Wireless power transmission system optimization method, device and medium based on machine learning
By using machine learning prediction models in the radio energy transmission system to determine whether the particles are Pareto frontier particles, the problem of slow generation rate is solved, and design efficiency and system optimization speed are improved.
Patent Information
- Application Number
- CN202311532938.3
- Authority / Receiving Office
- CN · China
- Patent Type
- Patents(China)
- Current Assignee / Owner
- Filing Date
- 2023-11-16
- Publication Date
- 2024-11-05
- Estimated Expiration
- 2043-11-16
AI Technical Summary
The existing radio energy transmission system has a slower rate at generating particles at the Pareto frontier, resulting in a longer design process.
The machine learning-based prediction model is used to predict the randomly generated available particles to determine whether they are particles at the Pareto frontier, thereby accelerating the particle generation rate and optimizing the radio energy transmission system through intelligent algorithms.
The design efficiency of the radio energy transmission system is improved, and faster system optimization is achieved by accelerating the multi-objective design process of the coupling mechanism.
Smart Images

Figure CN117669800B_ABST
Abstract
Description
Technical Field
[0001] The present invention relates to the field of wireless power transmission technology, and in particular to a wireless power transmission system optimization method, device and medium based on machine learning. Background Art
[0002] The coupling mechanism is a key component for the wireless power transmission (WPT) system to achieve energy coupling between the primary and secondary sides, and determines the efficiency, power, and anti-drift characteristics of the WPT system. At the same time, the coupling mechanism should also meet various constraints. In addition to the large number of design goals and constraints, the complexity of the coupling mechanism design is also reflected in the fact that there are many variables that affect the coupling performance.
[0003] The design of coupled coils is a typical complex multivariate problem with multiple constraints and multiple objectives. Multi-objective optimization can be achieved through intelligent algorithms. It is necessary to screen out particles on the Pareto frontier as the optimal solution set by comparing their target values among the particles that meet the constraints for subsequent analysis.
[0004] When optimizing through intelligent algorithms, after initializing the parameters, available particles are randomly generated, and the particles are restricted according to constraints during the optimization process. The particles at the Pareto front are obtained through a multi-objective optimization algorithm for subsequent analysis and decision-making. That is, the coupling mechanism design is the process of selecting Pareto front particles from many particles.
[0005] When multi-objective optimization is performed by randomly generating available particles, the process of generating particles is an iterative process. Since the number of available particles increases sharply with the increase of the dimension and value range of the design variables, the rate of generating particles on the Pareto front is slow, resulting in a long design process for the WPT system. Summary of the invention
[0006] The present invention provides a wireless power transmission system optimization method, device and medium based on machine learning, which are used to solve the defect of slow generation rate of particles at the Pareto front in the prior art and accelerate the generation rate of particles at the Pareto front.
[0007] The present invention provides a wireless power transmission system optimization method based on machine learning, comprising:
[0008] The randomly generated available particles are input into the trained prediction model to obtain the prediction results output by the prediction model; the prediction results include: whether the available particles are particles at the Pareto front; the prediction model is obtained by training based on the training set; the training set includes: ordinary particles and particles at the Pareto front obtained by multi-objective optimization of the coupling mechanism through an intelligent algorithm;
[0009] Based on the prediction results, the wireless power transmission system is optimized.
[0010] In some embodiments, optimizing the wireless power transmission system based on the prediction result includes:
[0011] When it is determined that the available particles are particles on the Pareto front, based on the available particles, an intelligent algorithm is used to optimize the wireless power transmission system;
[0012] The available particles are particles on the Pareto front and are determined based on the prediction result.
[0013] In some embodiments, the optimization of the wireless power transmission system based on the available particles using an intelligent algorithm includes:
[0014] Based on the available particles, the independent variables corresponding to the coupling mechanism, the constraints, the optimization objectives and the value range of the independent variables, an intelligent algorithm is used for optimization to obtain the first Pareto frontier result;
[0015] Based on the first Pareto front result, the wireless power transmission system is optimized.
[0016] In some embodiments, the training set is obtained by:
[0017] Based on the independent variables, constraints, optimization objectives and value ranges of the independent variables corresponding to the coupling mechanism, an intelligent algorithm is used for optimization to obtain the second Pareto frontier result;
[0018] Determine ordinary particles and particles at the Pareto front based on the second Pareto front results corresponding to coupling mechanisms of different power levels;
[0019] The common particles and the particles on the Pareto front are marked to obtain the training set.
[0020] In some embodiments, the prediction model is trained based on a machine learning algorithm.
[0021] The present invention also provides a wireless power transmission system optimization device based on machine learning, comprising:
[0022] An input module is used to input randomly generated available particles into a trained prediction model to obtain a prediction result output by the prediction model; the prediction result includes: whether the available particles are particles at the Pareto front; the prediction model is obtained by training based on a training set; the training set includes: ordinary particles and particles at the Pareto front obtained by multi-objective optimization of the coupling mechanism through an intelligent algorithm;
[0023] The optimization module is used to optimize the wireless power transmission system based on the prediction result.
[0024] In some embodiments, the optimization module is used to:
[0025] When it is determined that the available particles are particles on the Pareto front, based on the available particles, an intelligent algorithm is used to optimize the wireless power transmission system;
[0026] The available particles are particles on the Pareto front and are determined based on the prediction result.
[0027] The present invention also provides an electronic device, comprising a memory, a processor, and a computer program stored in the memory and running on the processor, wherein when the processor executes the program, it implements any of the above-mentioned wireless power transmission system optimization methods based on machine learning.
[0028] The present invention also provides a non-transitory computer-readable storage medium having a computer program stored thereon, which, when executed by a processor, implements any of the above-described wireless power transmission system optimization methods based on machine learning.
[0029] The present invention also provides a computer program product, comprising a computer program, which, when executed by a processor, implements any of the above-described wireless power transmission system optimization methods based on machine learning.
[0030] The wireless power transmission system optimization method, device and medium based on machine learning provided by the present invention predict the randomly generated available particles through a prediction model, determine whether they are on the Pareto front, accelerate the generation rate of particles on the Pareto front, and then accelerate the multi-objective design process of the coupling mechanism, thereby optimizing the wireless power transmission system and improving the design efficiency of the wireless power transmission system. BRIEF DESCRIPTION OF THE DRAWINGS
[0031] In order to more clearly illustrate the technical solutions in the present invention or the prior art, the following briefly introduces the drawings required for use in the embodiments or the description of the prior art. Obviously, the drawings described below are some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative work.
[0032] Figure 1 It is one of the flow charts of the wireless power transmission system optimization method based on machine learning provided in an embodiment of the present invention;
[0033] Figure 2It is a flowchart of a multi-objective optimization method based on a particle swarm algorithm provided in an embodiment of the present invention;
[0034] Figure 3 This is a second flow chart of a wireless power transmission system optimization method based on machine learning provided in an embodiment of the present invention;
[0035] Figure 4 is a cross-check result of a prediction model based on a naive Bayesian machine learning algorithm provided in an embodiment of the present invention;
[0036] Figure 5 It is a schematic diagram of the modular structure of the WPT system provided by an embodiment of the present invention;
[0037] Figure 6 is a structural schematic diagram of a wireless power transmission system optimization device based on machine learning provided in an embodiment of the present invention;
[0038] Figure 7 It is a schematic diagram of the structure of an electronic device provided by an embodiment of the present invention. DETAILED DESCRIPTION
[0039] In order to make the purpose, technical solution and advantages of the present invention clearer, the technical solution of the present invention will be clearly and completely described below in conjunction with the drawings of the present invention. Obviously, the described embodiments are part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.
[0040] The terms "first", "second", etc. in the specification and claims of the present invention are used to distinguish similar objects, and are not used to describe a specific order or sequence. It should be understood that the terms used in this way are interchangeable under appropriate circumstances, so that the embodiments of the present invention can be implemented in an order other than those illustrated or described herein, and the objects distinguished by "first" and "second" are generally of the same type, and the number of objects is not limited. For example, the first object can be one or more. In addition, "and / or" in the specification and claims means at least one of the connected objects, and the character " / " generally indicates that the objects associated with each other are in an "or" relationship.
[0041] Figure 1 is one of the flow charts of the wireless power transmission system optimization method based on machine learning provided by an embodiment of the present invention, such as Figure 1 As shown, the wireless power transmission system optimization method based on machine learning provided by the embodiment of the present invention includes:
[0042] Step 101: input the randomly generated available particles into the trained prediction model to obtain the prediction results output by the prediction model; the prediction results include: whether the available particles are particles at the Pareto front; the prediction model is obtained by training based on a training set; the training set includes: ordinary particles and particles at the Pareto front obtained by multi-objective optimization of the coupling mechanism through an intelligent algorithm;
[0043] Step 102: Optimize the wireless power transmission system based on the prediction result.
[0044] It should be noted that the execution subject of the wireless power transmission system optimization method based on machine learning provided by the present invention can be an electronic device, a component in an electronic device, an integrated circuit, or a chip. The electronic device can be a mobile electronic device or a non-mobile electronic device. Exemplarily, the mobile electronic device can be a mobile phone, a tablet computer, a laptop computer, a PDA, an in-vehicle electronic device, a wearable device, an ultra-mobile personal computer (UMPC), a netbook or a personal digital assistant (PDA), etc., and the non-mobile electronic device can be a server, a network attached storage (NAS), a personal computer (PC), a television (TV), a teller machine or a self-service machine, etc., which is not specifically limited by the present invention.
[0045] In step 101, randomly generated available particles are input into a trained prediction model to obtain a prediction result output by the prediction model; the prediction result includes: whether the available particles are particles at the Pareto front; the prediction model is obtained based on training of a training set; the training set includes: ordinary particles and particles at the Pareto front obtained by multi-objective optimization of a coupling mechanism through an intelligent algorithm.
[0046] Coupling mechanism design is the process of selecting Pareto frontier particles from many particles, which can correspond to the classification problem in machine learning. Particles with excellent comprehensive performance of coupling mechanisms have common characteristics.
[0047] Machine learning is an important discipline in the field of artificial intelligence. It uses computers to simulate human learning activities, establishes prediction models by learning the characteristics and rules of existing data sets, and measures the characteristics and metrics of unknown data.
[0048] Optionally, by identifying whether particles of different power levels are on the Pareto front to obtain a training set, a prediction model is established using a machine learning algorithm to determine whether a new particle is on the Pareto front, thereby accelerating the design process.
[0049] The randomly generated available particles are input into the trained prediction model to obtain the prediction result output by the prediction model, which is whether the available particles are particles on the Pareto front.
[0050] In some embodiments, the training set is obtained by:
[0051] Based on the independent variables, constraints, optimization objectives and value ranges of the independent variables corresponding to the coupling mechanism, an intelligent algorithm is used for optimization to obtain the second Pareto frontier result;
[0052] Determine ordinary particles and particles at the Pareto front based on the second Pareto front results corresponding to coupling mechanisms of different power levels;
[0053] The common particles and the particles on the Pareto front are marked to obtain the training set.
[0054] The independent variables corresponding to the coupling mechanism are parameters that may affect the performance of the coupling coil.
[0055] The premise for the stable and safe operation of the WPT system is to meet the limitations of the thermal management system and insulation capacity. Therefore, parameters such as system efficiency, current density of the coupling coil, loss density of the magnetic core, and terminal voltage of the resonant element are limited as constraints.
[0056] The optimization objectives include transmission efficiency η, weight power density G den , the area power density A at the receiving end den , Efficiency η in the offset state mis , tolerance ψ, etc.
[0057] According to the independent variables, constraints, optimization objectives and value ranges of the independent variables corresponding to the coupling mechanism, an intelligent algorithm can be used to optimize the independent variables to obtain the second Pareto frontier results. The intelligent algorithm can include ant colony algorithm, particle swarm algorithm and genetic algorithm.
[0058] Optionally, in the embodiment of the present invention, a particle swarm algorithm is used as an example. Figure 2 is a flow chart of a multi-objective optimization method based on a particle swarm algorithm provided in an embodiment of the present invention, such as Figure 2 As shown, the multi-objective optimization method based on particle swarm algorithm provided by the embodiment of the present invention includes:
[0059] Parameter initialization;
[0060] Randomly generate available particles;
[0061] Step 1: obtain the self-inductance and mutual inductance of the coupling mechanism through finite element simulation;
[0062] Step 1.1, obtain the self-inductance and mutual inductance of the primary coil;
[0063] Step 1.2, obtain the self-inductance of the secondary coil;
[0064] Step 2: Calculate the efficiency of the secondary rectifier, the current of each link of the circuit, etc.
[0065] Determine particle availability based on current density, capacitance and voltage;
[0066] Step 3: Perform finite element simulation based on the calculated current to obtain core loss, shielding loss, and extract coil magnetic field information;
[0067] Step 4, post-processing: calculate the loss, efficiency, weight, etc. of each part;
[0068] Determine particle availability based on magnetic flux leakage, magnetic core magnetic induction intensity, etc.;
[0069] Step 5: finite element simulation to obtain parameters such as the coupling mechanism at the offset position;
[0070] Step 6, post-processing: calculate the system loss and efficiency during offset;
[0071] Step 7: Filter particles that meet the conditions, update particles on the Pareto front, and select the next generation of available particles based on the particles on the Pareto front, and determine whether the optimization is completed:
[0072] If yes, the optimization is completed;
[0073] If no, repeat steps 1 to 7.
[0074] Optionally, in the embodiment of the present invention, the system parameters include: the primary input DC voltage U in is 510V, the secondary output DC voltage U O It is 510V, the rated air gap d is 200mm, the rated operating frequency is 85kHz, the horizontal lateral offset △X is +75mm, the horizontal longitudinal offset △Y is +100mm, and the vertical offset △d is +75mm.
[0075] In the optimization target of the coupling mechanism, the system weight power density is the ratio of the system output power to the weight of the coupling mechanism. Since the weight of the compensation capacitor and the converter is difficult to evaluate directly, only the weight of the coupling mechanism itself is considered when calculating the weight power density here.
[0076] In practical applications, the secondary side of the coupling mechanism needs to be installed inside a mobile electrical device, where the space is often small. Therefore, this embodiment only focuses on the area power density of the secondary side of the coupling mechanism, that is, the ratio of the output power to the area of the secondary side of the coupling mechanism.
[0077] The tolerance also takes into account the deviation of mutual inductance and primary compensation inductance. The coefficient of variation of mutual inductance is expressed as k. M Indicated by k, the compensation inductance variation coefficient is Lr Indicates that the value is equal to L r / L r0 , L r Represents the primary series compensation inductance, L r0 Indicates the ideal value of the primary series compensation inductor. System tolerance is measured by the extreme values of system characteristics within the range of coupling and compensation parameter deviation. k Lr The variation range of is set to ±5%, and the mutual inductance variation ratio is obtained by simulation.
[0078] The worst case of system characteristics usually occurs when the parameter deviation is the largest. Therefore, in addition to simulating the alignment condition, step 7 also simulates two extreme offset conditions, namely, the conditions where (ΔX, ΔY, Δd) are (0, 0, -30) and (+75, +100, +50) respectively, to obtain the maximum and minimum values of the coupled mutual inductance.
[0079] The optimization objectives of coupling mechanism efficiency, anti-deviation characteristics, etc. have different dimensions and are difficult to measure simply with a comprehensive indicator.
[0080] Therefore, in this embodiment, among the particles that meet the constraint conditions, the particles on the Pareto front are selected by comparing their target values as the optimal solution set for subsequent analysis. During the update iteration, particles with the same number of populations are randomly selected from the Pareto front particles for parameter update.
[0081] In order to obtain a coupling mechanism with good performance in all characteristics and to analyze the factors affecting the performance of the coupling mechanism more accurately and comprehensively, the embodiments of the present invention set the parameters that may affect the performance of the coupling coil as independent variables.
[0082] Taking the DD type coil as an example, the independent variables corresponding to the DD type coil and their value ranges during the optimization process are shown in Table 1.
[0083] Table 1
[0084]
[0085] The subscripts “P” and “S” in Table 1 represent the corresponding variables of the primary and secondary sides of the corresponding coupling mechanism, respectively.
[0086] During the particle generation process, non-continuous variables such as the number of coil turns and the number of parallel windings are discretized. The parameters of the randomly generated particles may be inconsistent. For example, the spacing between the generated particle coil turns may be less than zero. Therefore, before the finite element simulation, each generated particle is first tested for availability and unavailable particles are corrected.
[0087] The premise for the stable and safe operation of the system is to always meet the limitations of the thermal management system and insulation capacity. Therefore, the system efficiency, current density of the coupling coil, loss density of the magnetic core, voltage at the resonant element and other parameters are limited during system design. The determined constraints are shown in Table 2.
[0088] Table 2
[0089]
[0090] The "-mis" indicates the corresponding index of the offset working condition. When the magnetic flux density exceeds a certain value, the magnetic permeability of the core material will drop rapidly, resulting in magnetic failure, thus limiting the maximum magnetic induction intensity of the core.
[0091] In this embodiment, the leakage magnetic induction intensity S in the horizontal and vertical directions is f-h and S f-v The horizontal test point is located on a plane parallel to the coil, passing through the midpoints of the primary and secondary coils, and 1.5m away from the center of the coil. The vertical test point is located on the extension line of the center line of the primary and secondary coils under the alignment condition, and 0.5m away from the aluminum shielding plate of the secondary coil.
[0092] To speed up the simulation process, after obtaining the current of each link in step 2, the particle availability is judged according to the constraints of current density and capacitor voltage. The efficiency, leakage magnetic intensity, loss power density and other data under the alignment conditions obtained in step 4 are used to judge whether the current particle meets the constraints, and unavailable particles are discarded in time.
[0093] According to the independent variables, constraints, optimization objectives and the value range of the independent variables, an intelligent algorithm is used to optimize the independent variables. Among the particles that meet the constraints, the particles on the Pareto front are selected as the optimal solution set by comparing their target values, and the second Pareto front result is obtained.
[0094] Optionally, in order to determine the changes in system characteristics at different power levels and avoid the one-sidedness of the optimization result data distribution and characteristics, multi-objective optimization of systems with various power levels such as 15kW, 30kW, 50kW, 100kW, 150kW and 200kW was carried out based on the types and value ranges of independent variables, constraints and optimization objectives described in the above embodiments.
[0095] Optionally, by performing multi-objective optimization and simulation on the DD-type coupling mechanism of different power levels, there are 95,767 particles in the simulation results of 6 power levels, of which 4,014 particles are at the Pareto front. The particles at the Pareto front of each power level obtained by simulation are marked with ordinary particles, and ordinary particles refer to particles that are not at the Pareto front, so as to obtain a training set.
[0096] Optionally, a prediction model is established using a machine learning algorithm through the obtained training set. By predicting the randomly generated particles through the prediction model, it can be determined whether a new particle is on the Pareto frontier, thereby accelerating the design process.
[0097] In step 102, the wireless power transmission system is optimized based on the prediction result.
[0098] In some embodiments, optimizing the wireless power transmission system based on the prediction result includes:
[0099] When it is determined that the available particles are particles on the Pareto front, based on the available particles, an intelligent algorithm is used to optimize the wireless power transmission system;
[0100] The available particles are particles on the Pareto front and are determined based on the prediction result.
[0101] In some embodiments, the optimization of the wireless power transmission system based on the available particles using an intelligent algorithm includes:
[0102] Based on the available particles, the independent variables corresponding to the coupling mechanism, the constraints, the optimization objectives and the value range of the independent variables, an intelligent algorithm is used for optimization to obtain the first Pareto frontier result;
[0103] Based on the first Pareto front result, the wireless power transmission system is optimized.
[0104] Figure 3 This is a second flow chart of the wireless power transmission system optimization method based on machine learning provided by an embodiment of the present invention, such as Figure 3 As shown, after obtaining the trained prediction model, the randomly generated available particles are input into the prediction model to obtain the prediction results of the prediction model. According to the prediction results, it can be determined whether the available particles are particles on the Pareto frontier.
[0105] When it is determined that the available particle is not a particle on the Pareto front, an available particle is randomly generated again, and the prediction model is continued to be used to determine whether it is a particle on the Pareto front.
[0106] Until it is determined that the available particle is a particle at the Pareto front, the simulation and calculation of steps 1 to 7 in the above method are performed. According to the available particles, the independent variables, constraints, optimization objectives and value ranges of the independent variables corresponding to the coupling mechanism, an intelligent algorithm is used for optimization to obtain the first Pareto front result, obtain the efficiency and other indicators corresponding to the particle, and determine whether the optimization is completed.
[0107] When the optimization is completed, it ends;
[0108] If the optimization is not completed, randomly generate available particles again and repeat the above steps.
[0109] According to the obtained first Pareto front result, the particles on the Pareto front are determined, the design of the coupling mechanism is realized, and the multi-objective design process of the coupling mechanism is accelerated, thereby optimizing the wireless power transmission system and improving the design efficiency of the wireless power transmission system.
[0110] The wireless power transmission system optimization method based on machine learning provided in the embodiment of the present invention predicts the randomly generated available particles through a prediction model, determines whether they are on the Pareto front, accelerates the generation rate of particles on the Pareto front, and further accelerates the multi-objective design process of the coupling mechanism, thereby optimizing the wireless power transmission system and improving the design efficiency of the wireless power transmission system.
[0111] In some embodiments, the prediction model is trained based on a machine learning algorithm.
[0112] The prediction model is trained based on machine learning algorithms, which may include decision trees, clustering, Bayesian classification, support vector machines, EM, Adaboost, etc.
[0113] The embodiment of the present invention takes the Naive Bayesian algorithm as an example for explanation. The Naive Bayesian algorithm is efficient in calculation, highly accurate, and has a solid theoretical basis. The algorithm calculates the conditional probability of the sample to be judged belonging to each category based on the Bayesian theorem, and then judges it as the category with the highest probability.
[0114] In the embodiment of the present invention, the Pareto frontier particles and ordinary particles of each power level obtained by the above simulation are firstly marked, and the data set is trained and learned using the naive Bayes algorithm, and cross-validation is performed. Figure 4 is the cross-validation result of the prediction model based on the naive Bayesian machine learning algorithm provided by the embodiment of the present invention. The obtained validation matrix result is as follows: Figure 4 shown.
[0115] Among the marked values, there are 95767 particles in total, of which 4014 are on the Pareto front and 91753 are ordinary particles.
[0116] In the prediction results, 3428 Pareto frontier particles were accurately judged, and only 8015 ordinary particles were misjudged, with an overall accuracy rate of 91.0%.
[0117] On the other hand, among the particles judged as Pareto front particles by the model, 30% of the particles are true Pareto front particles, which greatly increases the probability of generating Pareto front particles and can accelerate the multi-objective design process of coupled mechanisms.
[0118] Optionally, Figure 3 Although the wireless power transmission system optimization method based on machine learning shown in the figure relies on a certain amount of historical data, the advantages of this method will become more and more significant as the data accumulates because the machine learning algorithm has a huge advantage in determining the accuracy of Pareto frontier particles.
[0119] In with Figure 2 When used in conjunction with the multi-objective optimization method based on the particle swarm algorithm shown, this method will accelerate the generation rate of Pareto frontier particles, which is conducive to the rapid completion of the evaluation and design of systems with different power levels.
[0120] For example, when supplying power to rail transit vehicles, the power level of the WPT system will reach the MW level. Figure 5 The modular structure diagram of the WPT system provided by the embodiment of the present invention is shown in FIG. Figure 5 When designing a modular system, the power level of a single module must first be determined based on constraints such as the installation space. Figure 2 The process shown in the figure is used to optimize the design of several power level systems, and then use Figure 3 The flow shown here will help speed up the overall system design and optimization process by completing more power level system exploration.
[0121] The wireless power transmission system optimization method based on machine learning provided by the embodiment of the present invention can determine whether a new particle is at the Pareto front by establishing a prediction model using a machine learning algorithm, thereby accelerating the design process and greatly increasing the probability of generating Pareto front particles, thereby accelerating the multi-objective design process of the coupling mechanism. The method will accelerate the generation rate of Pareto front particles, thereby improving the efficiency of wireless power transmission system optimization.
[0122] The wireless power transmission system optimization device based on machine learning provided by the present invention is described below. The wireless power transmission system optimization device based on machine learning described below and the wireless power transmission system optimization method based on machine learning described above can be referenced to each other.
[0123] Figure 6is a schematic diagram of the structure of a wireless power transmission system optimization device based on machine learning provided by an embodiment of the present invention, such as Figure 6 As shown, the wireless power transmission system optimization device based on machine learning provided by the embodiment of the present invention includes:
[0124] An input module 610 is used to input the randomly generated available particles into the trained prediction model to obtain the prediction results output by the prediction model; the prediction results include: whether the available particles are particles at the Pareto front; the prediction model is obtained by training based on a training set; the training set includes: ordinary particles and particles at the Pareto front obtained by performing multi-objective optimization on the coupling mechanism through an intelligent algorithm;
[0125] The optimization module 620 is used to optimize the wireless power transmission system based on the prediction result.
[0126] It should be noted here that the wireless power transmission system optimization device based on machine learning provided in the embodiment of the present invention can implement all the method steps implemented in the above-mentioned wireless power transmission system optimization method embodiment based on machine learning, and can achieve the same technical effect. The parts and beneficial effects of this embodiment that are the same as those of the method embodiment will not be described in detail here.
[0127] Optionally, the optimization module 620 is used to:
[0128] When it is determined that the available particles are particles on the Pareto front, based on the available particles, an intelligent algorithm is used to optimize the wireless power transmission system;
[0129] The available particles are particles on the Pareto front and are determined based on the prediction result.
[0130] Optionally, the optimization module 620 is further configured to:
[0131] Based on the available particles, the independent variables corresponding to the coupling mechanism, the constraints, the optimization objectives and the value range of the independent variables, an intelligent algorithm is used for optimization to obtain the first Pareto frontier result;
[0132] Based on the first Pareto front result, the wireless power transmission system is optimized.
[0133] Optionally, the training set is obtained by:
[0134] Based on the independent variables, constraints, optimization objectives and value ranges of the independent variables corresponding to the coupling mechanism, an intelligent algorithm is used for optimization to obtain the second Pareto frontier result;
[0135] Determine ordinary particles and particles at the Pareto front based on the second Pareto front results corresponding to coupling mechanisms of different power levels;
[0136] The common particles and the particles on the Pareto front are marked to obtain the training set.
[0137] Optionally, the prediction model is trained based on a machine learning algorithm.
[0138] Figure 7 An example of a physical structure diagram of an electronic device is shown in FIG. Figure 7 As shown, the electronic device may include: a processor 710, a communication interface 720, a memory 730 and a communication bus 740, wherein the processor 710, the communication interface 720 and the memory 730 communicate with each other through the communication bus 740. The processor 710 may call the logic instructions in the memory 730 to execute a wireless power transmission system optimization method based on machine learning, the method comprising: inputting randomly generated available particles into a trained prediction model to obtain a prediction result output by the prediction model; the prediction result comprises: whether the available particles are particles at the Pareto front; the prediction model is obtained by training based on a training set; the training set comprises: ordinary particles and particles at the Pareto front obtained by multi-objective optimization of the coupling mechanism by an intelligent algorithm; based on the prediction result, the wireless power transmission system is optimized.
[0139] In addition, the logic instructions in the above-mentioned memory 730 can be implemented in the form of a software functional unit and can be stored in a computer-readable storage medium when it is sold or used as an independent product. Based on such an understanding, the technical solution of the present invention, in essence, or the part that contributes to the prior art or the part of the technical solution, can be embodied in the form of a software product, and the computer software product is stored in a storage medium, including a number of instructions for a computer device (which can be a personal computer, a server, or a network device, etc.) to perform all or part of the steps of the method described in each embodiment of the present invention. The aforementioned storage medium includes: U disk, mobile hard disk, read-only memory (ROM, Read-Only Memory), random access memory (RAM, Random Access Memory), disk or optical disk and other media that can store program codes.
[0140] On the other hand, the present invention also provides a computer program product, which includes a computer program, and the computer program can be stored on a non-transitory computer-readable storage medium. When the computer program is executed by a processor, the computer can execute the wireless power transmission system optimization method based on machine learning provided by the above methods, and the method includes: inputting randomly generated available particles into a trained prediction model to obtain a prediction result output by the prediction model; the prediction result includes: whether the available particles are particles on the Pareto front; the prediction model is obtained based on training with a training set; the training set includes: ordinary particles and particles on the Pareto front obtained by performing multi-objective optimization of the coupling mechanism through an intelligent algorithm; based on the prediction result, the wireless power transmission system is optimized.
[0141] On the other hand, the present invention also provides a non-transitory computer-readable storage medium having a computer program stored thereon, which, when executed by a processor, is implemented to execute the wireless power transmission system optimization method based on machine learning provided by the above-mentioned methods, the method comprising: inputting randomly generated available particles into a trained prediction model to obtain a prediction result output by the prediction model; the prediction result comprises: whether the available particles are particles on the Pareto front; the prediction model is obtained based on training with a training set; the training set comprises: ordinary particles and particles on the Pareto front obtained by performing multi-objective optimization of the coupling mechanism through an intelligent algorithm; based on the prediction result, the wireless power transmission system is optimized.
[0142] The device embodiments described above are merely illustrative, wherein the units described as separate components may or may not be physically separated, and the components displayed as units may or may not be physical units, that is, they may be located in one place, or they may be distributed on multiple network units. Some or all of the modules may be selected according to actual needs to achieve the purpose of the scheme of this embodiment. Ordinary technicians in this field can understand and implement it without paying creative labor.
[0143] Through the description of the above implementation methods, those skilled in the art can clearly understand that each implementation method can be implemented by means of software plus a necessary general hardware platform, and of course, it can also be implemented by hardware. Based on this understanding, the above technical solution is essentially or the part that contributes to the prior art can be embodied in the form of a software product, and the computer software product can be stored in a computer-readable storage medium, such as ROM / RAM, a disk, an optical disk, etc., including a number of instructions for a computer device (which can be a personal computer, a server, or a network device, etc.) to execute the methods described in each embodiment or some parts of the embodiments.
[0144] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention, rather than to limit it. Although the present invention has been described in detail with reference to the aforementioned embodiments, those skilled in the art should understand that they can still modify the technical solutions described in the aforementioned embodiments, or make equivalent replacements for some of the technical features therein. However, these modifications or replacements do not deviate the essence of the corresponding technical solutions from the spirit and scope of the technical solutions of the embodiments of the present invention.
Claims
1. A wireless power transmission system optimization method based on machine learning, characterized in that: include: Inputting the randomly generated available particles into the trained prediction model to obtain the prediction results output by the prediction model; The prediction result includes: whether the available particle is a particle at the Pareto front; the prediction model is obtained based on training of a training set; the training set includes: ordinary particles and particles at the Pareto front obtained by multi-objective optimization of the coupling mechanism through an intelligent algorithm; Based on the prediction results, the wireless power transmission system is optimized; The step of optimizing the wireless power transmission system based on the prediction result includes: When it is determined that the available particles are particles on the Pareto front, based on the available particles, an intelligent algorithm is used to optimize the wireless power transmission system; The available particles are particles on the Pareto frontier, determined based on the prediction result; Wherein, based on the available particles, the wireless power transmission system is optimized by using an intelligent algorithm, including: Based on the available particles, the independent variables corresponding to the coupling mechanism, the constraints, the optimization objectives and the value range of the independent variables, an intelligent algorithm is used for optimization to obtain the first Pareto frontier result; Based on the first Pareto front result, the wireless power transmission system is optimized.
2. The wireless power transmission system optimization method based on machine learning according to claim 1 is characterized in that: The training set is obtained in the following way: Based on the independent variables, constraints, optimization objectives and value ranges of the independent variables corresponding to the coupling mechanism, an intelligent algorithm is used for optimization to obtain the second Pareto frontier result; Determine ordinary particles and particles at the Pareto front based on the second Pareto front results corresponding to coupling mechanisms of different power levels; The common particles and the particles on the Pareto front are marked to obtain the training set.
3. The wireless power transmission system optimization method based on machine learning according to claim 1, characterized in that: The prediction model is trained based on a machine learning algorithm.
4. A wireless power transmission system optimization device based on machine learning, characterized in that: include: An input module, used to input randomly generated available particles into a trained prediction model to obtain a prediction result output by the prediction model; The prediction result includes: whether the available particle is a particle at the Pareto front; the prediction model is obtained based on training of a training set; the training set includes: ordinary particles and particles at the Pareto front obtained by multi-objective optimization of the coupling mechanism through an intelligent algorithm; An optimization module, used for optimizing the wireless power transmission system based on the prediction result; Wherein, the optimization module is used to: When it is determined that the available particles are particles on the Pareto front, based on the available particles, an intelligent algorithm is used to optimize the wireless power transmission system; The available particles are particles on the Pareto frontier, determined based on the prediction result; Wherein, based on the available particles, the wireless power transmission system is optimized by using an intelligent algorithm, including: Based on the available particles, the independent variables corresponding to the coupling mechanism, the constraints, the optimization objectives and the value range of the independent variables, an intelligent algorithm is used for optimization to obtain the first Pareto frontier result; Based on the first Pareto front result, the wireless power transmission system is optimized.
5. An electronic device comprising a memory, a processor, and a computer program stored in the memory and running on the processor, characterized in that: When the processor executes the program, the wireless power transmission system optimization method based on machine learning as described in any one of claims 1 to 3 is implemented.
6. A non-transitory computer-readable storage medium having a computer program stored thereon, characterized in that: When the computer program is executed by a processor, the wireless power transmission system optimization method based on machine learning as described in any one of claims 1 to 3 is implemented.
7. A computer program product, comprising a computer program, characterized in that When the computer program is executed by a processor, the wireless power transmission system optimization method based on machine learning as described in any one of claims 1 to 3 is implemented.
