Energy storage frequency modulation instruction prediction method and system for assisting asymptote

Abstract

An energy storage frequency modulation instruction prediction method for assisting an asymptote comprises the following steps: acquiring a frequency modulation instruction signal; based on a preset algorithm, decomposing the frequency modulation instruction signal into a plurality of modal subsequences of which the number corresponds to that of decomposition layers; generating corresponding upper limit asymptotes and lower limit asymptotes for the plurality of modal subsequences; performing sequence decomposition on the plurality of modal subsequences by using corresponding upper limit asymptotes and lower limit asymptotes to generate a core sequence and an auxiliary sequence; predicting the plurality of core sequences and the plurality of auxiliary sequences by adopting a neural network to obtain a final prediction result; applying the final prediction result to the energy storage equipment for frequency modulation response; the prediction result of the frequency modulation instruction is more accurate, the size of the frequency modulation instruction is predicted in advance for energy storage regulation, and the response accuracy and the frequency modulation income are improved.

Description

Technical Field

[0001] The present application relates to the technical field of power grid frequency regulation, and in particular to a method and system for predicting energy storage frequency regulation instructions for auxiliary asymptotes. Background Art

[0002] At present, in major regional power grids in my country, large hydropower and thermal power units (coal-fired units / gas-fired units) are mainly used as power grid frequency modulation power sources. The frequency modulation power output is adjusted to respond to changes in system frequency. Hybrid energy storage frequency modulation instruction prediction refers to the establishment of a prediction model through the analysis of historical frequency modulation instructions and related data to predict future frequency modulation instruction requirements, thereby optimizing the control strategy of the hybrid energy storage system and improving its frequency modulation performance and efficiency. It can improve system performance and respond quickly. It can quickly adjust the output power in a short time, respond quickly to changes in grid frequency, effectively balance the supply and demand differences in the power system, and maintain grid frequency stability. Improve the regulation accuracy, be able to track frequency modulation instructions more accurately, reduce regulation errors, and improve the power supply quality and stability of the power grid. Enhance system stability. By reasonably allocating the power and energy of different energy storage devices, the overall risk of the system can be reduced, the reliability and stability of the system can be improved, and the risk of system failure caused by failure or performance degradation of a single energy storage device can be reduced. Extend equipment life and optimize charging and discharging strategies. According to the prediction results, a reasonable charging and discharging strategy can be formulated to avoid excessive or frequent charging and discharging of energy storage equipment, thereby extending its service life. Balance the use of equipment and reasonably allocate the use frequency and load of different energy storage equipment to make the aging degree of each equipment relatively balanced, reducing the maintenance cost and replacement frequency of equipment. Reduce costs and improve economic benefits. By improving frequency regulation performance and efficiency, the hybrid energy storage system can obtain more frequency regulation benefits, while reducing power outage losses and equipment damage costs caused by unstable grid frequency. Optimize investment costs. According to the prediction results, the capacity and quantity of different energy storage equipment can be reasonably configured to avoid over-investment and waste of resources and improve the return on investment.

[0003] The current traditional prediction methods (such as Figure 2 ), using methods such as VMD decomposition, the original frequency modulation instruction sequence is divided into a series of subsequences, which are then predicted separately and the predicted results are finally superimposed. The existing VMD decomposition method has shortcomings, and may decompose subsequences that are more difficult to predict. The nonlinearity of some subsequences is even higher than that of the original instruction sequence, which makes prediction more difficult. Inaccurate prediction results will cause certain response deviations, affect the accuracy of system response, and thus affect the frequency modulation benefits of power plants. Summary of the invention

[0004] In view of the deficiencies of the prior art, the present invention proposes a prediction method based on the asymptote method. First, the upper and lower asymptotes of each subsequence are made, and the parts of each subsequence that exceed the asymptote are superimposed together to form a new sequence for prediction, and the final result is obtained.

[0005] An energy storage frequency modulation command prediction method with auxiliary asymptotes, characterized by including, S101. Obtain a frequency modulation command signal; S102. Decompose the frequency modulation command signal into a plurality of modal subsequences corresponding to the number of decomposition layers based on a preset algorithm; S103. Generate corresponding upper asymptotes and lower asymptotes for a plurality of modal subsequences; S104. Perform sequence decomposition on a plurality of modal subsequences using the corresponding upper asymptotes and lower asymptotes to generate a core sequence and an auxiliary sequence; S105. Predict a plurality of core sequences and a plurality of auxiliary sequences using a neural network to obtain a final prediction result; apply the final prediction result to an energy storage device for frequency modulation response.

[0006] There is also provided an energy storage frequency modulation command prediction system with auxiliary asymptotes, characterized by including, A command acquisition module for acquiring a frequency modulation command signal; A command decomposition module for decomposing the frequency modulation command signal into a plurality of modal subsequences corresponding to the number of decomposition layers based on a preset algorithm; An asymptote generation module for generating corresponding upper asymptotes and lower asymptotes for a plurality of modal subsequences; A subsequence decomposition module for performing sequence decomposition on a plurality of modal subsequences using the corresponding upper asymptotes and lower asymptotes to generate a core sequence and an auxiliary sequence; A prediction processing module for predicting a plurality of core sequences and a plurality of auxiliary sequences using a neural network to obtain a final prediction result; applying the final prediction result to an energy storage device for frequency modulation response.

[0007] The beneficial effects of the present invention are: Reduce the non-linearity degree of the subsequences after VMD decomposition, make each subsequence of VMD smoother, greatly reduce the prediction difficulty, improve the accuracy of frequency modulation command prediction, and thus improve the response accuracy and frequency modulation benefit of the power plant. Description of the Drawings

[0008] Figure 1 It is a method step diagram of this application.

[0009] Figure 2 It is a schematic diagram of a traditional algorithm.

[0010] Figure 3 It is a schematic diagram of a data processing coordinate system.

[0011] Figure 4 It is a schematic diagram for generating asymptotes. Specific implementation manners

[0012] Next, the technical solutions in the embodiments of the present application will be clearly and completely described in conjunction with the accompanying drawings in the embodiments of the present application. Obviously, the described embodiments are only a part of the embodiments of the present application, rather than all the embodiments. Based on the embodiments in the present application, all other embodiments obtained by those of ordinary skill in the art without making creative efforts shall fall within the protection scope of the present application.

[0013] It should be understood that the step numbers used in the text are only for convenient description and do not limit the execution order of the steps.

[0014] It should be understood that the terms used in the specification of the present application are only for the purpose of describing specific embodiments and are not intended to limit the present application. As used in the specification of the present application and the appended claims, unless otherwise clearly specified in the context, the singular forms "a", "an" and "the" are intended to include the plural forms.

[0015] The terms "include" and "comprise" indicate the presence of the described features, wholes, steps, operations, elements and / or components, but do not exclude the presence or addition of one or more other features, wholes, steps, operations, elements, components and / or their combinations.

[0016] The term "and / or" refers to any combination and all possible combinations of one or more of the associated listed items, and includes these combinations.

[0017] The embodiment of the present disclosure provides a method for predicting a frequency modulation command of an auxiliary asymptote energy storage, as Figure 1 shown, including: S101. Obtain a frequency modulation command signal; S102. Decompose the frequency modulation command signal into a plurality of modal subsequences corresponding to the number of decomposition layers based on a preset algorithm; S103. Generate corresponding upper asymptotes and lower asymptotes for the plurality of modal subsequences; S104. Decompose the plurality of modal subsequences using the corresponding upper asymptotes and lower asymptotes to generate a core sequence and an auxiliary sequence; S105. Predict the plurality of core sequences using a neural network, and predict the plurality of auxiliary sequences after hybrid superposition using a neural network to obtain a final prediction result; apply the final prediction result to an energy storage device for frequency modulation response.

[0018] The frequency modulation command prediction method provided by the embodiments of the present disclosure uses a new type of supercapacitor and a lithium battery as energy storage devices to perform frequency modulation when the grid frequency in a power plant fluctuates. The state of the power plant grid is monitored in real time. When the power supply frequency in the power plant grid fluctuates, a corresponding frequency modulation command signal is generated. After obtaining the original signal, the power plant grid is frequency modulated by predicting future signals.

[0019] According to the prediction results, the output of the power source is adjusted in real time, such as the active power distribution of the energy storage unit under different working conditions, and the operation between each unit is coordinated to suppress power fluctuations and improve the frequency modulation performance of the power grid.

[0020] In another embodiment provided by the present disclosure, in the above step “S102. Decompose the frequency modulation command signal into a plurality of modal subsequences corresponding to the number of decomposition layers based on a preset algorithm”, it includes: Receive the frequency modulation command signal; let the frequency modulation command signal be Pt; Pt is a function of the time parameter t.

[0021] Decompose the frequency modulation command signal based on a preset algorithm.

[0022] Determine the number of decomposition layers, and decompose the frequency modulation command signal by using VMD decomposition; let the number of decomposition layers of VMD be K, and decompose the frequency modulation command signal into K subsequences IMF1, IMF2, IMF3, …, IMF K .

[0023] In another embodiment provided by the present disclosure, in the above step “S103. Generate corresponding upper asymptotes and lower asymptotes for a plurality of modal subsequences”, it includes: Let the frequency modulation command signal Pt become IMF1, IMF2, IMF3, …, IMF i , …, IMF K . after being decomposed by VMD. Among them, the subsequence IMF i = [X i1 , X i2 , X i3 , …, X ii , …, X iN .

[0024] Step 1. Obtain the intersection sequence of each subsequence.

[0025] Project each numerical point in the subsequence IMF i onto the data processing coordinate system as shown in Figure 3 .

[0026] As shown in the coordinate system in Figure 3 , the horizontal axis represents the time parameter t, and the vertical axis is the signal value X i1 , X i2, X i3 , …, X ii , …, X iN . Establish the connecting line segments between adjacent points and obtain the perpendicular bisectors of the connecting line segments; The adjacent perpendicular bisectors generate intersection points in the coordinate system, and the intersection points are O1, O2, O3, …, O N-1 , with a total of N - 1 intersection points, forming an intersection point sequence [O1, O2, O3, …, O N-1 .

[0027] Step 2: Obtain the upper limit point sequence and generate the upper limit asymptote.

[0028] Select the upper limit points from the intersection point sequence. The upper limit point is defined as O m > X i(m+1) , then this intersection point is the upper limit point.

[0029] As Figure 3 shown, O1 > X i2 , and O1 is the upper limit point.

[0030] All the selected upper limit points form an upper limit point sequence [O S1 , O S2 , O S3 , …, O Sq .

[0031] As Figure 4 shown, establish a quadratic function curve through two adjacent points in the upper limit point sequence, so that the enclosed area between the quadratic function curve and the line segment between the two adjacent points is the smallest.

[0032] In the specific process, pass through point O S1 and O S2 to generate a quadratic function X = a1t 2 + b1t + c1 with an upward opening. The area enclosed by this function and O S1 and O S2 is S1. Adjust a1, b1, c1 to make the area of S1 the smallest.

[0033] Continue to pass through point O S2 and O S3 to generate a quadratic function X = a2t 2 + b2t + c2 with an upward opening. The area enclosed by this function and O S2 and O S3 is S2. Adjust a2, b2, c2 to make the area of S2 the smallest.

[0034] Execute in this process sequence. Pass through point O Sq-1 and O Sq to generate a quadratic function X = aq-1 t 2 +b q-1 t + c q-1 , the area enclosed by this function and O Sq-1 and O Sq is S q-1 , adjust a q-1 , b q-1 , c q-1 to minimize S q-1 .

[0035] Through the above process, make the sum of the areas of S1 + S2 + … + S q-1 the smallest.

[0036] Use the above q - 1 quadratic function curves opening upward and connected to each other as the upper limit asymptote.

[0037] Step 3: Obtain the lower limit point sequence and generate the lower limit asymptote.

[0038] Select the lower limit points from the intersection point sequence. The lower limit point is defined as O m <X i(m+1) , then this intersection point is the upper limit point.

[0039] As shown in Figure 3 , O2 < X i3 , and O2 is the lower limit point.

[0040] All the selected lower limit points form the lower limit point sequence [O S1 、 , O S2 、 , O S3 、 , …, O Sq 、 .

[0041] Establish a quadratic function curve passing through two adjacent points in the lower limit point sequence, such that the enclosed area between the quadratic function curve and the line connecting the two adjacent points is the smallest.

[0042] In the specific process, pass through the points O S1 、 and O S2 、 to generate a quadratic function X = a1t 2 + b1t + c1 opening downward. The area enclosed by this function and O S1 、 and O S2 、 is S1 、 , adjust a1, b1, c1 to minimize the area of S1 、 .

[0043] Continue passing through point O S2 、 and O S3 、 Generate a quadratic function \(X = a_2t^2 + b_2t + c_2\) with a downward opening. The area enclosed by this function and O 2 +b2t+c2, and the area enclosed by this function and O S2 、 and O S3 、 is \(S_2\). Adjust \(a_2\), \(b_2\), \(c_2\) to minimize \(S_2\). 、 , adjust \(a_2\), \(b_2\), \(c_2\) so that \(S_2\) 、 has the minimum area.

[0044] Execute in the order of this process. Pass through point O Sq-1 、 and O Sq 、 Generate a quadratic function \(X = at^2 + bt + c\) with an upward opening. The area enclosed by this function and O q-1 t 2 +b q-1 t+c q-1 is \(S\). The area enclosed by this function and O Sq-1 、 and O Sq 、 is \(S\). Adjust \(a\) q-1 、 , adjust \(a\) q-1 , \(b\) q-1 , \(c\) q-1 so that \(S\) q-1 、 has the minimum area.

[0045] Through the above process, make \(S_1\) 、 +\(S_2\) 、 +…+\(S\) q-1 、 have the minimum sum of areas.

[0046] Use the curve connected by the above \(q - 1\) downward-opening quadratic function curves as the lower asymptote.

[0047] In another embodiment provided by the present disclosure, in the above step “S104. Use the corresponding upper asymptote and lower asymptote to decompose the sequence for several modal subsequences to generate a core sequence and an auxiliary sequence”, it includes: Use the upper asymptote and the lower asymptote as the threshold boundaries to decompose the subsequence IMF i into a core sequence IMF i 、 and an auxiliary sequence IMF i 、、 .

[0048] Decompose the subsequence IMF i =[Xi1 ,X i2 ,X i3 ,…,X ii ,…,X iN Compare each point in [] with the upper asymptote and the lower asymptote, where the numerical points located between the upper asymptote and the lower asymptote are retained, and other numerical points are reset to zero to form a new core sequence IMF i 、 .

[0049] In addition, for the subsequence IMF i = [X i1 ,X i2 ,X i3 ,…,X ii ,…,X iN compare each point with the upper asymptote and the lower asymptote, where the numerical points located outside the upper asymptote and the lower asymptote are retained, and other numerical points are reset to zero to form a new auxiliary sequence IMF i 、、 .

[0050] For example, if IMF i = [ X i1 ,X i2 ,X i3 ,X i4 ,X i5 ,X i6 ,X i7 ,X i8 ,X i9 , where the numerical points X i4 and X i7 are outside the upper asymptote and the lower asymptote, The core sequence IMF i 、 = [ X i1 ,X i2 ,X i3 ,0,X i5 ,X i6 ,0,X i8 ,X i9 , The auxiliary sequence IMF i 、、 = [0,0,0,X i4 ,0,0,X i7 ,0,0].

[0051] In the above step “S105. Use neural network prediction for several core sequences, and use neural network prediction after mixing and superimposing several auxiliary sequences to obtain the final prediction result”, it includes: Step 1. For the core sequence IMF1 、, IMF2 、 , IMF3 、 , …, IMF i 、 , …, IMF K 、 The GRU (Gated Recurrent Unit) network is used for prediction.

[0052] Step 2: For the auxiliary sequences IMF1 、、 , IMF2 、、 , IMF3 、、 , …, IMF i 、、 , …, IMF K 、、 First, perform hybrid superposition, and then use the GRU (Gated Recurrent Unit) network for prediction.

[0053] Suppose each auxiliary sequence IMF i 、、 has L i non-zero values and B i zero values. The hybrid superposition algorithm is as follows: exp(sigmoid(L1 / B1)) IMF1 、、 + exp(sigmoid(L2 / B2)) IMF2 、、 + exp(sigmoid(L3 / B3)) IMF3 、、 + … exp(sigmoid(L i / B i )) IMF i 、、 + … + exp(sigmoid(L k / B k )) IMF K 、、 , where sigmoid() is the activation function. After hybrid superposition, the GRU (Gated Recurrent Unit) network is used for prediction.

[0054] The core sequence and the auxiliary sequences are respectively input into the GRU network for prediction, and then the prediction results are superimposed to obtain the final prediction result. The final prediction result is applied to the power plant power grid for frequency modulation response.

[0055] According to the prediction result, the output of the frequency modulation power source is adjusted in real time, the active power distribution of the energy storage units under different working conditions is carried out, and the operation between the units is coordinated to suppress the power fluctuation and improve the frequency modulation performance of the power grid.

[0056] Corresponding to the method shown above Figure 1 , the embodiment of the present disclosure further provides an energy storage frequency modulation command prediction system with an auxiliary asymptote, including: An instruction acquisition module for acquiring a frequency modulation instruction signal; An instruction decomposition module for decomposing the frequency modulation instruction signal into a plurality of modal subsequences corresponding to the number of decomposition layers based on a preset algorithm; An asymptote generation module for generating corresponding upper asymptotes and lower asymptotes for a plurality of modal subsequences; A subsequence decomposition module for decomposing a plurality of modal subsequences using the corresponding upper asymptotes and lower asymptotes to generate a core sequence and an auxiliary sequence; A prediction processing module for predicting a plurality of core sequences and a plurality of auxiliary sequences using a neural network to obtain a final prediction result; and applying the final prediction result to an energy storage device for frequency modulation response.

[0057] In addition, the prediction processing module includes: A core sequence prediction module for predicting a plurality of core sequences using a neural network; An auxiliary sequence prediction module for predicting a plurality of auxiliary sequences after mixing and superposition using a neural network.

[0058] In order to further verify the advantages of the present invention, the method of the present invention and the method of GRU prediction are respectively used to predict the frequency modulation sequence, and the results are as follows.

[0059] The acquisition of the test frequency modulation instruction comes from a certain power plant in Gansu. The acquisition time of the frequency modulation sequence 1 is from 0:00 to 15:00 on November 1, 2022, and one data point is collected every 1 second. The acquisition time of the frequency modulation sequence 2 is from 0:00 to 15:00 on November 2, 2022, and one data point is collected every 1 second.

[0060] The performance evaluation is as shown in Table 1 below: Table 1

[0061] The four evaluation indexes are as shown in Table 2 below: Table 2

[0062] N represents the sample size, and respectively represent the actual value and the predicted value at time n.

[0063] It can be seen from the experimental results that all four evaluation indexes are reduced, indicating that the proposed model can well improve the prediction accuracy.

[0064] An embodiment of the present disclosure provides a computer device, including: a processor, a memory, and a bus. The memory stores machine-readable instructions executable by the processor. When the computer device runs, the processor communicates with the memory through the bus. When the machine-readable instructions are executed by the processor, the steps provided in any embodiment of the present disclosure are executed.

[0065] The computer device provided in an embodiment of the present application includes a processor, a memory, and a bus. Among them, the memory is used to store execution instructions, including an internal memory and an external memory; the internal memory here is also called the main memory, which is used to temporarily store the operation data in the processor and the data exchanged with external memories such as hard disks. The processor exchanges data with the external memory through the internal memory. When the electronic device runs, the processor communicates with the memory through the bus, so that the processor executes the following instructions: Obtain a frequency modulation instruction signal; Decompose the frequency modulation instruction signal into a number of modal subsequences corresponding to the decomposition level based on a preset algorithm; Generate corresponding upper asymptotes and lower asymptotes for a number of modal subsequences; Perform sequence decomposition on a number of modal subsequences using the corresponding upper asymptotes and lower asymptotes to generate a core sequence and an auxiliary sequence; Perform neural network prediction on a number of core sequences and a number of auxiliary sequences to obtain a final prediction result; apply the final prediction result to an energy storage device for frequency modulation response.

[0066] An embodiment of the present disclosure provides a computer-readable storage medium. A computer program is stored on the computer-readable storage medium. When the computer program is run by a processor, the steps provided in any embodiment of the present disclosure are executed. Among them, the storage medium can be a volatile or non-volatile computer-readable storage medium.

[0067] Through the description of the above embodiments, those skilled in the art can clearly understand that the embodiments of the present disclosure can be implemented by hardware or by means of software plus a necessary general hardware platform. Based on such an understanding, the technical solutions of the embodiments of the present disclosure can be embodied in the form of a software product. The software product can be stored in a non-volatile storage medium (which can be a read-only optical disc, a USB flash drive, a mobile hard disk, etc.), including several instructions for causing a computer device (which can be a personal computer, a server, or a network device, etc.) to execute the methods described in various embodiments of the present disclosure.

[0068] Those skilled in the art can understand that the drawings are only schematic diagrams of a preferred embodiment, and the modules or processes in the drawings are not necessarily essential for implementing the present disclosure.

[0069] Those skilled in the art can understand that the modules in the device in the embodiments can be distributed in the device in the embodiments as described in the embodiments, or can be correspondingly changed and located in one or more devices different from the present embodiments. The modules of the above embodiments can be combined into one module, or can be further split into multiple sub-modules.

[0070] The serial numbers of the above embodiments of the present disclosure are only for description and do not represent the advantages and disadvantages of the embodiments.

[0071] Obviously, those skilled in the art can make various changes and modifications to the present disclosure without departing from the spirit and scope of the present disclosure. Thus, if these modifications and variations of the present disclosure fall within the scope of the claims of the present disclosure and their equivalent technologies, the present disclosure also intends to include these modifications and variations.

[0072] Finally, it should be noted that the above is only an explanation of the present invention and is not used to limit the present invention. Although the present invention has been described in detail, for those skilled in the art, they can still modify the foregoing recorded technical solutions, or perform equivalent replacements for some of the technical features. Any modifications, equivalent replacements, improvements, etc. made within the spirit and principle of the present invention shall be included within the protection scope of the present invention.

Claims

1. A prediction method for energy storage frequency modulation commands of auxiliary asymptotes, characterized in that, including, S101. Obtain a frequency modulation command signal; S102. Based on a preset algorithm, decompose the frequency modulation command signal into a number of modal subsequences corresponding to the number of decomposition layers; S103. Generate corresponding upper asymptotes and lower asymptotes for the number of modal subsequences; S104. Use the corresponding upper asymptotes and lower asymptotes to decompose the number of modal subsequences to generate a core sequence and an auxiliary sequence; S105. Use a neural network to predict the number of core sequences and the number of auxiliary sequences to obtain a final prediction result; apply the final prediction result to an energy storage device for frequency modulation response.

2. The energy storage frequency modulation command prediction method with auxiliary asymptotes according to claim 1, characterized in that, In step S105, the number of core sequences is predicted using a neural network, and the number of auxiliary sequences are mixed and superimposed and then predicted using a neural network to obtain a final prediction result.

3. The energy storage frequency modulation command prediction method with auxiliary asymptotes according to claim 1, characterized in that, In step S102, determine the decomposition level, and decompose the frequency modulation command signal by using VMD decomposition; assume that the decomposition level of VMD is K, and decompose the frequency modulation command signal into K sub-sequences IMF1, IMF2, IMF3, …, IMF K .

4. The energy storage frequency modulation command prediction method with auxiliary asymptotes according to claim 3, characterized in that, In step S103, it includes: each subsequence IMF decomposed by VMD i =[ X i1 ,X i2 ,X i3 ,…,X ii ,…,X iN ; S1031. Obtain the intersection sequence of each subsequence; S1032. Based on the intersection sequence, obtain the upper limit point sequence and generate the upper asymptote; S1033. Based on the intersection sequence, obtain the lower limit point sequence and generate the lower asymptote.

5. The energy storage frequency modulation command prediction method with auxiliary asymptotes according to claim 4, characterized in that, In S1031, Project each numerical point in the subsequence IMF i onto the data processing coordinate system; establish the connecting line segments between adjacent points, and obtain the perpendicular bisectors of the connecting line segments; there are N - 1 intersection points generated by adjacent perpendicular bisectors in the coordinate system, forming an intersection point sequence [O1, O2, O3, …, O N-1 .

6. The energy storage frequency modulation command prediction method with auxiliary asymptotes according to claim 5, characterized in that, In S1032, Select the upper limit points from the intersection point sequence, and the upper limit points are defined as O m > X i(m+1) , then this intersection point is the upper limit point; All selected upper limit points form an upper limit point sequence [O S1 , O S2 , O S3 , …, O Sq ; Establish a quadratic function curve with an upward opening through two adjacent points in the upper limit point sequence, such that the enclosed area between the quadratic function curve and the line connecting the two adjacent points is the smallest; perform in this order to obtain a curve formed by connecting q - 1 quadratic function curves as the upper asymptote.

7. The energy storage frequency modulation command prediction method with auxiliary asymptotes according to claim 6, characterized in that, In S1033, Select the lower limit points from the intersection point sequence, where the lower limit points are defined as O m <X i(m+1) , then this intersection point is the lower limit point; All selected lower limit points form a lower limit point sequence [O S1 、 , O S2 、 , O S3 、 , …, O Sq 、 ; Establish a quadratic function curve with a downward opening through two adjacent points in the lower limit point sequence, such that the enclosed area between the quadratic function curve and the line connecting the two adjacent points is the smallest; perform in this order to obtain a curve formed by connecting q - 1 quadratic function curves as the lower asymptote.

8. The energy storage frequency modulation command prediction method with auxiliary asymptotes according to claim 7, characterized in that, In step S104, Taking the upper asymptote and the lower asymptote as threshold boundaries, decompose the subsequence IMF i into the core sequence IMF i 、 and the auxiliary sequence IMF i 、、 ; Compare each point in the subsequence IMF i = [X i1 , X i2 , X i3 , …, X ii , …, X iN with the upper asymptote and the lower asymptote. Retain the numerical points located between the upper asymptote and the lower asymptote, and reset other numerical points to zero to form the core sequence IMF i 、 ; Subsequence IMF i = [ X i1 , X i2 , X i3 , …, X ii , …, X iN Compare each point with the upper asymptote and the lower asymptote. Retain the numerical points outside the upper asymptote and the lower asymptote, and reset the other numerical points to zero to form the auxiliary sequence IMF i 、、 .

9. An energy storage frequency modulation command prediction system for auxiliary asymptotes, characterized in that, including, An instruction acquisition module for acquiring a frequency modulation command signal; An instruction decomposition module for decomposing the frequency modulation command signal into a number of modal subsequences corresponding to the number of decomposition layers based on a preset algorithm; An asymptote generation module for generating corresponding upper asymptotes and lower asymptotes for the number of modal subsequences; A subsequence decomposition module for using the corresponding upper asymptotes and lower asymptotes to decompose the number of modal subsequences to generate a core sequence and an auxiliary sequence; A prediction processing module for using a neural network to predict the number of core sequences and the number of auxiliary sequences to obtain a final prediction result; applying the final prediction result to an energy storage device for frequency modulation response.

10. The energy storage frequency modulation command prediction system for the auxiliary asymptote according to claim 9, wherein: The prediction processing module includes: A core sequence prediction module, configured to predict a plurality of core sequences using a neural network; An auxiliary sequence prediction module, configured to perform hybrid superposition on a plurality of auxiliary sequences and then predict using a neural network.

Images