribution are not carried out, so that the loss of various batteries is increased, and the economy of the ship is not facilitated. The ship adopting the PSO-BP optimization algorithm to manage energy can prolong the service life of the battery, greatly improve the economic performance and have remarkable advantages in energy conservation and utilization.
3. The hybrid double-electric ship for managing energy by using the PSO-BP optimization algorithm is cleaner in energy discharge, can respond immediately under variable working conditions, and can be more freely treated in the face of complex working conditions. The invention adopts Simulink simulation, can save the ship operation cost and is convenient to operate, and the obtained optimal result provides powerful support for actual operation.
Drawings
FIG. 1 is a detailed flow chart of the present invention;
FIG. 2 is a diagram illustrating an exemplary hybrid modeling of a fuel cell hybrid vessel according to the present invention;
FIG. 3 is a Matlab/Simulink diagram of a hybrid modeling of a fuel cell hybrid vessel according to the present invention;
FIG. 4 is a diagram of the PSO-BP optimization algorithm operation steps;
FIG. 5 is a flow chart of an example PSO-BP optimization algorithm.
Detailed Description
In order to make the objects, technical solutions and advantages of the present invention clearer, the following describes embodiments of the present invention in further detail with reference to the accompanying drawings, and the technical solutions in the embodiments of the present invention are clearly and completely described. Based on the embodiments of the present invention, including models and algorithms applied in the field, the present invention shall fall into the protection scope.
Embodiment 1, referring to fig. 1 to 5, a method for energy management of a hybrid twin-electric ship based on a PSO-BP optimization algorithm includes the following steps:
the method comprises the following steps: the modeling of the hybrid double electric ships comprises the following specific steps: modeling the fuel cell, including analyzing power and operating curves of the fuel cell; performing composite power supply modeling of a storage battery and a super capacitor, and constructing a composite power supply module with stable output; and constructing a DC/DC converter model for optimization management. The schematic diagram of the model is shown in fig. 2.
Step two: obtaining initialization parameters, wherein the initialization parameters comprise ship basic parameters, ship operation cost and the like. The basic parameters of the ship comprise the capacities of a storage battery, a super capacitor and a fuel cell, and the parameter limiting conditions are as follows:
P bat +P sc +P fc ＝P N
C＝N batref ·M bat +N scref ·M sc +P HESS ·M dcdc
wherein P is bat Is the output power of the battery, P sc Is the output power of the super capacitor, P fc Is the output power of the fuel cell.
Maximum value of super capacitance, i bat (t) is the battery current output, i sc (t) is the super-capacitor current output, i fc (t) Fuel cell Current output, N batref Setting the number of batteries, M bat To the battery price, N scref Is the number of super capacitors, M sc Unit price P of super capacitor HESS For the power requirements of the hybrid power supply, M dcdc Is the price of the dc converter.
And carrying out normalization processing on the data of the ship parameters. The formula is as follows:
step three: initializing BP neural data, and determining the topological structure of a BP network firstly, wherein the specific operations are as follows:
when the neural network is initialized and set, the number of nodes of the input layer and the output layer is fixed, and the number of nodes of the hidden layer is determined by the number of the optimized parameters; an arrow in the topology in the neural network structure diagram mainly represents the data output direction and has certain difference with the data flow during training; circles in the structure diagram represent "neurons", and connecting lines represent connections between "neurons". Each connecting line corresponds to a different weight, called weight.
Determining the topological structure of the BP network, and determining the number of nodes in a hidden layer, wherein the formula is as follows:
wherein n is the number of hidden layer nodes; i is the number of input layer nodes (input layer variables); o is the number of output layer nodes (output layer variable); a is a constant between 0 and 10.
Step four: a PSO-BP optimization algorithm is applied.
First, a population of particles (population size N) is initialized, including random positions and velocities, using the PSO algorithm. The fitness function of the ship running cost, the service life of the storage battery, the electric energy quality and the energy loss of the hybrid power system is as follows:
minJ＝aJ bat +b(J DC +J E )+cC
wherein J bat Evaluation of fitness function for Battery loss, J DC And J E The energy loss function of the hybrid power system is the energy loss function of the direct current power grid, and C is the ship running cost.
Then, the fitness is calculated, and the fitness of each particle is evaluated. For each particle, its current adaptation value is compared to its individual historical optimum position (P) best ) And comparing the adaptive value with the corresponding adaptive value to enable the particles with higher adaptability to be the optimal positions. For each particle, its current adaptation value is associated with the global optimum position (g) best ) Comparing with corresponding adaptive value, if the current adaptive value is higher, updating global optimum position with the current particle position, and continuously and iteratively updating P best ，g best . The limiting function is as follows:
P bat +P sc +P fc ＝P N
wherein P is bat Is the output power of the battery, P sc Is the output power of the super capacitor, P fc Is the output power of the fuel cell.
Maximum value of super capacitance, i bat (t) is the battery current output, i sc (t) is the super-capacitor current output, i fc (t) fuel cell current output.
Updated parameter P best 、g best And inputting the weight and the threshold value as the optimal values into a BP neural network, and entering the BP neural network to calculate errors to obtain an optimal solution. The method comprises the following specific steps: and initializing the position and the speed of the particles by the normalized data through a PSO algorithm, continuously updating and changing the fitness, and obtaining the individual optimal extreme value and the group optimal extreme value of each iterative particle, so as to transmit the individual optimal extreme value and the group optimal extreme value to the neural network optimal weight and the threshold value for optimal calculation. As shown in detail in fig. 4.
The iterative formula for particle velocity and position is as follows:
the expression of the weight is as follows, the optimization speed can be improved by selecting a larger value in the early optimization, and the optimization accuracy can be improved by selecting a smaller value in the later optimization.
In the formula, ω max 、ω min Is the maximum and minimum value of the weight; k denotes the current number of iterations, k max Indicating the maximum number of iterations.
According to the fitness function:
minJ＝aJ bat +b(J DC +J E )+cC
and limiting conditions, and finally obtaining the weight ratio a, b and c of 0.5,0.3 and 0.2.
And (4) performing error calculation by using the BP network, continuously updating data, and finally obtaining optimal data. The update formula that satisfies the accuracy is as follows:
w ij ＝w ij +ΔW ij
wherein eta is learning rate, 0< eta <1.
In the case of the example 2, the following examples are given,
and performing instantiation analysis through the specific shipping condition of the experimental ship, mainly analyzing the specific data of the equipped super capacitor and battery, optimizing the proportion by using a PSO-BP algorithm, and instantiating to obtain the optimal data. The optimization parameters mainly comprise super capacitor voltage, storage battery capacity and a low-pass filter constant T, and finally, iteration is carried out for multiple times to obtain an optimal data result. The method comprises the following specific steps:
the actual operating implementation parameters were simulated by simulating Matlab/Simulink as shown in table 1 below.
TABLE 1 Equipment parameters
| Parameter(s) | Super capacitor | Storage battery |
| Rated voltage | 48V | 3.6V |
| Operating voltage | 0-51V | 2.7-4V |
| Capacity of | 165F | 10Ah |
| Maximum current | 1900A | 20A |
The energy distribution problem of the composite energy storage system is controlled through a low-pass filter, the low-pass filter only has T variable parameters, and the transfer function expression of the low-pass filter in Simulink is as follows:
t is a time constant and the power of the composite energy storage system can be performed by adjusting the constant T. The specific value of T is determined according to the quantity and the bearing capacity of the storage battery and the super capacitor. Along with the increase of the T value, the passed high-frequency component is reduced and further distributed to the storage battery for bearing, and the high-frequency part is borne by the super capacitor with higher response speed.
The PSO-BP algorithm optimization is carried out by taking the capacity of the storage battery, the voltage of the super capacitor and the filter parameter time T as parameters, and the specific process is shown in FIG. 5. The intervals of the optimized parameters are shown in table 2 below.
TABLE 2 optimization intervals
| Optimizing parameters | Optimized interval |
| Super capacitor voltage | 384V-750V |
| Capacity of accumulator | 10Ah-60Ah |
| Low pass filter constant T | 1-10s |
And setting the iteration times as 100 times, and circularly iterating and optimizing to obtain the optimal battery capacity, the optimal super-capacitor voltage and the optimal time constant of the low-pass filter. The optimized data obtained from the final simulation according to the procedure of example 1 are shown in table 3 below.
TABLE 3 optimization results
| Optimizing parameters | Optimizing results |
| Super capacitor voltage | 420.85V |
| Capacity of accumulator | 40.25Ah |
| Low pass filter constant T | 3.22s |
And selecting the storage battery and the super capacitor which meet the specifications in the table 1 according to the optimized result in the table 3, and performing series-parallel connection to form an optimal grouping result meeting the table 3, so that the composite power supply ratio is obtained, and the optimal ship energy management strategy is obtained.
Claims (8)
1. A PSO-BP optimization algorithm-based energy management method for a hybrid electric double-ship is characterized by comprising the following steps:
s1: modeling the hybrid power double electric ships, and specifically comprising a fuel cell, a storage battery, a super capacitor and a bidirectional DC/DC converter module; acquiring basic parameters of a ship, wherein the basic parameters of the ship comprise parameters of a storage battery, a super capacitor and a fuel cell, and the running cost of the ship; carrying out normalization processing on the basic parameter data of the ship;
s2: initializing BP nerves, designing a neural network, determining the topological structure of the BP network, and determining the number of nodes in a hidden layer;
s3: determining an adaptive function, initializing the position and the speed of the particles by the initial normalized data in the S1 through a particle swarm optimization algorithm, continuously updating and changing the fitness, and obtaining an individual optimal extreme value and a group optimal extreme value of each iterative particle so as to transmit to the optimal weight and the threshold of the neural network; error calculation is carried out by using a BP network, data are continuously updated, data accuracy is improved, and optimal data are finally obtained;
s4: substituting the processed data into a model for simulation analysis, wherein the simulation analysis comprises power comparison curves of a battery, a super capacitor and a fuel cell under the premise of different energy control, front-back comparison of ship operation cost and direct-current bus voltage comparison; providing data for actual ship operation.
2. The PSO-BP optimization algorithm-based energy management method for a hybrid twin-electric ship according to claim 1, characterized in that:
and (3) establishing a simulation model for the hybrid electric double-ship, and obtaining the efficient and practicable ship management method under the condition of being based on a PSO-BP optimization algorithm.
3. The PSO-BP optimization algorithm-based energy management method for a hybrid twin-electric ship according to claim 1, characterized in that:
the simulation model of the hybrid double-electric ship is built by adopting Simulink components.
4. The PSO-BP optimization algorithm-based energy management method for a hybrid twin-electric ship according to claim 1, characterized in that:
in the particle swarm optimization algorithm, the data acquisition method of the optimal weight and the threshold value is as follows:
taking the ship operation cost, the service life of a storage battery, the quality of electric energy and the energy loss of a hybrid power system as standards for judging an optimization function, and combining the parameters;
performing parameter optimization by using a particle swarm optimization algorithm, wherein in the particle swarm optimization algorithm, each particle is described by using a position vector and a velocity vector, the position vector represents a possible solution of the problem, and the velocity vector represents the direction and the size of position change; judging the individual optimal position and the group optimal position at the time t;
and updating the speed position to update the fitness of the particles, and finally obtaining new updating data of the optimal individual and group positions.
5. The hybrid dual-electric-vessel-based particle swarm optimization algorithm according to claim 4, wherein: the calculation method of the data of the optimal weight and the threshold value is as follows:
a. firstly, initializing a particle group, wherein the size of the group is N, and the group comprises random positions and speeds; the fitness function of the ship running cost, the service life of the storage battery, the electric energy quality and the energy loss of the hybrid power system is as follows:
minJ＝aJ bat +b(J DC +J E )+cC
in the formula J bat Fitness function for battery loss evaluation, J DC And J E Respectively an electric energy quality function of a direct current power grid and an energy loss function of a hybrid power system, wherein C is the ship operation cost;
b. secondly, calculating the fitness and evaluating the fitness of each particle; for each particle, comparing the current adaptive value with the individual historical optimal position Pbest and the corresponding adaptive value, and enabling the particle with high fitness to become the optimal position;
c. for each particle, comparing the current adaptive value with the global optimal position gbest and the corresponding adaptive value, and updating the global optimal position by adopting the position of the particle with a high adaptive value;
d. continuously and iteratively updating Pbest and gbest;
e. the limiting function is as follows:
P bat +P sc +P fc ＝P N
in the formula P bat Is the output power of the battery, P sc Is the output power of the super capacitor, P fc Is the output power of the fuel cell;
maximum value of super capacitance, i bat (t) is the battery current output, i sc (t) is the super-capacitor current output, i fc (t) fuel cell current output.
6. The PSO-BP optimization algorithm-based energy management method for a hybrid twin-electric ship according to claim 1, characterized in that:
the initializing BP nerve and designing a neural network are methods for taking particle swarm optimization results as processing data of the neural network, and the methods are as follows:
a particle swarm algorithm is adopted to optimize a BP neural network, the basic BP neural network optimization algorithm is mainly a typical three-layer neural network and comprises an input layer, a hidden layer and an output layer, and the method comprises the following specific steps: firstly, initializing a network, secondly, calculating a hidden layer, then, calculating the input and output of an output layer, calculating errors, and adjusting specific weight and threshold value according to a mode of error gradient reduction.
7. The PSO-BP optimization algorithm-based energy management method for a hybrid twin-electric ship according to claim 6, characterized in that:
the method for taking the particle swarm optimization result as the processing data of the neural network mainly comprises the following steps:
a. determining the topological structure of the BP network, and determining the number of nodes in a hidden layer, wherein the formula is as follows:
wherein n is the number of hidden layer nodes; i is the number of nodes of an input layer, namely an input layer variable; o is the number of nodes of the output layer, namely the variable of the output layer; a is a constant between 0 and 10;
b. initializing a weight and a threshold of the BP neural network, obtaining an optimal weight and threshold through a PSO algorithm, and training the neural network to finally enable an error to meet the precision;
c. selecting the capacities of a storage battery, a super capacitor and a fuel cell, and carrying out normalization processing on data; the formula is as follows:
d. initializing the position and the speed of the particle through a PSO algorithm, continuously updating and changing the fitness, and obtaining an individual optimal extreme value and a group optimal extreme value of each iterative particle so as to transmit to the optimal weight and the threshold of the neural network; the iterative formula for particle velocity and position is as follows:
the expression of the weight is as follows, the optimization speed can be improved by selecting a larger value in the early optimization, and the optimization accuracy can be improved by selecting a smaller value in the later optimization;
e. error calculation is carried out by using a BP network, data are continuously updated, and optimal data are finally obtained; the update formula that satisfies the accuracy is as follows:
w ij ＝w ij +ΔW ij
8. the PSO-BP optimization algorithm-based energy management method for a hybrid twin-electric ship according to claim 1, characterized in that:
analyzing specific data of the battery and the super capacitor, and performing optimized proportioning by using a PSO-BP algorithm to obtain optimal data; the optimization parameters mainly comprise super capacitor voltage, storage battery capacity and a low-pass filter constant T.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202211373290.5A CN115906619A (en) | 2022-11-04 | 2022-11-04 | PSO-BP optimization algorithm-based energy management method for hybrid double electric ships |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202211373290.5A CN115906619A (en) | 2022-11-04 | 2022-11-04 | PSO-BP optimization algorithm-based energy management method for hybrid double electric ships |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN115906619A true CN115906619A (en) | 2023-04-04 |
Family
ID=86482730
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202211373290.5A Pending CN115906619A (en) | 2022-11-04 | 2022-11-04 | PSO-BP optimization algorithm-based energy management method for hybrid double electric ships |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN115906619A (en) |
Cited By (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN117048802A (en) * | 2023-04-26 | 2023-11-14 | 哈尔滨工业大学（威海） | Ship future motion attitude prediction method and system based on real sea state strong adaptation |
| CN117879128A (en) * | 2024-03-13 | 2024-04-12 | 西北工业大学宁波研究院 | Energy management system and management strategy of large container ship composite energy system |
-
2022
- 2022-11-04 CN CN202211373290.5A patent/CN115906619A/en active Pending
Cited By (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN117048802A (en) * | 2023-04-26 | 2023-11-14 | 哈尔滨工业大学（威海） | Ship future motion attitude prediction method and system based on real sea state strong adaptation |
| CN117879128A (en) * | 2024-03-13 | 2024-04-12 | 西北工业大学宁波研究院 | Energy management system and management strategy of large container ship composite energy system |
| CN117879128B (en) * | 2024-03-13 | 2024-05-10 | 西北工业大学宁波研究院 | Energy management system and management strategy of large container ship composite energy system |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| Xiong et al. | Towards a smarter hybrid energy storage system based on battery and ultracapacitor-A critical review on topology and energy management | |
| CN115906619A (en) | PSO-BP optimization algorithm-based energy management method for hybrid double electric ships | |
| CN110620402B (en) | Distributed planning operation joint optimization method and system for electricity-gas hybrid system | |
| CN108471143A (en) | Micro-grid multi-energy method for optimizing scheduling based on positive and negative feedback particle cluster algorithm | |
| WO2023060815A1 (en) | Energy storage capacity optimization configuration method for improving reliability of power distribution network | |
| CN111709850A (en) | New energy ship power system capacity optimization method considering ship rolling | |
| CN112668240A (en) | Virtual energy storage based economic dispatching method and system for all-electric ship | |
| CN113675890A (en) | TD 3-based new energy microgrid optimization method | |
| CN111891111A (en) | Hybrid electric vehicle interval II type fuzzy logic self-adaptive control method based on MCPSO | |
| CN111948562A (en) | Full life cycle monitoring and evaluating system for fuel cell | |
| Fan et al. | Development trend and hotspot analysis of ship energy management | |
| Liu et al. | Wind power short-term forecasting based on LSTM neural network with dragonfly algorithm | |
| Huangfu et al. | An optimal energy management strategy with subsection bi-objective optimization dynamic programming for photovoltaic/battery/hydrogen hybrid energy system | |
| CN108306346A (en) | A kind of distribution network var compensation power-economizing method | |
| CN114784831A (en) | Active power distribution network multi-objective reactive power optimization method based on mobile energy storage | |
| CN114928110A (en) | Three-layer robust optimization method for planning, scheduling and collaborative optimization based on N-K faults | |
| CN114444850A (en) | Park energy storage planning method, device, equipment and medium based on digital twin | |
| Li et al. | Optimization design of hybrid energy storage capacity configuration for electric ship | |
| CN115099522B (en) | Active and reactive carbon emission prediction method for special transformer user based on BP neural network | |
| CN113673141B (en) | Energy router modeling and optimization control method based on data driving | |
| CN116345502A (en) | Method, system, equipment and medium for capacity optimization configuration of hybrid energy storage double-layer model | |
| CN116050461A (en) | Improved method for determining membership function of fuzzy control theory by using convolutional neural network | |
| Zhou et al. | Voltage regulation based on deep reinforcement learning algorithm in distribution network with energy storage system | |
| CN112668764B (en) | Offshore wind farm energy storage system optimization configuration method based on cloud model and FCM algorithm | |
| CN112182945A (en) | Test microgrid construction and control method capable of being used for verification of various microgrid structures |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination |