Injection power optimal distribution method and system

Abstract

The invention discloses an optimal distribution method for injection power. The method comprises the following steps: acquiring an operation signal of a generator, decomposing the operation signal by using a variational mode decomposition algorithm to obtain a mode function, calculating the mode function, solving an analysis signal, and obtaining a unilateral frequency spectrum; adjusting by using a correction method to obtain a unilateral frequency spectrum consistent with a corresponding base frequency band, and calculating a gradient square norm of the analysis signal to obtain a bandwidth of a modal component; constructing a constrained variational model, and adding an augmented Lagrange coefficient to obtain a non-constrained variational model; solving the unconstrained variational model to obtain parameters of variational mode decomposition; carrying out optimization solution by utilizing an optimization algorithm to obtain an optimal parameter combination; charge and discharge power signals of the hybrid energy storage system are decomposed to obtain signal components, a demarcation point of power distribution is determined by using mutual information entropy, output power of energy storage equipment is adjusted according to a self-adaptive strategy to obtain power distribution, and higher solving precision is achieved.

Description

Technical Field

[0001] The present invention relates to the technical field of generator power distribution, and in particular to an injection power optimization distribution method and system. Background Art

[0002] During the operation of large generators, there is a risk of partial discharge of the stator bars in the stator slots. By detecting the contact state between the stator bars and the stator core slots, the corona discharge state in the stator bar slots can be effectively evaluated. The corona discharge state detection system in the stator bar slots of large generators is usually powered by lithium-ion batteries, which makes the system easier to wire when working on the generator maintenance site. Lithium-ion batteries have excellent performance and are widely used. They are very suitable for applications with high requirements for power energy density. In the application of large generator contact coefficient test, it is necessary to inject power that can be quickly and dynamically adjusted into the test point. Therefore, the existing power system using a single lithium-ion battery can be improved by using a high-power density supercapacitor. Reasonable energy management strategies and control methods for hybrid energy storage systems can give full play to the advantages of each energy storage component, effectively improve the safety and work efficiency of the power system, optimize the operation stability of the power system, and extend the service life of the energy storage components.

[0003] At present, the common power allocation methods of hybrid energy storage systems mainly include low-pass filtering, Fourier transform, wavelet transform and empirical mode decomposition. Among them, low-pass filters often produce a certain time delay during filtering, which may lead to inappropriate power allocation of hybrid energy storage systems. The effect of wavelet decomposition depends to a large extent on the selected basis function. Although the empirical mode decomposition method avoids the problem of relying on wavelet basis functions, it is difficult to accurately separate modal functions with similar frequencies during its recursive decomposition process, resulting in the existence of modal aliasing in each modal function. The variational mode decomposition (VMD) method can adaptively determine the optimal center frequency and bandwidth of each mode during the signal decomposition process, and can effectively separate the intrinsic mode function and perform frequency domain segmentation on the signal, so as to optimize the dynamic allocation of injected power of the hybrid energy storage system in the large generator contact coefficient test. Summary of the invention

[0004] In view of the above-mentioned problems, the present invention is proposed.

[0005] Therefore, the technical problem solved by the present invention is: the present invention solves the problem that in the process of variational mode decomposition, there is currently no unified default value to set the decomposition scale K in the variational mode decomposition algorithm, and the selection of the penalty factor has a significant impact on the bandwidth of each modal component.

[0006] In order to solve the above technical problems, the present invention provides the following technical solutions:

[0007] In a first aspect, the present invention provides a method for optimizing injection power allocation, comprising:

[0008] Acquire an operating signal of the generator, decompose the operating signal using a variational modal decomposition algorithm to obtain a modal function, calculate the modal function according to a function transformation method, obtain an analytical signal of the modal function, and obtain a unilateral spectrum of a modal component corresponding to the modal function;

[0009] The single-sided spectrum is adjusted by using a correction method to obtain a single-sided spectrum that is consistent with the corresponding baseband, and the square norm of the gradient of the analytical signal is calculated to obtain the bandwidth of the corresponding component of the modal function;

[0010] constructing a constrained variational model based on the bandwidth, and adding augmented Lagrangian coefficients to the constrained variational model to obtain an unconstrained variational model;

[0011] The unconstrained variational model is solved by using the alternating direction multiplier method and the energy conservation theorem to obtain the parameters of the variational mode decomposition;

[0012] Optimizing and solving the parameters of the variational mode decomposition using an optimization algorithm to obtain an optimal parameter combination;

[0013] The charging and discharging power signals of the hybrid energy storage system are decomposed using the optimal parameter combination of the variational mode decomposition algorithm to obtain signal components, and the demarcation point of power distribution is determined using mutual information entropy. For the signal components, the output power of the energy storage device is adjusted according to an adaptive strategy to obtain power distribution.

[0014] As a preferred solution of the injection power optimization allocation method of the present invention, wherein: adjusting the single-sided spectrum by using a correction method comprises:

[0015] The modal function is calculated using Hilbert transform, the analytical signal related to the modal function is obtained, and the estimated center frequency of the modal component corresponding to the modal function is obtained;

[0016] Using the exponential correction method, the unilateral spectrum of the modal component is adjusted so that the unilateral spectrum of the modal component falls on the corresponding baseband, which is expressed as:

[0017]

[0018] Among them, u k (t) is the mode function, is the estimated center frequency of the modal component, j is δ(t) is the Dirac equation, π is the circumference of a circle, and t is time.

[0019] As a preferred solution of the injection power optimization allocation method of the present invention, wherein: constructing a constrained variational model based on the bandwidth includes:

[0020] By calculating the square norm of the analytical signal gradient, the bandwidth of the modal component is obtained, and the constructed constrained variational model is expressed as:

[0021]

[0022] Among them, k is the number of decomposed modes, {u k} is the set of all modal components after decomposition, {ω k} is the set of center frequencies of the modal components, * is the convolution symbol, and f is the original signal.

[0023] As a preferred solution of the injection power optimization allocation method described in the present invention, adding augmented Lagrangian coefficients to the constrained variational model includes:

[0024] Add Lagrange multiplication operators and quadratic penalty factors to the constrained variational model.

[0025] As a preferred solution of the injection power optimization allocation method of the present invention, the solution of the unconstrained variational model includes:

[0026] The alternating direction multiplier method is combined with Parseval's theorem to solve the problem based on the equivalence of Fourier transform.

[0027] According to the Hermitian symmetry property of the signal, the unconstrained variational model formula is expressed by integrating in the non-negative frequency interval to obtain the first unconstrained variational model formula. The unconstrained variational model formula is derived and expressed as:

[0028]

[0029] The center frequency is similarly expressed as:

[0030]

[0031] in, is the center frequency of the current modal component, ω is the angular frequency, and α is the quadratic penalty factor introduced in the augmented Lagrangian method.

[0032] As a preferred solution of the injection power optimization allocation method described in the present invention, the modal components include:

[0033] In the process of obtaining the modal components, for each modal component, the square of the Euclidean norm of the difference between the modal component estimates obtained at the n+1th iteration and the nth iteration is calculated, divided by the square of the Euclidean norm of the modal component estimate obtained at the nth iteration, and the ratios of all modal components are summed;

[0034] If the sum is less than the preset positive number precision threshold, the iteration process of the modal component has met the precision requirement and the iteration can be stopped; if the sum is greater than the preset positive number precision threshold, the iteration needs to continue.

[0035] As a preferred solution of the injection power optimization allocation method described in the present invention, the optimization algorithm is used to optimize and solve the parameters of the variational mode decomposition, which includes:

[0036] Use the randomly generated parameter combination as the particle's initial position and initialize the velocity;

[0037] Under different particle position conditions, the variational mode decomposition algorithm is used to calculate the operating signal, the fitness value of each particle is calculated, and the global optimum of the particle is recorded;

[0038] Perform update operations to update the positions and velocities of the global optimal particle and other particles respectively, calculate the fitness values ​​of all particles, and update the global optimal particle;

[0039] If the loop iteration reaches the maximum number of iterations, the global optimal solution and particle position are output;

[0040] If the loop iteration does not reach the maximum number of iterations, the update operation continues.

[0041] In a second aspect, the present invention provides a system for optimizing injection power allocation, comprising:

[0042] A decomposition module is used to obtain an operating signal of the generator, decompose the operating signal using a variational modal decomposition algorithm to obtain a modal function, calculate the modal function according to a function transformation method, obtain an analytical signal of the modal function, and obtain a unilateral spectrum of a modal component corresponding to the modal function;

[0043] An adjustment module, used for adjusting the single-sided spectrum by using a correction method, so that the single-sided spectrum is consistent with the corresponding baseband, calculating the gradient square norm of the analytical signal, and obtaining the bandwidth of the corresponding component of the modal function;

[0044] A model building module, used for building a constrained variational model based on the bandwidth, adding augmented Lagrangian coefficients to the constrained variational model to obtain an unconstrained variational model;

[0045] A model solving module, used for solving the unconstrained variational model by using an alternating direction multiplier method and the energy conservation theorem to obtain parameters of variational mode decomposition;

[0046] A parameter solving module, used for optimizing and solving the parameters of the variational modal decomposition by using an optimization algorithm to obtain an optimal parameter combination;

[0047] The distribution module is used to decompose the charging and discharging power signals of the hybrid energy storage system using the optimal parameter combination of the variational mode decomposition algorithm to obtain signal components, determine the demarcation point of power distribution using mutual information entropy, and adjust the output power of the energy storage device according to the adaptive strategy for the signal components to obtain power distribution.

[0048] In a third aspect, the present invention provides a computing device, comprising:

[0049] Memory and processor;

[0050] The memory is used to store computer executable instructions, and the processor is used to execute the computer executable instructions. When the computer executable instructions are executed by the processor, the steps of the injection power optimization allocation method are implemented.

[0051] In a fourth aspect, the present invention provides a computer-readable storage medium storing computer-executable instructions, which, when executed by a processor, implement the steps of the injection power optimization allocation method.

[0052] Beneficial effects of the present invention: The parameter-optimized variational modal decomposition algorithm proposed in the present invention makes the decomposition result of the power signal of the hybrid energy storage system more uniform and regular, effectively improves the modal aliasing phenomenon caused by EMD decomposition, and can clearly distinguish each sub-signal, thereby ensuring the superiority of the VMD algorithm signal decomposition result, and can more effectively complete the power signal decomposition task, laying the foundation for the reasonable power allocation of hybrid energy storage; the GCPSO algorithm is used to optimize the parameters of VMD, and compared with the iterative optimization results of the PSO algorithm, the convergence speed of the GCPSO algorithm is greatly improved, and it has a higher solution accuracy. BRIEF DESCRIPTION OF THE DRAWINGS

[0053] In order to more clearly illustrate the technical solutions of the embodiments of the present invention, the following briefly introduces the drawings required for describing the embodiments. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without creative labor. Among them:

[0054] Figure 1 An overall flow chart of an injection power optimization allocation method provided by the present invention;

[0055] Figure 2 The GCPSO iterative calculation result of the injection power optimization allocation method provided by the present invention;

[0056] Figure 3 The PSO iterative calculation result of the injection power optimization allocation method provided by the present invention;

[0057] Figure 4 A VMD spectrum diagram of an injection power optimization allocation method provided by the present invention when K=7 and α=2000;

[0058] Figure 5 A VMD spectrum diagram of an injection power optimization allocation method provided by the present invention when K=9 and α=1992;

[0059] Figure 6 The present invention provides an EMD decomposition spectrum diagram of an injection power optimization allocation method. DETAILED DESCRIPTION

[0060] In order to make the above-mentioned purposes, features and advantages of the present invention more obvious and easy to understand, the specific implementation methods of the present invention are described in detail below in conjunction with the drawings of the specification. Obviously, the described embodiments are part of the embodiments of the present invention, but not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary persons in the art without creative work should fall within the scope of protection of the present invention.

[0061] In the following description, many specific details are set forth to facilitate a full understanding of the present invention, but the present invention may also be implemented in other ways different from those described herein, and those skilled in the art may make similar generalizations without violating the connotation of the present invention. Therefore, the present invention is not limited to the specific embodiments disclosed below.

[0062] Secondly, the term "one embodiment" or "embodiment" as used herein refers to a specific feature, structure, or characteristic that may be included in at least one implementation of the present invention. The term "in one embodiment" that appears in different places in this specification does not necessarily refer to the same embodiment, nor does it refer to a separate or selective embodiment that is mutually exclusive with other embodiments.

[0063] The present invention is described in detail with reference to schematic diagrams. When describing the embodiments of the present invention, for the sake of convenience, the cross-sectional diagrams showing the device structure will not be partially enlarged according to the general scale, and the schematic diagrams are only examples, which should not limit the scope of protection of the present invention. In addition, in actual production, the three-dimensional dimensions of length, width and depth should be included.

[0064] At the same time, in the description of the present invention, it should be noted that the directions or positional relationships indicated by the terms "upper, lower, inner and outer" are based on the directions or positional relationships shown in the drawings, and are only for the convenience of describing the present invention and simplifying the description, rather than indicating or implying that the device or element referred to must have a specific direction, be constructed and operated in a specific direction, and therefore cannot be understood as limiting the present invention. In addition, the terms "first, second or third" are only used for descriptive purposes and cannot be understood as indicating or implying relative importance.

[0065] In the present invention, unless otherwise clearly specified and limited, the terms "install, connect, connect" should be understood in a broad sense, for example: it can be a fixed connection, a detachable connection or an integral connection; it can also be a mechanical connection, an electrical connection or a direct connection, or it can be indirectly connected through an intermediate medium, or it can be the internal communication of two components. For ordinary technicians in this field, the specific meanings of the above terms in the present invention can be understood according to specific circumstances.

[0066] Example 1

[0067] Reference Figure 1 , as an embodiment of the present invention, provides a method for optimizing injection power allocation, comprising:

[0068] S100: obtaining an operating signal of the generator, decomposing the operating signal using a variational modal decomposition algorithm to obtain a modal function, calculating the modal function according to a function transformation method, obtaining an analytical signal of the modal function, and obtaining a unilateral spectrum of a modal component corresponding to the modal function;

[0069] Specifically, the variational mode decomposition algorithm assumes that the signal is composed of K finite bandwidth intrinsic mode function components, which show the characteristics of narrowband separation in frequency, and each component has its own unique center frequency. Each intrinsic mode function is expressed as a complex frequency-amplitude modulation signal, which can be expressed as:

[0070]

[0071]

[0072] Among them, u k (t) is the mode function, A k (t) is the envelope amplitude of the mode function, is the instantaneous phase, ω k (t) is the center frequency of the mode function, when When it is an ascending function, compared with The rate of change, A k (t) and ω k (t) changes much more slowly.

[0073] It should be noted that as a non-recursive, nonlinear and non-stationary signal processing method, the variational mode decomposition algorithm combines the concepts of Wiener filtering and Hilbert transform, and realizes effective signal decomposition by constructing and solving variational problems. This algorithm design can essentially avoid the common endpoint effect and mode aliasing problems in empirical mode decomposition, bringing more accurate and reliable analysis tools to the field of time-frequency analysis.

[0074] The variational mode decomposition algorithm can accurately determine the center frequency and bandwidth of each component of the signal through multiple iterative optimization processes, and then obtain the decomposed components, realize accurate separation of the signal, and significantly improve the computing efficiency. After its initial application in vibration signal analysis, the variational mode decomposition algorithm has also been widely used in power systems. During the signal decomposition process, the algorithm can adaptively determine the optimal center frequency and bandwidth of each mode, and can effectively separate the intrinsic mode function and perform frequency domain segmentation on the signal. Compared with the empirical mode decomposition algorithm, it successfully overcomes the problems of modal aliasing and difficult to identify frequency characteristics, is built on a more solid mathematical theoretical foundation, and exhibits strong robustness.

[0075] S200: using a correction method to adjust the single-sided spectrum, so that the single-sided spectrum is consistent with the corresponding baseband, calculating the square norm of the gradient of the analytical signal, and obtaining the bandwidth of the corresponding component of the modal function;

[0076] In the embodiment of the present application, the modal function is calculated using the Hilbert transform, the analytical signal related to the modal function is obtained, and the estimated center frequency of the modal component corresponding to the modal function is obtained;

[0077] Using the exponential correction method, the unilateral spectrum of the modal component is adjusted so that the unilateral spectrum of the modal component falls on the corresponding baseband, which is expressed as:

[0078]

[0079] Among them, u k (t) is the mode function, is the estimated center frequency of the modal component, j is δ(t) is the Dirac equation, π is the circumference of a circle, and t is time.

[0080] Specifically, the key idea of ​​variational mode decomposition is to construct and solve a variational problem. First, the variational problem needs to be constructed. It is assumed that the original signal f can be decomposed into K components that meet specific conditions. These conditions require that each component is of finite bandwidth and has its own center frequency. When decomposing, the algorithm tries to minimize the sum of the center frequency bandwidths of all modal components, while ensuring that the decomposed modal components are equal to the original signal f when superimposed;

[0081] Specifically, the present invention preferably uses Hilbert transform, and can also use Fourier transform and wavelet transform;

[0082] It should be noted that the core idea of ​​the Hilbert transform is to transform a real function into another real function. This transform can easily process the phase information of the signal. The principle is mainly to perform Fourier transform on the original signal, then set the negative frequency part of its spectrum to zero, and finally perform inverse Fourier transform to obtain the Hilbert transform. Compared with other transforms, an important property of the Hilbert transform is to extract the phase information of the signal in the frequency domain, so it is often used to analyze the instantaneous characteristics of the signal in signal processing.

[0083] Through Hilbert transform, we can get the instantaneous frequency and instantaneous amplitude of the signal, so as to better understand the characteristics of the signal. In addition, Hilbert transform can also be used for signal envelope extraction, modulation recognition and other applications.

[0084] S300: Construct a constrained variational model based on bandwidth, add augmented Lagrangian coefficients to the constrained variational model, and obtain an unconstrained variational model;

[0085] In the embodiment of the present application, the bandwidth of the modal component is obtained by calculating the square norm of the analytical signal gradient, and the constructed constrained variational model is expressed as:

[0086]

[0087] Among them, k is the number of decomposed modes, {u k} is the set of all modal components after decomposition, {ω k} is the set of center frequencies of the modal components, * is the convolution symbol, and f is the original signal.

[0088] In an embodiment of the present application, a Lagrange multiplication operator and a quadratic penalty factor are added to the constrained variational model.

[0089] Specifically, the augmented Lagrangian method is used to add the Lagrangian multiplication operator and the quadratic penalty factor to the variational model. The quadratic penalty factor is a classic method that can improve the reconstruction accuracy when processing data containing Gaussian white noise.

[0090] In practice, the size of this parameter is usually inversely proportional to the noise level in the data; for example, in the absence of noise, an infinite penalty parameter would theoretically be required to ensure the accuracy of the reconstruction. The Lagrangian operator, on the other hand, is a mathematical tool that ensures strict enforcement of constraints.

[0091] The augmented Lagrangian function is expressed as:

[0092]

[0093] It should be noted that combining the two aspects of the Lagrangian operator not only ensures that the quadratic penalty factor with finite weights has good convergence, but also ensures that the constraints can be strictly enforced.

[0094] S400: Using the alternating direction multiplier method and the energy conservation theorem, the unconstrained variational model is solved to obtain the parameters of the variational mode decomposition;

[0095] In the embodiment of the present application, the alternating direction multiplier method is combined with Parseval's theorem to solve the problem based on the equivalence of Fourier transform;

[0096] According to the Hermitian symmetry property of the signal, the unconstrained variational model formula is expressed by integrating in the non-negative frequency interval to obtain the first unconstrained variational model formula. The unconstrained variational model formula is derived and expressed as:

[0097]

[0098] The center frequency is similarly expressed as:

[0099]

[0100] in, is the center frequency of the current modal component, ω is the angular frequency, and α is the quadratic penalty factor introduced in the augmented Lagrangian method.

[0101] Specifically, using the alternating direction multiplier method Parseval theorem and taking into account the equivalence of Fourier transform, this can be solved to obtain:

[0102]

[0103] Taking into account the Hermite symmetry of the signal, the unconstrained variational model formula is expressed by integrating in the non-negative frequency interval, and the first unconstrained variational model formula is expressed as:

[0104]

[0105] Taking the derivative of the above formula, we get:

[0106]

[0107] The center frequency is obtained in the same way;

[0108] is the center frequency of the current modal component, Perform inverse Fourier transform, and the real part is λ n+1 The solution is:

[0109]

[0110] It should be noted that the alternating direction multiplier method is an optimization algorithm used to solve constrained optimization problems. It is particularly suitable for optimization problems containing coupled variables or continuous constraints. The main idea is to avoid introducing new variables or adding additional constraints by alternating the directions of different variables and using techniques such as multipliers, thereby reducing the complexity of the problem. Compared with others, its advantage is that it is simple and easy to implement, and can effectively solve some complex optimization problems.

[0111] In the embodiment of the present application, in the process of obtaining the modal components, for each modal component, the square of the Euclidean norm of the difference between the modal component estimation values ​​obtained at the n+1th iteration and the nth iteration is calculated, divided by the square of the Euclidean norm of the modal component estimation value obtained at the nth iteration, and the ratios of all modal components are summed;

[0112] If the sum is less than the preset positive number precision threshold, the iteration process of the modal component has met the precision requirement and the iteration can be stopped; if the sum is greater than the preset positive number precision threshold, the iteration needs to continue.

[0113] Specifically, in the process of obtaining K modal components, it is necessary to ensure that the preset discrimination accuracy condition of ε>0 is met. This setting is a prerequisite for obtaining the modal components and ensures that the decomposition process can achieve the expected accuracy conditions, which can be expressed as:

[0114]

[0115] in, is the estimated value of the modal component obtained at the n+1th iteration, is the estimated value of the modal component obtained at the nth iteration.

[0116] It should be noted that in this way, the algorithm can ensure that the iteration stops after reaching a certain accuracy, thereby avoiding unnecessary calculations, saving time and resources, and ensuring the accuracy of the results. In addition, this method can be applied to a variety of signal processing and system analysis occasions to improve the robustness and applicability of the algorithm.

[0117] S500: Optimizing and solving the parameters of variational mode decomposition using an optimization algorithm to obtain an optimal parameter combination;

[0118] In the embodiment of the present application, the randomly generated parameter combination is used as the initial position of the particle and the velocity is initialized;

[0119] Under different particle position conditions, the variational mode decomposition algorithm is used to calculate the operating signal, the fitness value of each particle is calculated, and the global optimum of the particle is recorded;

[0120] Perform update operations to update the positions and velocities of the global optimal particle and other particles respectively, calculate the fitness values ​​of all particles, and update the global optimal particle;

[0121] If the loop iteration reaches the maximum number of iterations, the global optimal solution and particle position are output;

[0122] If the loop iteration does not reach the maximum number of iterations, the update operation continues.

[0123] Specifically, the guaranteed convergence particle swarm algorithm can update the global optimal particle individually under the premise of ensuring local convergence, including the update of particle velocity and position, which can be expressed as:

[0124]

[0125] in, To reset the particle position to the current best position, is the current search direction, The ρ in ρ is called the scaling factor, which controls the range of random search performed by the particle when it is close to the optimal solution. It is worth noting that the value of the scaling factor ρ is not fixed, but is dynamically adjusted according to the search results of each iteration. It will be updated as follows, expressed as:

[0126]

[0127] The update of the best historical particle can be divided into two cases: success and failure. If the best historical position does not change after an update, the update is considered a failure, otherwise it is a success.

[0128] For n consecutive successes, #success is recorded as n. Once a failure occurs, #success is cleared and #failure is recorded as 1. Similarly, if #failure fails n times in a row, it is recorded as n. Once a success occurs, #failure is cleared and #success is recorded as 1. c and f c are all constants. The position update strategy of the best particle in history depends on the value of ρ, and its value determines the size of the search range.

[0129] It should be noted that existing population-based intelligent optimization algorithms usually take the entire population or sub-population as the evolutionary unit, and design some common optimization strategies based on the elite individuals of the population or sub-population. The search direction of the particle population is updated by sharing information, that is, there is a master-slave information exchange. In its algorithm framework, the particle swarm algorithm can enhance the ability to avoid local optimality through independent learning of particles and information exchange between multiple sub-groups, but in the search process, we still need to be vigilant about a potential risk. When a particle moves towards the position of the best particle in its own population and other sub-groups, if the current position of the particle happens to coincide with the global optimal position, that is, Then its speed update will depend only on If the particle's previous velocity is infinitely close to zero, once the particle reaches the current global optimal position, the particle swarm will be trapped in this optimal point and lose the ability to continue exploring in the solution space, which leads to premature convergence of the algorithm, that is, premature locking on a local optimal solution, and unable to further find the global optimal solution. The present invention preferably ensures that the convergence particle swarm algorithm optimizes the parameters of the variational mode decomposition. The algorithm can adaptively adjust the values ​​of K and α according to the signal characteristics, thereby solving the shortcoming that the parameters in the variational mode decomposition algorithm depend on human subjective determination.

[0130] S600: Decomposing the charging and discharging power signals of the hybrid energy storage system using the optimal parameter combination of the variational mode decomposition algorithm to obtain signal components, determining the demarcation point of power allocation using mutual information entropy, and adjusting the output power of the energy storage device according to the adaptive strategy for the signal components to obtain power allocation;

[0131] It should be noted that through precise power signal decomposition, the charging and discharging processes of different energy storage devices can be managed and allocated more effectively, thereby improving the energy conversion efficiency of the entire system. By using mutual information entropy to determine the dividing point, the best time and power level for charging and discharging of different energy storage devices can be found, thereby achieving optimal power allocation.

[0132] The above is a schematic scheme of a method for optimizing injection power allocation in this embodiment. It should be noted that the technical scheme of the device for optimizing injection power allocation and the technical scheme of the above-mentioned method for optimizing injection power allocation belong to the same concept, and the details of the technical scheme of the device for optimizing injection power allocation in this embodiment that are not described in detail can all be referred to the description of the technical scheme of the above-mentioned method for optimizing injection power allocation.

[0133] The device for optimizing injection power allocation in this embodiment includes:

[0134] A decomposition module is used to obtain an operating signal of the generator, decompose the operating signal using a variational modal decomposition algorithm to obtain a modal function, calculate the modal function according to a function transformation method, obtain an analytical signal of the modal function, and obtain a unilateral spectrum of a modal component corresponding to the modal function;

[0135] An adjustment module, used for adjusting the single-sided spectrum by using a correction method, so that the single-sided spectrum is consistent with the corresponding baseband, calculating the gradient square norm of the analytical signal, and obtaining the bandwidth of the corresponding component of the modal function;

[0136] A model building module, used for building a constrained variational model based on the bandwidth, adding augmented Lagrangian coefficients to the constrained variational model to obtain an unconstrained variational model;

[0137] A model solving module, used for solving the unconstrained variational model by using an alternating direction multiplier method and the energy conservation theorem to obtain parameters of variational mode decomposition;

[0138] A parameter solving module, used for optimizing and solving the parameters of the variational modal decomposition by using an optimization algorithm to obtain an optimal parameter combination;

[0139] The distribution module is used to decompose the charging and discharging power signals of the hybrid energy storage system using the optimal parameter combination of the variational mode decomposition algorithm to obtain signal components, determine the demarcation point of power distribution using mutual information entropy, and adjust the output power of the energy storage device according to the adaptive strategy for the signal components to obtain power distribution.

[0140] This embodiment further provides a computing device, which is applicable to the case of optimizing the allocation of injected power, including:

[0141] Memory and processor; the memory is used to store computer executable instructions, and the processor is used to execute computer executable instructions to implement the injection power optimization allocation method proposed in the above embodiment.

[0142] This embodiment further provides a storage medium on which a computer program is stored. When the program is executed by a processor, the method for optimizing injection power allocation proposed in the above embodiment is implemented.

[0143] The storage medium proposed in this embodiment and the method for realizing optimized allocation of injected power proposed in the above embodiment belong to the same inventive concept. The technical details not fully described in this embodiment can be referred to the above embodiment, and this embodiment has the same beneficial effects as the above embodiment.

[0144] Through the above description of the implementation methods, the technicians in the relevant field can clearly understand that the present invention can be implemented by means of software and necessary general hardware, and of course can also be implemented by hardware, but in many cases the former is a better implementation method. Based on such an understanding, the technical solution of the present invention 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 a computer floppy disk, read-only memory (ReadOnly, Memory, ROM), random access memory (RandomAccess Memory, RAM), flash memory (FLASH), hard disk or optical disk, etc., including a number of instructions for a computer device (which can be a personal computer, server, or network device, etc.) to perform the methods of various embodiments of the present invention.

[0145] Example 2

[0146] Reference Figure 2-6 As an embodiment of the present invention, a method for optimizing allocation of injected power is provided to verify and illustrate the technical effects adopted in the method.

[0147] The guaranteed convergence particle swarm algorithm (GCPSO) and particle swarm optimization (PSO) are used to optimize the VMD parameters and obtain the results of iterative calculation. Figure 2 and Figure 3 As shown;

[0148] By comparing files Figure 2 and Figure 3 It can be seen that the GCPSO algorithm found the minimum fitness value of -0.968219 at the 9th iteration, while the PSO algorithm found the minimum fitness value of -0.967338 at the 14th iteration;

[0149] And the time required for 20 iterations of GCPSO algorithm is 728 seconds, while that of PSO algorithm is 1934 seconds. The simulation results show that GCPSO algorithm is easier to jump out of local optimum than PSO algorithm, and it is superior to PSO algorithm in solution speed and calculation accuracy.

[0150] The optimal parameter combination obtained by optimizing VMD parameters using GCPSO is [K, α] = [7, 2000], and the optimal parameter combination obtained by PSO algorithm is [K, α] = [9, 1992]. The overall charging and discharging power signal of the hybrid energy storage is decomposed by VMD with different parameter combinations, and the intrinsic mode function (IMF) obtained by decomposition is Hilbert transformed to obtain the spectrum diagram as shown in the figure. Figure 4 and Figure 5As shown in the figure, the spectrum diagram obtained by optimizing the parameter combination of the PSO algorithm shows the phenomenon of over-decomposition of the signal, and a certain degree of modal aliasing occurs, resulting in a poor signal decomposition effect.

[0151] In order to verify the superiority of the signal decomposition results of the VMD algorithm, the modal decomposition (EMD) algorithm is used to compare with the VMD algorithm. The parameters of the EMD algorithm do not need to be set in advance. The VMD algorithm uses the parameter combination of [K, α] = [7, 2000] to perform Hilbert transform on the IMF obtained by decomposition, thereby obtaining their respective spectrum diagrams.

[0152] from Figure 6 It can be seen that the decomposition results obtained by the EMD algorithm show very serious modal aliasing in each sub-signal. Figure 4 The variational mode decomposition algorithm after parameter optimization has more uniform and regular decomposition results in the decomposition of power signals, and can clearly distinguish each sub-signal, thereby verifying the superiority of the signal decomposition results of the VMD algorithm, and can more effectively complete the decomposition task of the power signal, laying the foundation for the reasonable power allocation of hybrid energy storage.

[0153] It should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention and are not intended to limit the present invention. Although the present invention has been described in detail with reference to the preferred embodiments, those skilled in the art should understand that the technical solutions of the present invention may be modified or replaced by equivalents without departing from the spirit and scope of the technical solutions of the present invention, which should all be included in the scope of the claims of the present invention.

Claims

1. A method for optimizing injection power allocation, characterized in that: include: Acquire an operating signal of the generator, decompose the operating signal using a variational modal decomposition algorithm to obtain a modal function, calculate the modal function according to a function transformation method, obtain an analytical signal of the modal function, and obtain a unilateral spectrum of a modal component corresponding to the modal function; The single-sided spectrum is adjusted by using a correction method to obtain a single-sided spectrum that is consistent with the corresponding baseband, and the square norm of the gradient of the analytical signal is calculated to obtain the bandwidth of the corresponding component of the modal function; constructing a constrained variational model based on the bandwidth, and adding augmented Lagrangian coefficients to the constrained variational model to obtain an unconstrained variational model; The unconstrained variational model is solved by using the alternating direction multiplier method and the energy conservation theorem to obtain the parameters of the variational mode decomposition; Optimizing and solving the parameters of the variational mode decomposition using an optimization algorithm to obtain an optimal parameter combination; The charging and discharging power signals of the hybrid energy storage system are decomposed using the optimal parameter combination of the variational mode decomposition algorithm to obtain signal components, and the demarcation point of power distribution is determined using mutual information entropy. For the signal components, the output power of the energy storage device is adjusted according to an adaptive strategy to obtain power distribution.

2. The injection power optimization allocation method according to claim 1, characterized in that: The method of adjusting the single-sided spectrum by using a correction method includes: The modal function is calculated using Hilbert transform, the analytical signal related to the modal function is obtained, and the estimated center frequency of the modal component corresponding to the modal function is obtained; Using the exponential correction method, the unilateral spectrum of the modal component is adjusted so that the unilateral spectrum of the modal component falls on the corresponding baseband, which is expressed as: Among them, u k (t) is the mode function, is the estimated center frequency of the modal component, j is √(-1), δ(t) is the Dirac equation, π is pi, and t is time.

3. The injection power optimization allocation method according to claim 2, characterized in that: Constructing a constrained variational model based on the bandwidth includes: By calculating the square norm of the analytical signal gradient, the bandwidth of the modal component is obtained, and the constructed constrained variational model is expressed as: Among them, k is the number of decomposed modes, {u k } is the set of all modal components after decomposition, {ω k } is the set of center frequencies of the modal components, * is the convolution symbol, and f is the original signal.

4. The injection power optimization allocation method according to claim 3, characterized in that: Adding augmented Lagrangian coefficients to the constrained variational model includes: Add Lagrange multiplication operators and quadratic penalty factors to the constrained variational model.

5. The injection power optimization allocation method according to claim 4, characterized in that: Solving the unconstrained variational model includes: The alternating direction multiplier method is combined with Parseval's theorem to solve the problem based on the equivalence of Fourier transform. According to the Hermitian symmetry property of the signal, the unconstrained variational model formula is expressed by integrating in the non-negative frequency interval to obtain the first unconstrained variational model formula. The unconstrained variational model formula is derived and expressed as: The center frequency is similarly expressed as: in, is the center frequency of the current modal component, ω is the angular frequency, and α is the quadratic penalty factor introduced in the augmented Lagrangian method.

6. The injection power optimization allocation method according to claim 1 or 5, characterized in that: The modal components include: In the process of obtaining the modal components, for each modal component, the square of the Euclidean norm of the difference between the modal component estimates obtained at the n+1th iteration and the nth iteration is calculated, divided by the square of the Euclidean norm of the modal component estimate obtained at the nth iteration, and the ratios of all modal components are summed; If the sum is less than the preset positive number precision threshold, the iteration process of the modal component has met the precision requirement and the iteration can be stopped; if the sum is greater than the preset positive number precision threshold, the iteration needs to continue.

7. The injection power optimization allocation method according to claim 6, characterized in that: Optimizing and solving the parameters of the variational mode decomposition using an optimization algorithm includes: Use the randomly generated parameter combination as the particle's initial position and initialize the velocity; Under different particle position conditions, the variational mode decomposition algorithm is used to calculate the operating signal, the fitness value of each particle is calculated, and the global optimum of the particle is recorded; Perform update operations to update the positions and velocities of the global optimal particle and other particles respectively, calculate the fitness values ​​of all particles, and update the global optimal particle; If the loop iteration reaches the maximum number of iterations, the global optimal solution and particle position are output; If the loop iteration does not reach the maximum number of iterations, the update operation continues.

8. A system for optimizing injection power allocation method, characterized in that: include: A decomposition module is used to obtain an operating signal of the generator, decompose the operating signal using a variational modal decomposition algorithm to obtain a modal function, calculate the modal function according to a function transformation method, obtain an analytical signal of the modal function, and obtain a unilateral spectrum of a modal component corresponding to the modal function; An adjustment module, used for adjusting the single-sided spectrum by using a correction method, so that the single-sided spectrum is consistent with the corresponding baseband, calculating the gradient square norm of the analytical signal, and obtaining the bandwidth of the corresponding component of the modal function; A model building module, used for building a constrained variational model based on the bandwidth, adding augmented Lagrangian coefficients to the constrained variational model to obtain an unconstrained variational model; A model solving module, used for solving the unconstrained variational model by using an alternating direction multiplier method and the energy conservation theorem to obtain parameters of variational mode decomposition; A parameter solving module, used for optimizing and solving the parameters of the variational modal decomposition by using an optimization algorithm to obtain an optimal parameter combination; The distribution module is used to decompose the charging and discharging power signals of the hybrid energy storage system using the optimal parameter combination of the variational mode decomposition algorithm to obtain signal components, determine the demarcation point of power distribution using mutual information entropy, and adjust the output power of the energy storage device according to the adaptive strategy for the signal components to obtain power distribution.

9. An electronic device, characterized in that: The device comprises: processor; a memory for storing processor-executable instructions; The processor is configured to call the instructions stored in the memory to execute the method according to any one of claims 1 to 7.

10. A computer-readable storage medium having computer program instructions stored thereon, characterized in that: When the computer program instructions are executed by a processor, the method according to any one of claims 1 to 7 is implemented.

Images