Load curve signal decomposition method and system, storage medium and equipment

Abstract

The embodiment of the invention discloses a load curve signal decomposition method, and the method comprises the steps: collecting an original load curve signal X (t), adding Gaussian white noise delta 0 omega n (t) to the original load curve signal X (t), and obtaining a load curve signal Xn (t); processing the load curve signal according to complete set multivariate variational mode decomposition to obtain a mode number and a plurality of corresponding mode components; setting a preset condition, and judging whether the modal quantity and the plurality of corresponding modal components meet the preset condition or not; and if the number of modes and the plurality of corresponding mode components meet a preset condition, taking the number of modes and the plurality of corresponding mode components as a decomposition result of the load curve. According to the method, the effective decomposition result of the load curve is obtained through the steps of decomposing the signal by using the complete set multivariate variational mode and setting the preset condition for screening and verification, so that the internal structure and characteristics of the load curve can be understood, and a data basis is provided for subsequent prediction support.

Description

Technical Field

[0001] The present invention relates to the field of signal processing technology, and in particular to a load curve signal decomposition method, system, storage medium and device. Background Art

[0002] Load curve signals often contain a large amount of noise components and nonlinear characteristics, which makes it difficult for traditional signal decomposition methods to extract useful information accurately and efficiently.

[0003] Traditional load curve signal decomposition methods, such as Fourier transform and wavelet transform, can achieve signal decomposition to a certain extent, but they are prone to lose local signal characteristics when processing nonlinear and non-stationary signals, and have poor robustness to noise, making it difficult to accurately extract modal components in load curve signals.

[0004] In addition, with the rapid development of smart grids and new energy, load curve signals have become more complex and changeable. Traditional decomposition methods can no longer meet actual needs. Therefore, a more efficient and accurate load curve signal decomposition method is needed to cope with complex and changeable load curve signals and improve the accuracy and efficiency of signal processing. Summary of the invention

[0005] Based on this, it is necessary to propose a load curve signal decomposition method to address the above problems.

[0006] A load curve signal decomposition method, the method comprising:

[0007] Collect the original load curve signal X(t) and add Gaussian white noise δ0ω to it n (t), and obtain the load curve signal X n (t);

[0008] Processing the load curve signal according to the complete set multivariate variational modal decomposition to obtain the modal number and the corresponding modal components;

[0009] Setting a preset condition, and determining whether the modal quantity and the corresponding modal components meet the preset condition;

[0010] If the modal quantity and the corresponding modal components meet a preset condition, the modal quantity and the corresponding modal components are used as the decomposition result of the load curve.

[0011] In the above scheme, the load curve signal is:

[0012] X n (t) = X(t) + δ0ω n (t)

[0013] Among them, X(t) is the original load curve signal, δ0ω n (t) is the added Gaussian white noise, X n (t) is the load curve signal;

[0014] The constraint expression corresponding to the load curve signal is:

[0015]

[0016] The augmented Lagrangian expression corresponding to the load curve signal is:

[0017]

[0018] Among them, α is the penalty factor; λ c is the Lagrangian multiplication operator; L() is the augmented Lagrangian function; is the analytical modulation signal with channel number c, mode number k, and n-dimensional white noise added; u k,c (t) is the kth mode of the cth channel; ω k is the center frequency of the kth mode; x c (t)+δ0ω n (t) is the signal after adding white noise.

[0019] In the above scheme, before the load curve signal is processed according to the complete set multivariate variational modal decomposition to obtain the modal number and the corresponding modal components, the method further includes:

[0020] According to the load curve signal X n (t) determine abnormal noise points and remove them;

[0021] Get the load curve signal X n The missing values ​​in (t) are filled according to the trend of the original load curve signal X(t).

[0022] In the above scheme, the load curve signal is processed according to the complete set multivariate variational modal decomposition to obtain the modal number and the corresponding modal components, specifically including:

[0023] The load curve signal X n (t) performing EMD decomposition to obtain a number of first intrinsic mode function IMFs sets;

[0024] Averaging the first intrinsic mode function IMFs set to obtain a first modal component;

[0025] Determining a first residual signal between the preliminary modal component and the original load curve signal X(t);

[0026] Determining whether the first residual signal satisfies a threshold range;

[0027] If satisfied, output the first modal component;

[0028] Constructing a multivariate variational modal decomposition framework, wherein the multivariate variational modal decomposition framework includes a constraint expression corresponding to the load curve signal and an augmented Lagrangian expression corresponding to the load curve signal;

[0029] Iteratively optimizing the first modal component according to the augmented Lagrangian expression corresponding to the load curve signal to obtain a modal component that satisfies the constraint expression;

[0030] Output the number of modes that satisfy the constraint expression and the corresponding modal components.

[0031] In the above scheme, the setting of preset conditions and judging whether the number of modes and the corresponding modal components meet the preset conditions specifically include:

[0032] The preset conditions include loss function e, permutation entropy pe, modal aliasing analysis and center frequency interval size;

[0033] The loss function e is:

[0034]

[0035] Among them, X n (t) is the load curve signal, and u(t) is the multivariable modulated oscillation signal.

[0036] In the above scheme, the setting of preset conditions and judging whether the number of modes and the corresponding modal components meet the preset conditions specifically include:

[0037] Determining a multivariable modulated oscillation signal u(t) according to the modal quantity and the corresponding plurality of modal components;

[0038] Determine the multivariable modulated oscillation signal u(t) and the load curve signal X n (t) determining the loss function value;

[0039] When the loss function value satisfies the loss range, determining the permutation entropy pe of the multivariate variational mode decomposition framework;

[0040] When the permutation entropy pe satisfies a threshold range, determining whether modal mixing occurs in the plurality of modal components;

[0041] When the modal components do not have modal mixing, determining the center frequency intervals of adjacent modal functions in the modal components;

[0042] When the heart rate interval satisfies the interval range, the plurality of modal quantities and the corresponding plurality of modal components are used as decomposition results of the load curve.

[0043] The above scheme also includes: when the loss function value does not meet the loss range, when the several modal components have modal mixing and when the heart rate interval does not meet the interval range, adjusting the modal number K of the multivariate variational modal decomposition framework, and re-determining whether the modal number and the corresponding several modal components meet the preset conditions.

[0044] The present application also proposes a load curve signal decomposition system, the system comprising: a signal acquisition unit, a signal processing unit, a judgment unit and a signal output unit;

[0045] The signal acquisition unit is used to collect the original load curve signal X(t) and add Gaussian white noise δ0ω to it. n (t), and obtain the load curve signal X n (t);

[0046] The signal processing unit is used to process the load curve signal according to the complete set multivariate variational modal decomposition to obtain the modal number and the corresponding modal components;

[0047] The judging unit is used to set a preset condition and judge whether the modal quantity and the corresponding modal components meet the preset condition;

[0048] The signal output unit is used to use the modal numbers and the corresponding modal components as decomposition results of the load curve if the modal numbers and the corresponding modal components meet preset conditions.

[0049] The present application also proposes a readable storage medium storing a computer program. When the computer program is executed by a processor, the processor executes the following steps:

[0050] Collect the original load curve signal X(t) and add Gaussian white noise δ0ω to it n (t), and obtain the load curve signal X n (t);

[0051] Processing the load curve signal according to the complete set multivariate variational modal decomposition to obtain the modal number and the corresponding modal components;

[0052] Setting a preset condition, and determining whether the modal quantity and the corresponding modal components meet the preset condition;

[0053] If the modal quantity and the corresponding modal components meet a preset condition, the modal quantity and the corresponding modal components are used as the decomposition result of the load curve.

[0054] The present application also proposes a computer device, comprising a memory and a processor, wherein the memory stores a computer program, and the processor executes the following steps of the computer program:

[0055] Collect the original load curve signal X(t) and add Gaussian white noise δ0ω to it n (t), and obtain the load curve signal X n (t);

[0056] Processing the load curve signal according to the complete set multivariate variational modal decomposition to obtain the modal number and the corresponding modal components;

[0057] Setting a preset condition, and determining whether the modal quantity and the corresponding modal components meet the preset condition;

[0058] If the modal quantity and the corresponding modal components meet a preset condition, the modal quantity and the corresponding modal components are used as the decomposition result of the load curve.

[0059] The embodiment of the present invention has the following beneficial effects: firstly, the original load curve signal X(t) is collected and Gaussian white noise δ0ω is added to it. n (t), and obtain the load curve signal X n (t); Process the load curve signal according to the complete set multivariate variational modal decomposition to obtain the number of modes and the corresponding modal components; Set preset conditions and determine whether the number of modes and the corresponding modal components meet the preset conditions; If the number of modes and the corresponding modal components meet the preset conditions, then use the number of modes and the corresponding modal components as the decomposition results of the load curve. This scheme aims to obtain effective decomposition results of the load curve by collecting the original load curve signal and adding Gaussian white noise, applying the complete set multivariate variational modal decomposition, setting preset conditions for screening and verification, etc. This not only helps to deeply understand the inherent structure and characteristics of the load curve, but also provides a reliable data basis for subsequent analysis, prediction and decision support. BRIEF DESCRIPTION OF THE DRAWINGS

[0060] In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings required for use in the embodiments or the description of the prior art will be briefly introduced below. 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 paying creative work.

[0061] in:

[0062] Figure 1 FIG. 1 is a flow chart of a method for decomposing a load curve signal in an embodiment. DETAILED DESCRIPTION

[0063] The technical solutions in the embodiments of the present invention will be clearly and completely described below in conjunction with the drawings in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, rather than all the embodiments; based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without making creative work are within the scope of protection of the present invention.

[0064] In the following description, a large number of specific details are given in order to provide a more thorough understanding of the present invention; however, it is obvious to those skilled in the art that the present invention can be implemented without one or more of these details; in other examples, in order to avoid confusion with the present invention, some technical features known in the art are not described, and it should be understood that the present invention can be implemented in different forms and should not be interpreted as limited to the embodiments presented here; on the contrary, providing these embodiments will make the disclosure thorough and complete and fully convey the scope of the present invention to those skilled in the art.

[0065] The purpose of the terms used herein is only to describe specific embodiments and is not intended to be limiting of the present invention. When used herein, the singular forms "a", "an" and "said / the" are also intended to include the plural forms, unless the context clearly indicates otherwise, and it should also be understood that the terms "consisting of" and / or "comprising", when used in this specification, determine the presence of the features, integers, steps, operations, elements and / or parts, but do not exclude the presence or addition of one or more other features, integers, steps, operations, elements, parts and / or groups. When used herein, the term "and / or" includes any and all combinations of the relevant listed items.

[0066] Variational mode decomposition (VMD) describes the decomposition of univariate data. When using multivariate variational mode decomposition (MVMD), its main goal is to extract K predetermined multivariate modulated oscillation signals u from the input data X(t) containing C data channels. k (t).

[0067]

[0068] The input data X(t) under multiple input channels is a multivariate signal, X(t) = [x1(t), x2(t), ..., x C (t)],U k (t)=[u1(t),u2(t),...,u C (t)], its goal will also become to extract the multivariate modulation signal so that the sum of the bandwidths of the same center frequency between the multidimensional modal signals is minimized. The constraint expression of the multivariate variational modal decomposition is shown in formula (2). By calculating the L of the given demodulated signal 2 The constrained variational model is established by using the squared norm, and the mathematical model of the convex optimization problem is shown in formula (3).

[0069]

[0070] The Lagrange multiplication operator is introduced to transform the constrained variational problem into an unconstrained variational problem. Two penalty terms are added: one is the quadratic term α used to enhance the fidelity coefficient of signal reconstruction, and the other is the Lagrange multiplier λ to ensure that the constraints are strictly met. The transformed formula (3) is as follows:

[0071]

[0072] On this basis, this application proposes the ANCEMVMD (Complete Set Multivariate Variational Mode) decomposition method, which adds Gaussian white noise δ0w to the original nonlinear data X(t). n (t), and obtain the load curve signal X n (t); Processing the load curve signal according to the complete set multivariate variational modal decomposition to obtain the modal number and the corresponding modal components; Setting a preset condition, and determining whether the modal number and the corresponding modal components meet the preset condition; If the modal number and the corresponding modal components meet the preset condition, taking the modal number and the corresponding modal components as the decomposition result of the load curve.

[0073] In order to thoroughly understand the present invention, a detailed structure will be presented in the following description to illustrate the technical solution proposed by the present invention; optional embodiments of the present invention are described in detail as follows, but in addition to these detailed descriptions, the present invention may also have other implementation methods.

[0074] like Figure 1 As shown, in one embodiment, a load curve signal decomposition method is provided, and the load curve signal decomposition method includes steps S101 to S104, which are described in detail as follows:

[0075] S101, collect the original load curve signal X(t) and add Gaussian white noise δ0ω to it n (t), and obtain the load curve signal X n (t);

[0076] This step makes the load curve signal closer to real-world data by simulating noise interference in actual situations. Gaussian white noise is a commonly used noise model, which assumes that the energy of noise at each frequency is uniform, which helps to test the performance of subsequent signal processing algorithms in noisy environments. By adding noise, the robustness and stability of the algorithm under complex and non-ideal conditions can be evaluated, providing a more meaningful test environment for practical applications.

[0077] In some embodiments, the load curve signal is:

[0078] X n (t) = X(t) + δ0ω n (t)(4)

[0079] Among them, X(t) is the original load curve signal, δ0ω n (t) is the added Gaussian white noise, X n (t) is the load curve signal;

[0080] The constraint expression corresponding to the load curve signal is:

[0081]

[0082] The augmented Lagrangian expression corresponding to the load curve signal is:

[0083]

[0084] Among them, α is the penalty factor; λ c is the Lagrangian multiplication operator; L() is the augmented Lagrangian function; is the analytical modulation signal with channel number c, mode number k, and n-dimensional white noise added; u k,c (t) is the kth mode of the cth channel; ω k is the center frequency of the kth mode; x c (t)+δ0ω n (t) is the signal after adding white noise.

[0085] It can be seen that when white noise is added to MVMD, its constraint expression and the corresponding augmented Lagrangian expression will also be updated accordingly. The updated expressions are shown in formula (4) and formula (5).

[0086] S102, processing the load curve signal according to the complete set multivariate variational modal decomposition to obtain the modal number and a number of corresponding modal components;

[0087] The complete set multivariate variational modal decomposition proposed in this application is a signal processing technology that aims to decompose complex signals into multiple simple modal components. This method can effectively extract different frequency components in the signal and help understand the intrinsic structure and characteristics of the signal. By decomposing the load curve signal, more detailed and in-depth signal characteristics can be obtained, providing a basis for subsequent analysis and prediction.

[0088] In some embodiments, before processing the load curve signal according to the complete set multivariate variational modal decomposition to obtain the modal number and the corresponding modal components, the method further includes:

[0089] According to the load curve signal X n (t) determine abnormal noise points and remove them;

[0090] Get load curve signal X n The missing values ​​in (t) are filled according to the trend of the original load curve signal X(t).

[0091] In some embodiments, the load curve signal is processed according to the complete set multivariate variational modal decomposition to obtain the modal number and the corresponding modal components, specifically including:

[0092] Load curve signal X n (t) performing EMD decomposition to obtain a number of first intrinsic mode function IMFs sets;

[0093] Averaging the first intrinsic mode function IMFs set to obtain the first mode component;

[0094] Determining a first residual signal between the preliminary modal component and the original load curve signal X(t);

[0095] Determining whether the first residual signal satisfies a threshold range;

[0096] If satisfied, output the first modal component;

[0097] Construct a multivariate variational mode decomposition framework, which includes constraint expressions corresponding to load curve signals and augmented Lagrangian expressions corresponding to load curve signals;

[0098] The first modal component is iteratively optimized according to the augmented Lagrangian expression corresponding to the load curve signal to obtain the modal component that satisfies the constraint expression;

[0099] Output the number of modes that satisfy the constraint expression and the corresponding modal components.

[0100] Preferably, in order to transform the complex optimization problem into multiple simple sub-optimization problems, the alternating direction multiplier method is introduced to update the parameters Find the optimal solution of the expression The equivalent optimization expression of is shown in formula (7):

[0101]

[0102] Transform the above iterative update problem into the frequency domain, as shown in formula (8):

[0103]

[0104] Simplify the calculation of formula (8), and finally solve the modal component The update formula (9) in the frequency domain is as follows:

[0105]

[0106] The next step is the corresponding frequency domain update optimization problem, considering that the last two terms of the Lagrangian formula (6) do not depend on ω k , the formula can be simplified to (10). The Plancherel theorem is used to connect the inner products of the functions in the time domain and the frequency domain, and the frequency domain equivalent transformation formula (11) is obtained. The value of the quadratic function is minimized by setting the first-order derivative of the quadratic function to zero. Then, simple algebraic operations are performed to obtain the following center frequency update relationship (12):

[0107]

[0108] Finally, the update of the Lagrangian operator is shown in formula (13):

[0109]

[0110] Where: the superscript “^” indicates the analytical form of the corresponding variable. w is the frequency; n is the number of iterations; is the original signal in the frequency domain of channel c; is the original signal in the frequency domain of channel c after adding white noise; is the frequency domain signal of the ith modal component of channel c; τ is the fidelity coefficient between the decomposed mode and the original signal.

[0111] S103, setting a preset condition, and determining whether the number of modes and the corresponding modal components meet the preset condition;

[0112] By screening and verifying the decomposition results, we ensure that the modal components and modal numbers obtained meet specific requirements or standards. The preset conditions may involve parameters such as the amplitude, frequency range, and correlation of the modal to ensure the accuracy and reliability of the decomposition results. By screening the preset conditions, we can exclude the modal components that do not meet the requirements and improve the accuracy of subsequent analysis and prediction. At the same time, this step also helps to discover potential problems and anomalies, providing a basis for further optimization and improvement.

[0113] In some embodiments, setting a preset condition and determining whether the number of modes and the corresponding modal components meet the preset condition specifically includes:

[0114] The preset conditions include loss function e, permutation entropy pe, modal aliasing analysis and center frequency interval size;

[0115] The loss function e is:

[0116]

[0117] Among them, X n (t) is the load curve signal, and u(t) is the multivariable modulated oscillation signal.

[0118] In some embodiments, setting a preset condition and determining whether the number of modes and the corresponding modal components meet the preset condition specifically includes:

[0119] Determine the multivariable modulated oscillation signal u(t) according to the number of modes and the corresponding several modal components;

[0120] Determine the multivariable modulated oscillation signal u(t) and the load curve signal X n (t) determining the loss function value;

[0121] When the loss function value satisfies the loss range, the permutation entropy pe of the multivariate variational mode decomposition framework is determined;

[0122] When the permutation entropy pe meets the threshold range, it is determined whether modal mixing occurs in several modal components;

[0123] When no modal mixing occurs in several modal components, the center frequency intervals of adjacent modal functions in the several modal components are determined;

[0124] When the center frequency interval meets the interval range, several modal numbers and corresponding several modal components are taken as the decomposition results of the load curve.

[0125] In some embodiments, it also includes:

[0126] When the loss function value does not meet the loss range, when several modal components have modal mixing and the center frequency interval does not meet the interval range, adjust the mode number K of the multivariate variational modal decomposition framework, and re-judge whether the mode number and the corresponding modal components meet the preset conditions.

[0127] Specifically, in formula (7), the ratio of the square of the 2-norm of the residual signal to the original signal is defined as the loss coefficient e, and the permutation entropy is introduced as the main indicator for updating the number of modes K. When the loss coefficient is less than a certain threshold μ1 (μ1=0.01), the reconstructed signal is closer to the original signal. Otherwise, we can try to increase the number of modes K to reduce the difference between the reconstructed signal and the original signal.

[0128] Specifically, when the regularity of the signal's time series is better, the complexity is lower and the permutation entropy pe is smaller; on the contrary, when the time series is more random, the unpredictability is higher and the permutation entropy pe is larger; when the normalized pe is less than the threshold μ2 (μ2=0.4), all test functions contain only a small amount of noise or no noise, and when the normalized pe is greater than the threshold μ2, the number of modes K is increased.

[0129] Furthermore, for the modal function identified by the permutation entropy pe, it is necessary to further determine whether modal mixing occurs inside it. This can be determined by the number and amplitude of the extreme points of the modal function. When the ratio r is greater than the threshold μ3 (μ3 = 0.2), it is considered that mode mixing occurs inside the modal function. In this case, the mode number K needs to be increased to re-run MVMD.

[0130] On the other hand, if the ratio r is less than the threshold μ3, the modal function is regarded as an eigenmodal function mixed with a small amount of noise. At this time, the threshold condition is met, and several modal numbers and corresponding modal components are taken as the decomposition results of the load curve.

[0131] S104: If the number of modes and the corresponding modal components meet preset conditions, the number of modes and the corresponding modal components are used as decomposition results of the load curve.

[0132] If the number of modes and modal components meet the preset conditions, then these results can be regarded as valid decomposition results of the load curve and can be output as decomposition results for subsequent analysis, prediction or decision support. Confirming the validity of the decomposition results is a key step to ensure the accuracy of the entire signal processing process. By outputting decomposition results that meet the conditions, reliable data support can be provided for subsequent research and applications.

[0133] In summary, if the output decomposition result meets all the preset conditions, the decomposition result is considered ideal and the decomposition process can be stopped; if some of the preset conditions are not met: adjustments need to be made according to the unsatisfied conditions, for example:

[0134] If the loss function value is too large, the number of modes K can be increased to better approximate the original signal;

[0135] If the permutation entropy is too large, adjust the parameters of MVMD, such as the penalty factor α or the tolerance ε, to improve the decomposition result;

[0136] If there is modal aliasing, increase the number of modes K to separate the aliased frequency components;

[0137] If the center frequency interval is too small, the number of modes K can be reduced to avoid over-decomposition;

[0138] Repeat the above judgment steps until all preset conditions are met or the maximum number of iterations is reached.

[0139] Among them, the maximum number of iterations can be set according to the actual application scenario.

[0140] The present application also proposes a load curve signal decomposition system, the system comprising: a signal acquisition unit, a signal processing unit, a judgment unit and a signal output unit;

[0141] Signal acquisition unit, used to collect the original load curve signal X(t) and add Gaussian white noise δ0ω to it n (t), and obtain the load curve signal X n (t);

[0142] A signal processing unit, used for processing the load curve signal according to the complete set multivariate variational modal decomposition to obtain the modal number and the corresponding modal components;

[0143] A judgment unit, used to set a preset condition and judge whether the number of modes and the corresponding modal components meet the preset condition;

[0144] The signal output unit is used to use the modal numbers and the corresponding modal components as decomposition results of the load curve if the modal numbers and the corresponding modal components meet preset conditions.

[0145] The present application also proposes a readable storage medium storing a computer program. When the computer program is executed by a processor, the processor executes the following steps:

[0146] Collect the original load curve signal X(t) and add Gaussian white noise δ0ω to it n (t), and obtain the load curve signal X n (t);

[0147] The load curve signal is processed according to the complete set multivariate variational modal decomposition to obtain the modal number and the corresponding modal components;

[0148] Set preset conditions and determine whether the number of modes and the corresponding modal components meet the preset conditions;

[0149] If the number of modes and the corresponding modal components meet preset conditions, the number of modes and the corresponding modal components are used as the decomposition results of the load curve.

[0150] The present application also proposes a computer device, including a memory and a processor, wherein the memory stores a computer program, and the computer program is executed by the processor to perform the following steps:

[0151] Collect the original load curve signal X(t) and add Gaussian white noise δ0ω to it n (t), and obtain the load curve signal X n (t);

[0152] The load curve signal is processed according to the complete set multivariate variational modal decomposition to obtain the modal number and the corresponding modal components;

[0153] Set preset conditions and determine whether the number of modes and the corresponding modal components meet the preset conditions;

[0154] If the number of modes and the corresponding modal components meet preset conditions, the number of modes and the corresponding modal components are used as the decomposition results of the load curve.

[0155] Those skilled in the art will understand that all or part of the processes in the above-mentioned embodiments can be completed by instructing the relevant hardware through a computer program, and the program can be stored in a non-volatile computer-readable storage medium. When the program is executed, it can include the processes of the embodiments of the above-mentioned methods. Among them, any reference to memory, storage, database or other media used in the embodiments provided in the present application may include non-volatile and / or volatile memory. Non-volatile memory may include read-only memory (ROM), programmable ROM (PROM), electrically programmable ROM (EPROM), electrically erasable programmable ROM (EEPROM) or flash memory. Volatile memory may include random access memory (RAM) or external cache memory. As an illustration and not limitation, RAM is available in various forms, such as static RAM (SRAM), dynamic RAM (DRAM), synchronous DRAM (SDRAM), double data rate SDRAM (DDRSDRAM), enhanced SDRAM (ESDRAM), synchronous link (Synchlink) DRAM (SLDRAM), memory bus (Rambus) direct RAM (RDRAM), direct memory bus dynamic RAM (DRDRAM), and memory bus dynamic RAM (RDRAM).

[0156] The technical features of the above embodiments can be combined arbitrarily. To make the description concise, not all possible combinations of the technical features in the above embodiments are described. However, as long as there is no contradiction in the combination of these technical features, they should be considered to be within the scope of this specification.

[0157] The above-mentioned embodiments only express several implementation methods of the present application, and the descriptions thereof are relatively specific and detailed, but they cannot be understood as limiting the scope of the patent of the present application. It should be pointed out that, for ordinary technicians in this field, several modifications and improvements can be made without departing from the concept of the present application, and these all belong to the protection scope of the present application. The above disclosure is only a preferred embodiment of the present invention, and of course it cannot be used to limit the scope of the rights of the present invention. Therefore, equivalent changes made according to the claims of the present invention are still within the scope covered by the present invention.

Claims

1. A load curve signal decomposition method, characterized in that: The method comprises: Collect the original load curve signal X(t) and add Gaussian white noise δ0ω to it n (t), and obtain the load curve signal X n (t); Processing the load curve signal according to the complete set multivariate variational modal decomposition to obtain the modal number and the corresponding modal components; Setting a preset condition, and determining whether the modal quantity and the corresponding modal components meet the preset condition; If the modal quantity and the corresponding modal components meet a preset condition, the modal quantity and the corresponding modal components are used as the decomposition result of the load curve.

2. The load curve signal decomposition method according to claim 1, characterized in that: The load curve signal is: X n (t)=X(t)+δ0ω n (t) Among them, X(t) is the original load curve signal, δ0ω n (t) is the added Gaussian white noise, X n (t) is the load curve signal; The constraint expression corresponding to the load curve signal is: The augmented Lagrangian expression corresponding to the load curve signal is: Among them, α is the penalty factor; λ c is the Lagrangian multiplication operator; L() is the augmented Lagrangian function; is the analytical modulation signal with channel number c, mode number k, and n-dimensional white noise added; u k,c (t) is the kth mode of the cth channel; ω k is the center frequency of the kth mode; x c (t)+δ0ω n (t) is the signal after adding white noise.

3. The load curve signal decomposition method according to claim 1, characterized in that: Before the load curve signal is processed according to the complete set multivariate variational modal decomposition to obtain the modal quantity and the corresponding modal components, the method further includes: According to the load curve signal X n (t) determine abnormal noise points and remove them; Get the load curve signal X n The missing values ​​in (t) are filled according to the trend of the original load curve signal X(t).

4. The load curve signal decomposition method according to claim 2, characterized in that: The process of processing the load curve signal according to the complete set multivariate variational modal decomposition to obtain the modal quantity and the corresponding modal components specifically includes: The load curve signal X n (t) performing EMD decomposition to obtain a number of first intrinsic mode function IMFs sets; Averaging the first intrinsic mode function IMFs set to obtain a first modal component; Determining a first residual signal between the preliminary modal component and the original load curve signal X(t); Determining whether the first residual signal satisfies a threshold range; If satisfied, output the first modal component; Constructing a multivariate variational modal decomposition framework, wherein the multivariate variational modal decomposition framework includes a constraint expression corresponding to the load curve signal and an augmented Lagrangian expression corresponding to the load curve signal; Iteratively optimizing the first modal component according to the augmented Lagrangian expression corresponding to the load curve signal to obtain a modal component that satisfies the constraint expression; Output the number of modes that satisfy the constraint expression and the corresponding modal components.

5. The load curve signal decomposition method according to claim 1, characterized in that: The setting of preset conditions and judging whether the number of modes and the corresponding modal components meet the preset conditions specifically includes: The preset conditions include loss function e, permutation entropy pe, modal aliasing analysis and center frequency interval size; The loss function e is: Among them, X n (t) is the load curve signal, and u(t) is the multivariable modulated oscillation signal.

6. The load curve signal decomposition method according to claim 1, characterized in that: The setting of the preset conditions and judging whether the number of modes and the corresponding modal components meet the preset conditions specifically includes: Determining a multivariable modulated oscillation signal u(t) according to the modal quantity and the corresponding plurality of modal components; Determine the multivariable modulated oscillation signal u(t) and the load curve signal X n (t) determining the loss function value; When the loss function value satisfies the loss range, determining the permutation entropy pe of the multivariate variational mode decomposition framework; When the permutation entropy pe satisfies a threshold range, determining whether modal mixing occurs in the plurality of modal components; When the modal components do not have modal mixing, determining the center frequency intervals of adjacent modal functions in the modal components; When the heart rate interval satisfies the interval range, the plurality of modal quantities and the corresponding plurality of modal components are used as decomposition results of the load curve.

7. The load curve signal decomposition method according to claim 6, characterized in that: Also includes: When the loss function value does not meet the loss range, when modal mixing occurs among the several modal components, and when the heart rate interval does not meet the interval range, the modal number K of the multivariate variational modal decomposition framework is adjusted, and it is re-determined whether the modal number and the corresponding several modal components meet the preset conditions.

8. A load curve signal decomposition system, characterized in that: The system comprises: a signal acquisition unit, a signal processing unit, a judgment unit and a signal output unit; The signal acquisition unit is used to collect the original load curve signal X(t) and add Gaussian white noise δ0ω to it. n (t), and obtain the load curve signal X n (t); The signal processing unit is used to process the load curve signal according to the complete set multivariate variational modal decomposition to obtain the modal number and the corresponding modal components; The judging unit is used to set a preset condition and judge whether the modal quantity and the corresponding modal components meet the preset condition; The signal output unit is used to use the modal numbers and the corresponding modal components as decomposition results of the load curve if the modal numbers and the corresponding modal components meet preset conditions.

9. A readable storage medium storing a computer program, wherein when the computer program is executed by a processor, the processor is caused to execute the steps of the method according to any one of claims 1 to 7.

10. A computer device comprising a memory and a processor, wherein the memory stores a computer program, and when the computer program is executed by the processor, the processor executes the steps of the method according to any one of claims 1 to 7.

Images