Optimization method for minimum magnetic core loss and maximum transmission magnetic energy based on dual-objective function

Abstract

The invention relates to the technical field of new energy grid connection, in particular to an optimization method for minimum magnetic core loss and maximum transmission magnetic energy based on double objective functions, which comprises the following steps: step 1, defining objective functions according to optimal design, the objective function 1 being a minimum magnetic core design, and the objective function 2 being a maximum transmission magnetic energy; 2, decision variables are determined, wherein the model mainly comprises five variables including temperature, frequency, a magnetic flux density peak value, a magnetic core material and an excitation waveform; the method comprises the following steps of: (1) determining a constraint condition to ensure the feasibility and rationality of a model, and effectively improving the practicability and accuracy of an optimization model, (4) carrying out double-objective function operation on five decision variables of experimental data, and (5) calculating an overall condition meeting the optimal design of a magnetic element according to a particle swarm optimization algorithm. Through the innovative modeling and optimization scheme, the magnetic core loss can be effectively reduced, the transmission magnetic energy is increased, and the reasonability and effectiveness of the model are optimized.

Description

Technical Field

[0001] The present invention relates to the technical field of new energy grid connection, and in particular to an optimization method based on a dual objective function of minimum magnetic core loss and maximum transmission magnetic energy. Background Art

[0002] With the rapid development of energy transformation and information technology, high-frequency, high-efficiency, and high-power density power conversion technology has become a key support. Especially in the fields of new energy grid connection, efficient power supply for data centers, and electric vehicle charging, the performance of magnetic components directly determines the system performance and cost. However, the core loss of magnetic components under the action of high-frequency alternating magnetic flux is difficult to accurately predict due to material complexity and multivariable coupling, and existing models are difficult to meet actual needs.

[0003] By systematically collecting loss data of various core materials under a wide range of temperature, frequency and flux density conditions, and combining electromagnetic field theory with data processing technology, we explore the nonlinear law of core loss and provide solid theoretical and technical support for the optimal design and high-efficiency application of magnetic components. Summary of the invention

[0004] The purpose of the present invention is to solve the shortcomings existing in the prior art, and proposes an optimization method based on the dual objective function of minimum core loss and maximum transmission magnetic energy. By systematically collecting loss data of various core materials under a wide range of temperature, frequency and flux density conditions, combining electromagnetic field theory and data processing technology, the optimal design and high-efficiency application of magnetic components can be achieved.

[0005] In order to achieve the above object, the present invention adopts the following technical solutions:

[0006] An optimization method based on dual objective functions of minimum core loss and maximum transmission magnetic energy comprises the following steps:

[0007] Step 1: Define the objective function according to the optimization design, where objective function 1 is to minimize the core design and objective function 2 is to maximize the transmitted magnetic energy;

[0008] Step 2: Determine the decision variables, which include five in the objective function: temperature, frequency, peak value of magnetic flux density, core material, and excitation waveform;

[0009] Step 3: Determine constraints, including: physical and engineering constraints, design specifications and standards, economic and environmental constraints, and overall constraints;

[0010] Step 4: Calculate how each decision variable satisfies the minimization of core loss and maximization of transmitted magnetic energy individually;

[0011] Step 5: Calculate the overall conditions that meet the optimal design of magnetic components based on the particle swarm optimization algorithm.

[0012] Preferably, in step 1, core loss generally refers to energy loss in magnetic materials due to hysteresis effect, eddy current effect and other factors. This loss can be predicted by different models, among which the commonly used models include the improved Steinmetz equation. The objective function 1 can be expressed as:

[0013] L(T,f,Bm,Waveform,Material)=k1(Material)·fα 1 ·Bm β1 ·C(T) (1)

[0014] Where L is the core loss (W / m3), T is the temperature (°C), f is the frequency (Hz), and B m is the peak value of magnetic flux density (T), Waveform is the excitation waveform type (such as sine wave, triangle wave, trapezoidal wave), Material is the core material type (such as material 1, material 2, etc.), k1(Material) is the loss parameter of different materials, usually obtained through experimental data, and C(T) is the temperature correction coefficient.

[0015] The goal of minimizing core losses is to reduce the energy loss of magnetic components under specific operating conditions, thereby improving energy efficiency. By optimizing conditions (such as temperature, frequency, waveform, and flux density), the overall efficiency of the power converter will be improved. This will not only help reduce energy consumption, but also extend the service life of the equipment and improve its reliability, thus providing better theoretical support for the design of high-efficiency equipment.

[0016] Among them, objective function 2: maximize the transmission magnetic energy

[0017] Transmitted magnetic energy refers to the energy that magnetic materials can transmit under certain conditions. Since the concept of transmitted magnetic energy is relatively complex, a simplified representation method is used here:

[0018] E(f,B m )=f·B m (2)

[0019] Where, E is the transmitted magnetic energy (W), f is the frequency (Hz), B m is the peak magnetic flux density (T).

[0020] By maximizing the transmission of magnetic energy, efficiency can be improved, energy loss can be minimized, and the overall system efficiency can be improved. At the same time, a clear optimization direction is provided to guide the selection of the best frequency and flux density under specific conditions, which ultimately helps to optimize the design of magnetic components.

[0021] In actual optimization, these two objectives need to be combined to form a multi-objective optimization model. Usually, the objectives are:

[0022] minL(T,f,B m ,Waveform,Material),maxE(f,B m )(3)

[0023] In order to deal with these two objective functions, we can use weighting method, Pareto optimization and other methods to merge them into a single objective function. For example, we can define a comprehensive objective function:

[0024] Z=ω1·L-ω2·E(4)

[0025] Among them, ω1 and ω2 are the weights of loss and transmitted magnetic energy, indicating the importance of different goals. By adjusting the weights, the relationship between core loss and transmitted magnetic energy can be flexibly balanced in practical applications. Figure 1 The three-dimensional relationship between the transmitted magnetic energy, frequency and peak magnetic flux density can be well demonstrated.

[0026] Preferably, in step 2, the dual objective function is firstly calculated on the five decision variables to obtain the conditions satisfied by each of them.

[0027] Preferably, in step three, the final model is:

[0028]

[0029] Among them, ω1 and ω2 are the weights of loss and transmission magnetic energy, regulating the importance between the two.

[0030] Make

[0031] Among them, B m is the peak value of magnetic flux density, ω1 and ω2 are the weights of loss and transmitted magnetic energy, which adjust the importance between the two.

[0032] Preferably, in step five, the genetic algorithm is compared with the particle swarm algorithm, and a better algorithm is used to seek the overall conditions that satisfy the optimization of the magnetic element.

[0033] Compared with the prior art, the present invention has the following beneficial effects:

[0034] 1. The present invention systematically collects loss data of various magnetic core materials under a wide range of temperature, frequency and flux density conditions, and combines electromagnetic field theory with data processing technology to achieve optimal design and high-efficiency application of magnetic components.

[0035] 2. The present invention can effectively reduce core loss, increase transmission magnetic energy, and ensure the rationality and effectiveness of the optimization model through innovative modeling and optimization schemes.

[0036] 3. The present invention utilizes adaptive differential evolution algorithm and particle swarm algorithm to realize dual-objective optimization problem, namely, minimum core loss and maximum transmission magnetic energy problem, through Python.

[0037] 4. The present invention optimizes the minimum core loss and the maximum transmitted magnetic energy problems, finds the best conditions under the decision variables such as temperature, frequency, waveform, material, and magnetic flux density, and provides solid theoretical and technical support for the optimal design and high-efficiency application of magnetic components. BRIEF DESCRIPTION OF THE DRAWINGS

[0038] Figure 1 A three-dimensional relationship diagram of the transmission magnetic energy, frequency, and peak value of magnetic flux density in the present invention;

[0039] Figure 2 It is the convergence curve diagram of the PSO algorithm in the present invention;

[0040] Figure 3 The figure is a relationship diagram between the transmission magnetic energy and the core loss obtained by the present invention after calculating the experimental data through the optimization model;

[0041] Figure 4 It is a flow chart of the present invention. DETAILED DESCRIPTION

[0042] The technical solutions in the embodiments of the present invention will be described clearly and completely below in conjunction with the accompanying drawings, so that those skilled in the art can better understand the advantages and features of the present invention, thereby making a clearer definition of the protection scope of the present invention. The embodiments described in the present invention are only a part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments in the present invention, all other embodiments obtained by ordinary technicians in the field without making creative work are within the scope of protection of the present invention.

[0043] Example:

[0044] Step 1: Define the objective function

[0045] In multi-objective optimization, defining the objective function is a key step, which directly affects the solution direction and final result of the model. For the optimization problem of magnetic components, two objective functions are established: objective function 1 minimizes the core loss and objective function 2 maximizes the transmitted magnetic energy.

[0046] Step 2: Determine the decision variables

[0047] ① Temperature

[0048] First of all, temperature is the core factor affecting the performance of magnetic materials, usually expressed in degrees Celsius (℃), and its selection range should be based on the material's use environment and working conditions. Temperature changes will directly affect the loss characteristics and magnetic properties of the core. High temperature may cause changes in resistance and permeability, which in turn affects the loss.

[0049] ②Frequency

[0050] Frequency refers to the rate of change of the excitation waveform, measured in Hertz (Hz). The specific value of the selection range depends on the operating frequency of the power electronic equipment. Different frequencies will significantly affect the loss characteristics of the magnetic core. Generally, higher frequencies will lead to greater magnetic losses, so they must be selected with caution.

[0051] ③Excitation waveform type

[0052] The excitation waveform type is the current or voltage waveform applied to the magnetic core. In this study, sine wave, triangle wave and trapezoidal wave are selected. The effect of different waveform shapes on the change of magnetic flux density is directly related to core loss and energy transmission efficiency.

[0053] ④ Magnetic flux density peak

[0054] The peak flux density refers to the maximum flux density applied to the core, usually in Tesla (T). The selection range should be set according to the specific material and design requirements. The change of flux density will directly affect the loss and transmission efficiency of the core. Higher flux density usually leads to higher losses, but it may also improve energy transmission capacity.

[0055] ⑤Core material type

[0056] The core material type refers to the type of material used to make the core. The magnetic properties of different materials, such as permeability and loss factor, significantly affect the overall performance of the core. Therefore, choosing the right material is particularly important for optimizing the design.

[0057] By setting these decision variables appropriately, the best conditions can be found in the optimization model to achieve the goals of minimum core loss and maximum transmitted magnetic energy.

[0058] Step 3: Construct constraints

[0059] Constructing constraints is indeed crucial in the optimization model, which ensures the feasibility and rationality of the model. It can effectively improve the practicality and accuracy of the optimization model and ensure that the final solution is feasible and meaningful in reality.

[0060] ① Physical and engineering constraints

[0061] Physical and engineering constraints are fundamental to ensure that the design results meet actual operating conditions. For example, temperature constraints should set the minimum and maximum operating temperatures (Tmin and T max ) to prevent material performance degradation or failure; the frequency range must also be limited to the effective working range (f min and f max ) to avoid increased losses or decreased efficiency. In addition, the peak value of the magnetic flux density (B m ) needs to be within a reasonable range to ensure that the material is not saturated and the performance is stable.

[0062] ②Design specifications and standards

[0063] Design specifications and standards are also an important part of constraints. These standards usually come from industry requirements or specific design goals. For example, material selection constraints can ensure that core materials with specific performance are selected. Different waveform types can be represented by integer encoding, where 0 represents a sine wave, 1 represents a triangle wave, and 2 represents a trapezoidal wave. The waveform type constraint ensures that only specific excitation waveforms are allowed, and integer encoding is used to represent different materials. 0 represents material 1, 1 represents material 2, 2 represents material 3, and 3 represents material 4.

[0064] ③Economic and environmental constraints

[0065] Economic and environmental considerations are also constraints that cannot be ignored, but they are not considered in this study.

[0066] ④ Overall constraints

[0067] The overall constraints also need to consider the relationship between decision variables. For example, a specific waveform type may only be valid within a certain frequency range, and this dependency should also be reflected in the model. By constructing these constraints reasonably, the solution space of the optimization model can be effectively limited, thereby obtaining the best design solution that meets actual needs and improving the performance and reliability of magnetic components.

[0068] Step 4: Calculate the dual objective function for the five decision variables of the experimental data

[0069] ① Temperature

[0070] The optimal temperature is 63.03℃. As the temperature continues to rise, the core loss tends to increase, but when it reaches 63.03℃, the magnetic properties of the material have not degraded significantly, and it can still provide high magnetic energy transmission efficiency and perform well in loss control. At this time, the loss at this temperature is maintained at a low level (3741.64W / m 3 ), indicating that the working condition at this temperature is ideal.

[0071] ②Frequency

[0072] The optimal frequency is 267665.32Hz. The frequency is positively correlated with the transmitted magnetic energy, so when the transmitted magnetic energy is significantly increased to 22831.85Hz·T in a high-frequency environment, the frequency is 267665.32Hz. However, high frequency will also increase the core loss, so loss control at this frequency is also very important. According to the analysis results, the minimum loss can still be maintained at 3741.64W / m in this high-frequency environment. 3 level, indicating that the performance of the core material and waveform at this frequency balances transmission efficiency and loss.

[0073] ③ Waveform

[0074] The optimal waveform is a triangular wave. The triangular wave has the characteristics of linear rise and fall, which causes the magnetic flux density to change rapidly over time, so it will produce higher eddy current losses at high frequencies. However, from the optimization results, the triangular wave can achieve a larger transmission magnetic energy (2831.85W) under this condition, and the loss is controlled at 3741.64W / m 3 This indicates that under the current conditions, the combination of the triangle wave with other variables (such as temperature, frequency, and magnetic flux density) is more balanced and suitable for efficient energy transmission compared with the other two waveforms.

[0075] ④ Materials

[0076] The best material is material 1. The four different core materials have different magnetic properties and thermal stability, and are suitable for different application scenarios. In this optimization, material 1 can achieve lower core loss under the previous optimal conditions (high frequency and high temperature), indicating that its hysteresis loss and eddy current loss are small, and it is a suitable material choice for high frequency and high temperature occasions.

[0077] ⑤ Magnetic flux density peak

[0078] The optimal magnetic flux density peak is 0.0853T (Tesla), which belongs to a smaller magnetic flux density range. A lower magnetic flux density can effectively reduce core loss, because the core material is prone to hysteresis loss and eddy current loss at a higher magnetic flux density. Therefore, by limiting the peak magnetic flux density, the core loss can be minimized while ensuring the maximum transmission of magnetic energy.

[0079] Step 5: Calculate the overall conditions that meet the optimal design of magnetic components based on the particle swarm optimization algorithm.

[0080] The specific steps of the particle swarm algorithm are as follows:

[0081] Step 1: Initialize the particle swarm

[0082] In the particle swarm optimization (PSO) algorithm, initializing the particle swarm is a crucial first step. Each particle represents a possible solution, that is, a combination of the optimal conditions for core loss and transmitted magnetic energy. The following are the specific steps to initialize the particle swarm:

[0083] ①Number of particles

[0084] For the current problem, the present invention sets the number of particle swarms to 50, which means that during the iteration process of the algorithm, the present invention will simultaneously track and evaluate 50 potential solutions.

[0085] ② Initialization of particle position

[0086] In the particle swarm optimization algorithm, the position of the particle is a visual representation of the combination of decision variables. These decision variables, including but not limited to temperature, frequency, waveform, peak flux density, and core material, constitute the key parameter set that needs to be solved in the optimization problem. Each decision variable is limited by its physical properties and has clear upper and lower limits. These restrictions can be set specifically based on experimental data to ensure that the position of the particle can accurately map the possible value range of the decision variable in the actual situation.

[0087] Among them, the setting range of temperature T (decision variable 1) is [25℃, 90℃]; the setting range of frequency f (decision variable 2) is [50000Hz, 500000Hz]; waveform type (decision variable 3) is a discrete variable, representing three waveforms (1: sine wave, 2: triangle wave, 3: trapezoidal wave); the peak value of magnetic flux density B m The range of (decision variable 4) is determined by experimental data; the core material (decision variable 5) is a discrete variable representing different core materials (0: material 1, 1: material 2, 2: material 3, 3: material 4).

[0088] The particle position initialization formula is:

[0089]

[0090] in, represents the initial position of the i-th particle on the j-th decision variable, L j and U j are the upper and lower limits of the decision variable, rand(L j ,U j ) generates a random number within the range of the change amount.

[0091] ③ Initialization of particle velocity

[0092] Particle velocity initialization formula:

[0093]

[0094] in, represents the initial velocity of the i-th particle on the j-th decision variable, is the maximum velocity value of the variable, usually a ratio of the variable's value range.

[0095] Step 2: Update particle status

[0096] In the framework of the particle swarm optimization (PSO) algorithm, the state of each particle in the search space is characterized by its position information and velocity information. In each iteration of the algorithm, the state of the particle is updated in a timely manner according to the evaluation result of the fitness function (i.e., the objective function). This update mechanism mainly relies on two core elements: one is the best position found by the particle itself so far, that is, the individual best position (pbest); the other is the best position explored by the entire particle group so far, that is, the global best position (gbest). By combining the information of these two elements, the particle can continuously adjust its own state in order to find a better solution in the search space.

[0097] ① Calculate fitness

[0098] In each iteration of the particle swarm optimization algorithm, the present invention first calculates the fitness value of each particle according to its current position, that is, the value of the objective function. The objective function is usually a weighted sum designed for a multi-objective problem, and its core purpose is to minimize the core loss and maximize the transmitted magnetic energy. Therefore, the specific form of this function needs to be precisely defined according to the actual design goals and requirements.

[0099] ② Update the best individual position (pbest)

[0100] For each particle i, a historical best position can be obtained in the previous iteration, which is the best position obtained by this particle. If the particle fitness f(x i (t)) is better than its individual best historical fitness f(pbest i ), then update the individual best position:

[0101] pbest i =x i (t) (9)

[0102] Then the current position x of the current particle i (t) will become the new individual optimal position.

[0103] ③Update the global best position (gbest)

[0104] The global best position gbest is the best position reached by all particles in the entire group. It indicates the best solution of the particle in all solution spaces; if the fitness of the current particle f(x i (t)) is better than the current global best fitness f(gbest), then the best position will be updated accordingly:

[0105] gbest=x i (t) (10)

[0106] It means that the optimal solution in the particle is replaced by the current position of the particle.

[0107] ④Update speed and position

[0108] After updating the individual best pbest and the global best gbest, the speed and position of each particle are updated according to the update formula of the particle swarm algorithm:

[0109] v i (t+1)=w·v i (t)+c1·r1·(pbest i -x i (t))+c2·r2·(gbest-x i (t)) (11)

[0110] Where w is the inertia weight, which controls the decay of the particle velocity; c1 and c2 are learning factors or acceleration constants; r1 and r2 are random numbers between [0,1] used to introduce randomness; v i (t) is the velocity of the particle at the current iteration.

[0111] The new position of the particle will also change according to the corresponding formula, that is, the new position is equal to the old position plus the particle's velocity.

[0112] x i (t+1)=x i (t)+v i (t+1) (12)

[0113] ⑤Iterative process

[0114] In the particle swarm optimization algorithm, the above process will be executed repeatedly, and the speed and position information of each particle will be updated through continuous iteration. After each iteration, the particle will adjust its movement direction according to the guidance of the individual best position (pbest) and the global best position (gbest), and gradually approach the global optimal solution. This iterative process will continue until the preset maximum number of iterations is reached, or when the change of the global best fitness value in multiple consecutive iterations becomes very small, which means that the algorithm has found a near-optimal solution and the solution has stabilized. And the convergence curve of the particle swarm algorithm is obtained.

[0115] Through the particle swarm optimization (PSO) algorithm, when a higher frequency, moderate temperature, triangular wave excitation waveform and material 1 are met, the maximum transmission magnetic energy (22831.85W) and the minimum core loss (3741.64W / m 3 ).

[0116] In addition, the basic symbols in the embodiments of the present invention are explained as follows:

[0117]

[0118] In summary, the present invention can effectively improve the working efficiency of magnetic components through innovative modeling and optimization schemes, and make the transmitted magnetic energy as large as possible while minimizing the core loss, so that the magnetic components can be more efficient.

[0119] The description and practice disclosed in the present invention are easy to think and understand for ordinary technicians in the technical field, and several improvements and modifications can be made without departing from the principles of the present invention. Therefore, modifications or improvements made without departing from the spirit of the present invention should also be regarded as the protection scope of the present invention.

Claims

1. An optimization method based on the dual objective function of minimum core loss and maximum transmission magnetic energy, characterized in that: The steps include: Step 1: Define the objective function according to the optimization design, where objective function 1 is to minimize the core design and objective function 2 is to maximize the transmitted magnetic energy; Step 2: Determine the decision variables, which include five in the objective function: temperature, frequency, peak value of magnetic flux density, core material, and excitation waveform; Step 3: Determine constraints, including: physical and engineering constraints, design specifications and standards, economic and environmental constraints, and overall constraints; Step 4: Calculate how each decision variable satisfies the minimization of core loss and maximization of transmitted magnetic energy individually; Step 5: Calculate the overall conditions that meet the optimal design of magnetic components based on the particle swarm optimization algorithm.

2. The optimization method according to claim 1, characterized in that: In step 1, the objective function 1 is expressed as: Where L is the core loss W / m 3 , T is temperature °C, f is frequency Hz, B m is the peak value of magnetic flux density T, Waveform is the excitation waveform type, Material is the core material type, k1(Material) is the loss parameter of different materials obtained through experimental data, and C(T) is the temperature correction coefficient; The representation method used by objective function 2 is: E(f,B m )=f·B m (2) Where E is the transmitted magnetic energy W, f is the frequency Hz, B m is the peak value of magnetic flux density T; Combining the above two objectives, a multi-objective optimization model is formed, whose goals are: minL(T,f,B m ,Waveform,Material),maxE(f,B m )(3) In order to deal with these two objective functions, the weight method and Pareto optimization method are used to merge them into a single objective function; a comprehensive objective function is defined: Z=ω1·L-ω2·E (4) Among them, ω1 and ω2 are the weights of loss and transmitted magnetic energy, indicating the importance of different objectives. By adjusting the weights, the relationship between core loss and transmitted magnetic energy can be flexibly balanced in practical applications.

3. The optimization method based on the dual objective function of minimum core loss and maximum transmission magnetic energy according to claim 1, characterized in that: In step three, the final model is: Among them, B m is the peak value of magnetic flux density, ω1 and ω2 are the weights of loss and transmitted magnetic energy, which adjust the importance between the two.

4. The optimization method based on the dual objective function of minimum core loss and maximum transmission magnetic energy according to claim 1, characterized in that: In step 2, the dual objective function is calculated for each of the five decision variables to obtain the conditions that each of them satisfies.

5. The optimization method based on the dual objective function of minimum core loss and maximum transmission magnetic energy according to claim 1, characterized in that: In step five, the genetic algorithm is compared with the particle swarm algorithm, and the better algorithm is used to find the overall conditions that meet the optimization of the magnetic components.

Images