Optimized and corrected energy storage frequency modulation instruction prediction method and system

Abstract

The energy storage frequency modulation instruction prediction method comprises the following steps: acquiring an original frequency modulation instruction sequence; aiming at each signal in the original frequency modulation instruction sequence, obtaining an optimization correction parameter, wherein the optimization correction parameter comprises an optimal point position, an optimal straight slope angle and a hit number; correcting the original frequency modulation instruction sequence signal by using the optimal point position, the optimal straight slope angle and the hit number in the optimized correction parameter to obtain a corrected frequency modulation instruction sequence; predicting the corrected frequency modulation instruction sequence by using a neural network to obtain a final prediction result; 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 with optimal correction. Background Art

[0002] Currently, large hydropower and thermal power units (coal-fired and gas-fired) are the primary frequency regulation sources in my country's major regional power grids. Frequency regulation power output is adjusted to respond to system frequency changes. Hybrid energy storage frequency regulation command prediction involves analyzing historical frequency regulation commands and related data to develop a predictive model to predict future frequency regulation command requirements. This allows for the optimization of the hybrid energy storage system's control strategy and improves its frequency regulation performance and efficiency. This system improves system performance and response speed, enabling rapid output power adjustments to respond quickly to grid frequency changes, effectively balancing supply and demand within the power system and maintaining grid frequency stability. Improved regulation accuracy allows for more precise tracking of frequency regulation commands, reducing regulation errors and improving power supply quality and stability. Enhanced system stability, through the rational allocation of power and energy across different energy storage devices, can reduce overall system risk, improve system reliability and stability, and mitigate the risk of system failure due to failure or performance degradation of a single energy storage device. Extending equipment life and optimizing charging and discharging strategies: Based on the prediction results, appropriate charging and discharging strategies can be formulated to avoid excessive or frequent charging and discharging of energy storage devices, thereby extending their service life. Balanced equipment usage: Rationally allocating the usage frequency and load of different energy storage devices ensures relatively even aging across all devices, reducing maintenance costs and replacement frequency. Reduced costs and improved economic benefits: By improving frequency regulation performance and efficiency, hybrid energy storage systems can achieve greater frequency regulation benefits while reducing power outages and equipment damage costs caused by unstable grid frequency. Optimized investment costs: Based on forecast results, the capacity and quantity of different energy storage devices can be rationally allocated, avoiding overinvestment and waste of resources, and improving return on investment.

[0003] The traditional prediction method is to put the original FM sequence directly into the GRU network for prediction. The disadvantage of this method is that the original sequence has great nonlinearity and irregularity. Directly putting the original sequence into the prediction will lead to inaccurate prediction results and large prediction errors. Therefore, the original FM sequence must be preprocessed before prediction, such as Figure 2 As shown in Figure 1, common methods such as VMD and CEEMD decomposition decompose the original FM sequence into a series of low-frequency subsequences, which are then fed into a neural network for prediction. However, while combining VMD (Variational Mode Decomposition) and CEEMD (Complementary Ensemble Empirical Mode Decomposition) for decomposition and prediction of the original FM sequence can improve the accuracy of non-stationary signal processing, it also has the following disadvantages: 1. High computational complexity. The decomposition process is time-consuming: Both VMD and CEEMD require multiple iterations (CEEMD requires multiple noise additions and decompositions). Combining the two doubles the computational effort, especially for long time series data, which suffers from poor real-time performance. Parameter tuning is complex: VMD requires presetting the number of modes (K) and penalty factor (α), while CEEMD requires setting the noise amplitude and number of integrations. Improper parameter selection can lead to over- or under-decomposition.

[0004] 2. Modal redundancy and over-decomposition risk. Sensitivity to the number of modalities: If the number of VMD modes (K) is set too high, it may decompose into noise modes without physical meaning; if it is set too low, it may fail to capture key features. Information fragmentation: CEEMD may generate redundant modes due to residual noise or over-decomposition, resulting in an excessive number of subsequences and increasing the complexity of the subsequent prediction model.

[0005] 3. Modal aliasing and frequency confusion. Residual modal aliasing: While CEEMD reduces modal aliasing compared to traditional EMD, it does not completely eliminate it. In particular, overlap may still occur in areas with abrupt changes in the original FM sequence or high-frequency noise. VMD frequency boundary issues: VMD assumes that modes are narrowband signals, but the actual original FM sequence may contain broadband components, resulting in blurred frequency boundaries between modes after decomposition.

[0006] 4. Difficulty integrating prediction models. Multimodal modeling complexity: Each decomposed subsequence must be modeled separately or predicted jointly, which increases the number of models exponentially. Integration strategies (such as linear superposition and neural network fusion) may introduce errors. Timing synchronization issues: The decomposed subsequences must be time-aligned. If phase shifts occur during the decomposition process (such as the end-point effect of CEEMD), the prediction results may be distorted.

[0007] 5. Information loss and reconstruction error. Decomposition-reconstruction distortion: VMD and CEEMD require reconstructing submodals into the original signal after decomposition. Decomposition errors or inappropriate parameters can cause the reconstructed signal to deviate from the true value, affecting the quality of the predicted input. High-frequency noise amplification: Random fluctuations in the original FM sequence may be decomposed into high-frequency modes. If directly ignored or simply processed, critical transient information may be lost.

[0008] 6. Inadequate handling of sudden changes and non-stationarity. Delayed response to extreme events: The corresponding mode may be dispersed across multiple subsequences, making it difficult for the forecasting model to capture sudden patterns in a timely manner. Non-stationary residuals: If the decomposition fails to completely separate the stationary and non-stationary components, the residuals will still contain complex fluctuations, reducing forecast accuracy. Summary of the Invention

[0009] In view of the deficiencies in the prior art, in order to address the shortcomings of VMD / CEEMD and improve the accuracy of prediction, the present invention proposes an optimization correction method to process the original frequency modulation sequence.

[0010] A method for predicting energy storage frequency modulation instructions by optimizing and correcting, characterized by comprising: S101, obtaining an original frequency modulation instruction sequence; S102, obtaining optimized correction parameters for each signal in the original frequency modulation instruction sequence, wherein the optimized correction parameters include the optimal point position, the optimal straight line slope angle, and the number of hits; S103, using the optimal point position, optimal straight line slope angle, and hit number in the optimization correction parameters to correct the original frequency modulation instruction sequence signal to obtain a corrected frequency modulation instruction sequence; S104: Using a neural network to predict the modified frequency modulation instruction sequence to obtain a final prediction result.

[0011] An energy storage frequency modulation instruction prediction system with optimized correction is characterized by including: Instruction acquisition module, obtains the original frequency modulation instruction sequence; The optimization correction parameter calculation module obtains the optimization correction parameters for each signal in the original frequency modulation instruction sequence, and the optimization correction parameters include the optimal point position, the optimal straight line slope angle, and the number of hits; The correction module corrects the original frequency modulation instruction sequence signal by using the optimal point position, the optimal straight line slope angle, and the hit number in the optimized correction parameters to obtain a corrected frequency modulation instruction sequence; The prediction processing module uses a neural network to predict the modified frequency modulation instruction sequence to obtain the final prediction result.

[0012] The beneficial effects of the present invention are: Reducing the nonlinearity of the original signal sequence makes the corrected sequence smoother, greatly reduces the difficulty of prediction, and improves the accuracy of frequency regulation instruction prediction, thereby improving the response accuracy of the power plant and the frequency regulation benefits. BRIEF DESCRIPTION OF THE DRAWINGS

[0013] Figure 1 A diagram of the method steps.

[0014] Figure 2 This is an illustration of the traditional algorithm.

[0015] Figure 3 Schematic diagram of the optimal point.

[0016] Figure 4 Schematic diagram of the optimal straight line. DETAILED DESCRIPTION

[0017] The following will be combined with the drawings in the embodiments of this application to clearly and completely describe the technical solutions in the embodiments of this application. Obviously, the embodiments described are only part of the embodiments of this application, not all of the embodiments. Based on the embodiments in this application, all other embodiments obtained by ordinary technicians in this field without making creative efforts are within the scope of protection of this application.

[0018] It should be understood that the step numbers used herein are only for convenience of description and are not intended to limit the order in which the steps are to be executed.

[0019] It should be understood that the terms used in this specification are only for the purpose of describing specific embodiments and are not intended to limit this application. As used in this specification and the appended claims, the singular forms "a", "an", and "the" are intended to include the plural forms unless the context clearly indicates otherwise.

[0020] The terms “include” and “comprising” indicate the presence of described features, integers, steps, operations, elements and / or components, but do not preclude the presence or addition of one or more other features, integers, steps, operations, elements, components and / or groups thereof.

[0021] The term "and / or" refers to and includes any and all possible combinations of one or more of the associated listed items.

[0022] The embodiment of the present disclosure provides a method for predicting energy storage frequency modulation instructions by optimizing and correcting the energy storage frequency modulation instructions. Figure 1 Shown, including: S101, obtaining an original frequency modulation instruction sequence; S102, obtaining optimized correction parameters for each signal in the original frequency modulation instruction sequence, wherein the optimized correction parameters include the optimal point position, the optimal straight line slope angle, and the number of hits; S103, using the optimal point position, optimal straight line slope angle, and hit number in the optimization correction parameters to correct the original frequency modulation instruction sequence signal to obtain a corrected frequency modulation instruction sequence; S104: Using a neural network to predict the modified frequency modulation instruction sequence to obtain a final prediction result.

[0023] The frequency modulation command prediction method based on supercapacitor energy storage, provided in the embodiments of the present disclosure, uses novel supercapacitors and lithium batteries as energy storage devices and modulates the frequency of a power plant's power grid when its frequency fluctuates. The method monitors the power plant's power grid in real time and generates a corresponding frequency modulation command signal when the power supply frequency fluctuates. After acquiring this original signal, the method then predicts future signals to implement frequency modulation of the power plant's power grid.

[0024] Wherein, step S101, obtaining the original frequency modulation instruction signal sequence, includes: The original frequency modulation command signal sequence is Pt=[ X1,X2,X3,…,X i ,…,X N ]. The original frequency modulation instruction signal sequence Pt is a function of time t.

[0025] Project each numerical point in the original frequency modulation command signal sequence Pt onto the plane coordinate system, where the vertical axis represents the magnitude of the numerical value and the horizontal axis represents the number of sampling times.

[0026] like Figure 3 As shown in the coordinate system, the horizontal axis represents the number of sampling times, and the vertical axis represents the signal values ​​X1, X2, X3, ..., X i ,…,X N .

[0027] In another embodiment disclosed in the present application, for the original frequency modulation instruction signal sequence, the stronger the degree of nonlinearity, the more difficult it is to predict. An algorithm is used to obtain appropriate optimization parameters, and the optimization parameters are used to correct the original frequency modulation instruction sequence.

[0028] The above S102, obtaining optimized correction parameters for each signal in the original frequency modulation instruction sequence, wherein the optimized correction parameters include the optimal point position, the optimal straight line slope angle, and the optimal straight line hit number; includes: Step 1: Obtain an optimal point X for each signal in the original frequency modulation instruction sequence p .

[0029] like Figure 3 As shown, set X p and [X1,X2,X3,…,X i ,…,X N ] form a distance sequence [L1,L2,L3,…,L N ], where the sum of distances is the shortest, X p That is the optimal point.

[0030] Optimal Point X p The sum of the corresponding distances L sum =L1+L2+L3+…+L N , optimal point X p The distance and L sum is the minimum value.

[0031] Step 2: Obtain the slope angle of an optimal straight line for each signal in the original FM instruction sequence θ ; like Figure 4 As shown, a straight line d is generated through the coordinate origin, and the sum of the vertical distances of each signal in the sequence to the straight line d is required to be the shortest.

[0032] Determine the angle between line d and the X-axis as θ .

[0033] Step 3: Determine the hit number M of each signal in the original frequency modulation instruction sequence on the optimal straight line; Set Pt=[X1,X2,X3,…,X i ,…,X N ]The number of points on this straight line is M.

[0034] Among them, the optimal position and L sum It reflects the strength of the volatility of the frequency modulation sequence.

[0035] The angle of the optimal line d θ The hit number M value reflects the strength of the regularity of the FM sequence.

[0036] The above S103, using the optimal point position, optimal straight line slope angle, and hit number in the optimization correction parameters to correct the original frequency modulation instruction sequence signal to obtain a corrected frequency modulation instruction sequence, includes: According to the optimal point position, optimal point distance and L sum , optimal straight line slope angle θ , and the optimal straight line hit number M respectively modify the original sequence. Specifically including the following situations: 1) When the optimal point X p Contained in [X1,X2,X3,…,X i ,…,X N ], and the optimal position and L sum ≤max(X1,X2,X3,…,X i ,…,X N ), indicating that the volatility of the series is weak and there is no need to adjust the volatility of the series.

[0037] The correction method of the original sequence is as follows: .

[0038] The corrected sequence is identical to the original sequence.

[0039] 2) When the optimal point X p Contained in [X1,X2,X3,…,X i ,…,X N ], and the optimal position and L sum ≥max(X1,X2,X3,…,X i ,…,X N ), indicating that the volatility of the sequence is stronger than in 1), but still acceptable, and only requires fine-tuning of the volatility of the sequence.

[0040] Let the optimal point X p It is exactly the sequence [X1,X2,X3,…,X i ,…,X N ] in X r .

[0041] sigmoid() is the activation function.

[0042] Each signal in the original sequence is corrected according to the formula.

[0043] 3) When the optimal point X p Not included in [X1,X2,X3,…,X i ,…,X N ], regardless of the optimal position and L sum and max(X1,X2,X3,…,X i ,…,X N ), indicating that volatility is very high and needs to be adjusted.

[0044] At this time, let X p The point closest to the line is the sequence [X1,X2,X3,…,X i ,…,X N ] in X r , the distance between the lines is LR.

[0045] .

[0046] Each signal in the original sequence is corrected according to the formula.

[0047] 4) When the hit number M ≥ N / 3, it indicates that the sequence has strong regularity and there is no need to fine-tune the regularity of the sequence.

[0048] That is, no regular adjustments are made, and the corrected sequence is identical to the original sequence.

[0049] 5) When the hit number M is less than N / 3, it indicates that the sequence regularity is weak and the sequence needs to be adjusted regularly.

[0050] .

[0051] Each signal in the original sequence is corrected according to the formula.

[0052] The original FM sequence [X1,X2,X3,…,X i ,…,X NWhen making corrections after the optimization method, first determine whether the regularity belongs to 4) or 5) based on the number of hits, then determine whether the volatility belongs to 1), 2), or 3) based on the optimal point position, and finally output the final result.

[0053] .

[0054] Each signal in the original sequence is corrected according to the formula, and regularity adjustment and volatility adjustment are performed independently on the original sequence, and the regularity adjustment results and volatility adjustment results are summed.

[0055] The original FM sequence [X1,X2,X3,…,X i ,…,X N ] The sequence after the above optimization correction is [X1 、 ,X2 、 ,X3 、 ,…,X i 、 ,…,X N 、 ].

[0056] With the above Figure 1 Corresponding to the method shown, the embodiment of the present disclosure further provides an energy storage frequency modulation instruction prediction system with optimized correction, which is characterized by including: Instruction acquisition module, obtains the original frequency modulation instruction sequence; The optimization correction parameter calculation module obtains the optimization correction parameters for each signal in the original frequency modulation instruction sequence, and the optimization correction parameters include the optimal point position, the optimal straight line slope angle, and the number of hits; The correction module corrects the original frequency modulation instruction sequence signal by using the optimal point position, the optimal straight line slope angle, and the hit number in the optimized correction parameters to obtain a corrected frequency modulation instruction sequence; The prediction processing module uses a neural network to predict the modified frequency modulation instruction sequence to obtain the final prediction result.

[0057] In order to further verify the advantages of the present invention, the present invention uses the method of the present invention and the ANFIS prediction method to predict the frequency modulation sequence, and the results are as follows.

[0058] The experimental frequency modulation instructions were obtained from a power plant in Gansu. The acquisition time of frequency modulation sequence 1 was from 4:00 to 18:00 on a certain day in January 2021, with one data point collected every 1 second. The acquisition time of frequency modulation sequence 2 was from 0:00 to 19:00 on a certain day in March 2022, with one data point collected every 1 second.

[0059]

[0060] Four evaluation indicators

[0061] N represents the sample size, and represent the actual value and predicted value at time n respectively.

[0062] From the experimental results, we can see that all four evaluation indicators have been reduced, which shows that the proposed model can well improve the prediction accuracy.

[0063] An embodiment of the present disclosure provides a computer device, comprising: a processor, a memory, and a bus, wherein the memory stores machine-readable instructions executable by the processor. When the computer device is running, the processor and the memory communicate via the bus, and when the machine-readable instructions are executed by the processor, the steps provided in any embodiment of the present disclosure are performed.

[0064] The computer device provided in the embodiment of the present application includes a processor, a memory, and a bus. The memory is used to store and execute instructions, and includes internal memory and external memory. The internal memory here is also called internal memory, which is used to temporarily store the calculation data in the processor and the data exchanged with the external memory such as the hard disk. The processor exchanges data with the external memory through the internal memory. When the electronic device is running, the processor and the memory communicate through the bus, so that the processor executes the following instructions: Get the original frequency modulation instruction sequence; Obtaining optimized correction parameters for each signal in the original frequency modulation instruction sequence, wherein the optimized correction parameters include the optimal point position, the optimal straight line slope angle, and the number of hits; The original frequency modulation instruction sequence signal is corrected by using the optimal point position, optimal straight line slope angle and hit number in the optimization correction parameters to obtain a corrected frequency modulation instruction sequence; The modified frequency modulation instruction sequence is predicted using a neural network to obtain the final prediction result.

[0065] The present disclosure provides a computer-readable storage medium having a computer program stored thereon, which, when executed by a processor, executes the steps provided in any embodiment of the present disclosure. The storage medium may be a volatile or non-volatile computer-readable storage medium.

[0066] Through the above description of the embodiments, those skilled in the art will clearly understand that the embodiments of the present disclosure can be implemented through hardware or through software plus the necessary general hardware platform. Based on this understanding, the technical solutions of the embodiments of the present disclosure can be embodied in the form of a software product, which can be stored in a non-volatile storage medium (such as a read-only optical disk, a USB flash drive, a mobile hard disk, etc.) and includes a number of instructions for causing a computer device (such as a personal computer, a server, or a network device) to execute the methods described in the various embodiments of the present disclosure.

[0067] Those skilled in the art will understand that the accompanying drawings are merely schematic diagrams of a preferred embodiment, and the modules or processes in the accompanying drawings are not necessarily required for implementing the present disclosure.

[0068] Those skilled in the art will appreciate that the modules in the devices of the embodiments may be distributed in the devices of the embodiments as described in the embodiments, or may be located in one or more devices different from the embodiments with corresponding changes. The modules of the above embodiments may be combined into one module or further split into multiple submodules.

[0069] The serial numbers of the above-mentioned embodiments of the present disclosure are for description only and do not represent the advantages or disadvantages of the embodiments.

[0070] Obviously, those skilled in the art may 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 equivalents, the present disclosure is intended to include these modifications and variations.

[0071] Finally, it should be noted that the foregoing description is merely an explanation of the present invention and is not intended to limit the present invention. Although the present invention has been described in detail, those skilled in the art will be able to modify the aforementioned technical solutions or substitute equivalents for some of the technical features. Any modifications, equivalent substitutions, and improvements made within the spirit and principles of the present invention shall be included within the scope of protection of the present invention.

Claims

1. A method for predicting energy storage frequency modulation instructions by optimizing and correcting, characterized in that: include, S101, obtaining an original frequency modulation instruction sequence; S102, obtaining optimized correction parameters for each signal in the original frequency modulation instruction sequence, wherein the optimized correction parameters include the optimal point position, the optimal straight line slope angle, and the number of hits; S103, using the optimal point position, optimal straight line slope angle, and hit number in the optimization correction parameters to correct the original frequency modulation instruction sequence signal to obtain a corrected frequency modulation instruction sequence; S104: Using a neural network to predict the modified frequency modulation instruction sequence to obtain a final prediction result.

2. The energy storage frequency modulation instruction prediction method with optimization and correction according to claim 1 is characterized in that: In step S101, The original frequency modulation command signal sequence is Pt=[ X1,X2,X3,…,X i ,…,X N ], Project each numerical point in the original frequency modulation command signal sequence Pt onto the plane coordinate system, where the vertical axis represents the magnitude of the numerical value and the horizontal axis represents the number of sampling times.

3. The energy storage frequency modulation instruction prediction method with optimization and correction according to claim 1 is characterized in that: Step S102 includes: Step 1: Obtain an optimal point X for each signal in the original frequency modulation instruction sequence p ; Step 2: Obtain the slope angle of an optimal straight line for each signal in the original FM instruction sequence θ ; Step 3: Determine the hit number M of each signal in the original frequency modulation instruction sequence on the optimal straight line.

4. The energy storage frequency modulation instruction prediction method with optimization and correction according to claim 3 is characterized in that: In step one; Set X p and [X1,X2,X3,…,X i ,…,X N ] form a distance sequence [L1,L2,L3,…,L N ], where the sum of distances is the shortest, X p That is the optimal point; Optimal Point X p The sum of the corresponding distances L sum =L1+L2+L3+…+L N , optimal point X p The distance and L sum is the minimum value.

5. The energy storage frequency modulation instruction prediction method with optimization and correction according to claim 4 is characterized in that: In step 2; Generate a straight line d through the origin of the coordinate system, and require that the sum of the vertical distances from each signal in the sequence to the straight line d be the shortest. Determine the angle between line d and the X-axis as θ .

6. The energy storage frequency modulation instruction prediction method with optimization and correction according to claim 5 is characterized in that: Set Pt=[ X1,X2,X3,…,X i ,…,X N ]The number of points on the optimal straight line is M.

7. The energy storage frequency modulation instruction prediction method with optimization and correction according to claim 6 is characterized in that: S103; Step 1: determine the regularity of the sequence based on the number of hits, and make adjustments to generate a first correction result; Step 2: Determine the volatility of the sequence based on the optimal point position and make adjustments to generate a second correction result; Step 3: sum the first correction result and the second correction result to obtain the final correction result.

8. The energy storage frequency modulation instruction prediction method with optimization and correction according to claim 7 is characterized in that: In step one, When the number of hits M≥N / 3, X i 、 =X i ; When the hit number M is less than N / 3, .

9. The energy storage frequency modulation instruction prediction method with optimization and correction according to claim 7 is characterized in that: In step 2, When the optimal point X p Contained in [X1,X2,X3,…,X i ,…,X N ], and the optimal position and L sum ≤max(X1,X2,X3,…,X i ,…,X N ), X i 、 =X i ; When the optimal point X p Contained in [X1,X2,X3,…,X i ,…,X N ], and the optimal position and L sum ≥max(X1,X2,X3,…,X i ,…,X N ), let the optimal point X p It is exactly the sequence [X1,X2,X3,…,X i ,…,X N ] in X r ； sigmoid() is the activation function; When the optimal point X p Not included in [X1,X2,X3,…,X i ,…,X N ], let X p The closest point to the line is the sequence X1, X2, X3, ..., X i ,…,X N ] in X r , the distance between the lines is LR, 。 10. An energy storage frequency modulation instruction prediction system with optimization and correction, characterized in that: include, Instruction acquisition module, obtains the original frequency modulation instruction sequence; The optimization correction parameter calculation module obtains the optimization correction parameters for each signal in the original frequency modulation instruction sequence, and the optimization correction parameters include the optimal point position, the optimal straight line slope angle, and the number of hits; The correction module corrects the original frequency modulation instruction sequence signal by using the optimal point position, the optimal straight line slope angle, and the hit number in the optimized correction parameters to obtain a corrected frequency modulation instruction sequence; The prediction processing module uses a neural network to predict the modified frequency modulation instruction sequence to obtain the final prediction result.

Images