Frequency modulation instruction prediction method of sodium ion super-capacity energy storage system and related equipment

Abstract

The invention belongs to the field of electric power energy storage and frequency modulation, and discloses a frequency modulation instruction prediction method of a sodium ion super-capacitance energy storage system and related equipment, and the method comprises the steps: collecting an original frequency modulation instruction sequence of the sodium ion super-capacitance energy storage system; performing approximate integratable processing on the original frequency modulation instruction sequence to obtain an approximate integratable quantity; performing periodic processing on the approximate integratable quantity to obtain a periodic frequency modulation instruction sequence; and inputting the periodic frequency modulation instruction sequence into a pre-trained prediction model to obtain a frequency modulation instruction prediction result. According to the method, the problem of large prediction error caused by direct use of original data in a traditional prediction method is effectively solved, and the prediction precision is improved, so that the power distribution strategy of the super capacitor and the lithium battery is optimized, the optimal configuration of energy storage resources is realized, and the operation efficiency and the stability of the power frequency modulation system are remarkably improved.

Description

Technical Field

[0001] The present invention belongs to the technical field of electric energy storage and frequency modulation, and in particular relates to a frequency modulation instruction prediction method and related equipment for a sodium ion supercapacitor energy storage system. Background Art

[0002] With the large-scale grid connection of renewable energy power generation, the frequency stability of the power system faces severe challenges. As the traditional main frequency modulation power source, thermal power units have problems such as increased coal consumption, reduced reliability, and reduced operating life in the long-term frequency modulation process. At the same time, the scarcity of high-quality and efficient frequency modulation power sources and the constraints of environmental pressure on the regulation capacity of units have further increased the demand for power frequency modulation. In order to meet this challenge, energy storage technology has been widely used in the frequency modulation of power systems.

[0003] Supercapacitors and lithium batteries are two common energy storage devices. Supercapacitors have the advantages of high power density, fast charging and discharging speed, and long cycle life, but their energy density is relatively low. Lithium batteries have high energy density, but their power density and charging and discharging speed are relatively slow. Therefore, the supercapacitor-coupled lithium battery energy storage system (referred to as the sodium ion supercapacitor energy storage system) composed of supercapacitors and lithium batteries can give full play to the advantages of both and achieve efficient and stable operation of power frequency modulation. However, for supercapacitor-coupled lithium battery energy storage systems, traditional prediction methods have serious defects in frequency modulation instruction prediction. Traditional prediction methods often directly input the original frequency modulation instructions into the neural network for prediction, but due to the strong nonlinearity and irregularity of the original frequency modulation instructions, direct prediction will result in large errors. Inaccurate prediction leads to unreasonable power distribution of supercapacitors and lithium batteries in the energy storage system, and the optimal utilization of energy storage resources cannot be achieved, which reduces the frequency modulation effect.

[0004] It can be seen that in terms of frequency modulation instruction prediction technology, the traditional prediction method uses the original frequency modulation instruction data to be put into the neural network for prediction. Since this unprocessed data is directly used for prediction, it will produce a large error. Summary of the invention

[0005] The present invention provides a frequency modulation instruction prediction method and related equipment for a sodium ion supercapacitor energy storage system, so as to solve the technical problem that the traditional prediction method directly inputs the original frequency modulation instruction into a neural network for prediction, resulting in inaccurate prediction results.

[0006] In order to achieve the above object, the present invention adopts the following technical solution: A method for predicting frequency modulation instructions of a sodium ion supercapacitor energy storage system, comprising: Collect the original frequency modulation instruction sequence of the sodium ion supercapacitor energy storage system; The original frequency modulation instruction sequence is processed approximately as integrable to obtain an approximately integrable quantity; The approximately integrable quantity is processed periodically to obtain a periodic frequency modulation instruction sequence; The periodic frequency modulation instruction sequence is input into the pre-trained prediction model to obtain the frequency modulation instruction prediction result.

[0007] Furthermore, the process of performing approximate integrability processing on the original frequency modulation instruction sequence to obtain an approximate integrable quantity includes: The smooth function is used to smooth the collected original frequency modulation instruction sequence to obtain an approximate integrable quantity.

[0008] Furthermore, the smoothing function is used to smooth the collected original frequency modulation instruction sequence, and an incommensurable quantity is also obtained; the incommensurable quantity is used to be input into the pre-trained prediction model simultaneously with the periodic frequency modulation instruction sequence to predict the frequency modulation instruction; wherein the incommensurable quantity is expressed as: X2(t)=X(t)-X1(t) Where X(t) represents the original frequency modulation instruction sequence; X1(t) represents the approximately integrable quantity; X2(t) represents the non-integrable quantity.

[0009] Furthermore, the approximate integrable quantity is subjected to periodic processing to obtain a periodic frequency modulation instruction sequence, and the specific formula for the periodic processing is as follows:

[0010]

[0011] In the formula, X1(t) represents the approximately integrable quantity; represents the first coefficient associated with the approximately integrable quantity; represents the second coefficient related to the approximately integrable quantity; k represents the frequency component; t represents the time variable; j represents the length of the original frequency modulation instruction sequence.

[0012] Furthermore, before inputting the periodic frequency modulation instruction sequence into the pre-trained prediction model to obtain the frequency modulation instruction prediction result, the method further includes: Performing linear merging transformation on the periodic frequency modulation instruction sequence to obtain merging item data; Input the merged item data into the pre-trained prediction model and output the frequency modulation instruction prediction result.

[0013] Furthermore, the specific formula for performing linear merging and transformation processing on the periodic frequency modulation instruction sequence is as follows:

[0014] Where X3(t) represents the merged data; θ represents the angle variable; j represents the length of the original frequency modulation instruction sequence; t represents the time variable; a j represents the first coefficient related to frequency; b j Represents the second frequency-dependent coefficient.

[0015] Furthermore, the prediction model is trained based on historical frequency modulation instruction sequences and historical frequency modulation instruction prediction results; the basic model of the prediction model is one or more combinations of a neural network model and a support vector machine model.

[0016] A frequency modulation instruction prediction system for a sodium ion supercapacitor energy storage system, comprising: A sequence acquisition module, used to acquire the original frequency modulation instruction sequence of the sodium ion supercapacitor energy storage system; An integrable processing module is used to perform approximate integrable processing on the original frequency modulation instruction sequence to obtain an approximate integrable quantity; A periodic processing module is used to perform periodic processing on the approximately integrable quantity to obtain a periodic frequency modulation instruction sequence; The prediction module is used to input the periodic frequency modulation instruction sequence into the pre-trained prediction model to obtain the frequency modulation instruction prediction result.

[0017] A device comprising: Memory for storing computer programs; A processor is used to implement the steps of the frequency modulation instruction prediction method of the above-mentioned sodium ion supercapacitor energy storage system when executing the computer program.

[0018] A computer-readable storage medium stores a computer program, which, when executed by a processor, is used to implement the steps of the frequency modulation instruction prediction method of the above-mentioned sodium ion supercapacitor energy storage system.

[0019] Compared with the prior art, the present invention has the following beneficial effects: The present invention provides a method for predicting frequency modulation instructions for a sodium ion supercapacitor energy storage system, which collects the original frequency modulation instruction sequence and performs approximate integrable processing on it to smooth the data and reduce the nonlinear influence; then the processed data is periodically processed, and the periodic characteristics of the frequency modulation instruction are used to improve the accuracy of the prediction. This processing flow can convert the originally nonlinear and irregular raw data into a form that is easier to predict. Finally, the processed periodic frequency modulation instruction sequence is input into a pre-trained prediction model to obtain an accurate frequency modulation instruction prediction result. This method effectively solves the problem of large prediction errors caused by the direct use of raw data in traditional prediction methods, improves the prediction accuracy, thereby optimizing the power allocation strategy of supercapacitors and lithium batteries, achieving the optimal configuration of energy storage resources, and significantly improving the operating efficiency and stability of the power frequency modulation system. BRIEF DESCRIPTION OF THE DRAWINGS

[0020] Figure 1 A flow chart of a method for predicting frequency modulation instructions for a sodium ion supercapacitor energy storage system provided in an embodiment of the present invention; Figure 2 A schematic diagram of approximate integrability processing provided by an embodiment of the present invention; Figure 3 A flow chart of a method for predicting frequency modulation instructions for a sodium ion supercapacitor energy storage system provided by the present invention; Figure 4 A structural schematic diagram of a frequency modulation instruction prediction system for a sodium ion supercapacitor energy storage system provided by the present invention. DETAILED DESCRIPTION

[0021] In order to further understand the content of the present invention, the present invention is described in detail below in conjunction with the accompanying drawings and specific embodiments. It should be understood that the embodiments are only for explaining the present invention and are not intended to limit it.

[0022] The following is an explanation of the technical terms involved in the present invention: The smooth function is a powerful data smoothing tool. By selecting the appropriate smoothing method and window size, it can effectively remove data noise and improve the accuracy of data analysis.

[0023] As described in the background technology, in the power frequency modulation of the supercapacitor coupled lithium battery energy storage system, the traditional frequency modulation instruction prediction method is used. The traditional prediction method often directly inputs the original frequency modulation instruction into the neural network for prediction. However, due to the nonlinearity and irregularity of the original frequency modulation instruction, direct prediction will lead to large errors. Inaccurate prediction leads to unreasonable power distribution of supercapacitors and lithium batteries in the energy storage system, and the optimal utilization of energy storage resources cannot be achieved, which affects the frequency modulation efficiency and reliability.

[0024] In order to solve the above problems, this embodiment provides a method for predicting frequency modulation instructions of a sodium ion supercapacitor energy storage system. This method optimizes the power distribution of supercapacitors and lithium batteries in the energy storage system by accurately predicting the frequency modulation instructions, thereby improving the efficiency and reliability of power frequency modulation, reducing the negative impact of frequency modulation of thermal power units, and improving the economic benefits of frequency modulation of thermal power plants using energy storage systems. The specific steps are as follows: S1, collects the original frequency modulation instruction sequence of the sodium ion supercapacitor energy storage system; S2, the original frequency modulation sequence is smoothed using the smooth function to obtain an approximate integrable quantity; this step is intended to eliminate the noise in the frequency modulation instruction data and improve the predictability of the data.

[0025] S3, convert the approximate integrable quantity X1 into a series of periodic functions through the periodic formula. The core of the periodic formula is to decompose the integrable function into periodic components of different frequencies, which is convenient for the identification and processing of the subsequent prediction model.

[0026] On the basis of periodic processing, the present invention proposes a linear merging transformation, which linearly merges the periodic components of different frequencies to form a periodic frequency modulation instruction sequence as new prediction input data. The purpose of the linear merging transformation is to reduce the dimension of the data while ensuring the periodicity of the data and improve the operating efficiency of the prediction model.

[0027] S4, with the help of machine learning algorithms such as neural networks, the periodic frequency modulation instruction sequence is input into the pre-trained prediction model to obtain the frequency modulation instruction prediction result. Among them, the prediction model is responsible for learning the periodicity and change trend of the frequency modulation instruction data to achieve accurate prediction of future frequency modulation instructions.

[0028] According to the prediction results of frequency modulation instructions, the power allocation strategy of supercapacitors and lithium batteries is formulated. In the period when the frequency modulation instructions change greatly, supercapacitors are used first for rapid response; in the period when the frequency modulation instructions are relatively stable, the participation of lithium batteries is gradually increased to improve the overall efficiency and reliability of the energy storage system.

[0029] The frequency modulation instruction prediction method provided by this embodiment is further explained below in conjunction with the accompanying drawings: For example, Figure 1 As shown, this embodiment provides a method for predicting frequency modulation instructions of a sodium ion supercapacitor energy storage system, and the specific steps are as follows: The original frequency modulation instruction sequence of the sodium ion supercapacitor energy storage system is collected. In the actual prediction process, the original frequency modulation instruction sequence here refers to the original frequency modulation instruction sequence to be predicted; in the process of building and training the prediction model, the original frequency modulation instruction sequence here can also be the historical frequency modulation instruction sequence ( Figure 1 The historical frequency modulation instruction data in the training data is used as the data source of the training data.

[0030] See also Figure 2 , Figure 2 This is a schematic diagram of approximate integrability processing. Figure 2 In the figure, the thick solid line represents the original FM sequence; the thin solid line represents the approximate integrable quantity.

[0031] The original frequency modulation instruction sequence is approximately integrable, and the approximately integrable quantity X1(t) and the non-integrable quantity X2(t) are obtained.

[0032] Among them, the integrable quantity is expressed as: X2(t)=X(t)-X1(t) Where X(t) represents the original frequency modulation instruction sequence; X1(t) represents the approximately integrable quantity; X2(t) represents the non-integrable quantity.

[0033] The approximate integrable quantity is periodized using the periodization formula to obtain a periodic frequency modulation instruction sequence. The specific formula is as follows:

[0034]

[0035] In the formula, X1(t) represents the approximately integrable quantity; represents the first coefficient associated with the approximately integrable quantity; represents the second coefficient related to the approximately integrable quantity; k represents the frequency component; t represents the time variable; j represents the length of the original frequency modulation instruction sequence.

[0036] The periodic frequency modulation instruction sequence is subjected to a linear merging transformation process, which is also called a similarity merging transformation process, to obtain merging item data X3 (t) and residual item data X4 (t).

[0037] Finally, the merged item data X3(t), the residual item data X4(t) and the integrable quantity X2(t) are input into the prediction model for prediction, and the final frequency modulation instruction prediction result is output.

[0038] In this embodiment, the prediction model is a combination of one or more neural network models and support vector machine models, and is trained based on historical frequency modulation instruction sequences and historical frequency modulation instruction prediction results.

[0039] Simulation verification was carried out for the traditional prediction method and the prediction method provided by the present invention, and the specific verification results are shown in Table 1. In Table 1, the RMSE (full name: Root-Mean-Squared Error) of the traditional prediction method for frequency modulation instruction 1 is 9.9, while that of the prediction method provided by the present invention is 1.2; the RMSE of frequency modulation instruction 2 is 8.7, while that of the prediction method provided by the present invention is 3.3; it can be seen that the use of this prediction method significantly improves the accuracy of the prediction results.

[0040] Table 1 shows the root mean square error verification results.

[0041] The prediction method provided by the present invention is further explained below in conjunction with application examples: For example, regarding the frequency modulation instruction prediction and power allocation application of a thermal power plant, the frequency modulation instruction prediction method provided by the present invention is implemented and applied, as follows: Data collection and preprocessing: The frequency modulation instruction data of a thermal power plant within a week was selected as the research object, and the data collection frequency was 100ms. First, the original frequency modulation sequence was approximately integrable, and the Gaussian smoothing function was selected as the smoothing function. According to the noise level and fluctuation of the data, the standard deviation (σ=0.05) was determined. Through the convolution operation of the Gaussian smoothing function and the original frequency modulation sequence, the approximate integrable quantity was obtained, and then the integrable quantity was calculated.

[0042] Periodic formula and linear merging transformation: For approximately integrable quantities, determine the signal period based on statistical analysis of historical data. Apply the periodic formula to calculate and When calculating a1, the number of data sampling points in one cycle is N=10 (because the sampling frequency is 100ms, there are 10 sampling points in 1s), the sampling interval is Δt=0.1s, and after calculating the values ​​of a1 and other coefficients, the periodized function X is obtained. j (t). Then, a linear merging transformation is performed, and the coefficients a=0.3 and b=0.5 are determined by the least square method based on historical data, and the merging term X3(t) and the residual term X4(t) are calculated.

[0043] Prediction model training and prediction: The Long Short-Term Memory (LSTM) model is used for prediction. Build the LSTM model, set the number of input layer nodes to 2 (corresponding to the merged term X3 (t) and the residual term X4 (t)), set the hidden layer to 3 layers, the number of nodes in each layer is 64, 32, and 16 respectively, and the number of output layer nodes is 1 (the predicted frequency modulation command value). Use the first six days of the collected historical data for one week as the training set, and the last day of data as the test set. During training, the learning rate is set to 0.001, and the number of training rounds is 100. After the training is completed, use the trained model to predict the test set data to obtain the predicted frequency modulation command value.

[0044] Power allocation decision: According to the predicted frequency modulation command value, the power allocation algorithm based on dynamic programming is used to allocate the power of supercapacitors and lithium batteries. Considering the maximum power charging power of supercapacitors is , the maximum power discharge power is , the initial SOC is ;The maximum charging power of lithium battery is , the maximum power discharge power is , the initial SOC is . Taking the minimum total energy loss of the energy storage system as the objective function and combining the above constraints, the power allocation scheme of supercapacitors and lithium batteries under different predicted frequency modulation instructions is calculated. For example, at a certain moment, the predicted frequency modulation instruction requires an output power of 80kW. According to the power allocation algorithm, the supercapacitor is allocated an output power of 60kW and the lithium battery is allocated an output power of 20kW, which not only meets the frequency modulation requirements but also optimizes the energy loss of the energy storage system.

[0045] Effect evaluation: The prediction results of the present invention were compared with those of the traditional prediction model, and the root mean square error (RMSE) was calculated. The results showed that the RMSE of the present invention on the test data of the thermal power plant was 2.5, while the RMSE of the traditional model was 7.8. At the same time, by comparing the frequency regulation effect of the energy storage system before and after the application of the present invention, it was found that after the application, the frequency fluctuation range of the power grid was reduced from ±0.2Hz to ±0.1Hz, and the number of times the thermal power units participated in frequency regulation was reduced by 35%, which effectively verified the effectiveness of the present invention.

[0046] For example, Figure 3 As shown, this embodiment also provides a method for predicting frequency modulation instructions of a sodium ion supercapacitor energy storage system, comprising the following steps: Collect the original frequency modulation instruction sequence of the sodium ion supercapacitor energy storage system; The original frequency modulation instruction sequence is processed approximately as integrable to obtain an approximately integrable quantity; The approximately integrable quantity is processed periodically to obtain a periodic frequency modulation instruction sequence; The periodic frequency modulation instruction sequence is input into the pre-trained prediction model to obtain the frequency modulation instruction prediction result.

[0047] In this embodiment, the process of performing approximate integrability processing on the original frequency modulation instruction sequence to obtain an approximate integrable quantity includes: The smooth function is used to smooth the collected original frequency modulation instruction sequence to obtain an approximate integrable quantity.

[0048] Here, the original frequency modulation instruction sequence is smoothed by using the smooth function to obtain an approximate integrable quantity, which effectively reduces the nonlinearity and volatility of the original data, makes the data more stable, provides a good foundation for subsequent processing, and improves the accuracy of the prediction.

[0049] In this embodiment, the collected original frequency modulation instruction sequence is smoothed by using the smooth function to obtain an incommensurable quantity; the incommensurable quantity is used to be input into the pre-trained prediction model simultaneously with the periodic frequency modulation instruction sequence to predict the frequency modulation instruction; wherein the incommensurable quantity is expressed as: X2(t)=X(t)-X1(t) Where X(t) represents the original frequency modulation instruction sequence; X1(t) represents the approximately integrable quantity; X2(t) represents the non-integrable quantity.

[0050] It can be seen that while smoothing, the integrable quantities are separated and input into the prediction model together with the periodic frequency modulation instruction sequence, which fully considers all the information in the original data, avoids information loss, and further improves the comprehensiveness and accuracy of the prediction.

[0051] In this embodiment, the approximately integrable quantity is subjected to periodic processing to obtain a periodic frequency modulation instruction sequence; the specific formula for the periodic processing is as follows:

[0052]

[0053] In the formula, X1(t) represents the approximately integrable quantity; represents the first coefficient associated with the approximately integrable quantity; represents the second coefficient related to the approximately integrable quantity; k represents the frequency component; t represents the time variable; j represents the length of the original frequency modulation instruction sequence.

[0054] Here, the approximate integrable quantity is converted into a periodic sequence through a specific periodic processing formula, and the periodic characteristics of the frequency modulation instruction are utilized, so that the prediction model can better capture the changing patterns of the data and improve the accuracy and stability of the prediction.

[0055] In this embodiment, before inputting the periodic frequency modulation instruction sequence into the pre-trained prediction model to obtain the frequency modulation instruction prediction result, the method further includes: Performing linear merging transformation on the periodic frequency modulation instruction sequence to obtain merging item data; Input the merged item data into the pre-trained prediction model and output the frequency modulation instruction prediction result.

[0056] It can be seen that before the periodic frequency modulation instruction sequence is input into the prediction model, a linear merging transformation is performed to obtain the merged item data. This step further simplifies the data and improves the data processability and efficiency of the prediction model.

[0057] In this embodiment, the specific formula for performing linear merging transformation processing on the periodic frequency modulation instruction sequence is as follows:

[0058] Where X3(t) represents the merged data; θ represents the angle variable; j represents the length of the original frequency modulation instruction sequence; t represents the time variable; a j represents the first coefficient related to frequency; b j Represents the second frequency-dependent coefficient.

[0059] Here, a specific formula for linear merging transformation processing is given, which provides clear guidance for practical operations, makes the data transformation process more normalized and standardized, and helps to improve the consistency and reliability of predictions.

[0060] In this embodiment, the prediction model is trained based on historical frequency modulation instruction sequences and historical frequency modulation instruction prediction results; the basic model of the prediction model is one or more combinations of a neural network model and a support vector machine model.

[0061] Here, the prediction model is trained based on the historical frequency modulation instruction sequence and the historical frequency modulation instruction prediction results, and multiple models or combinations such as neural network models and support vector machine models can be selected, making the prediction model more flexible and adaptable, and able to select the optimal model according to different application scenarios and needs, thereby improving the generalization ability and practicality of the prediction.

[0062] like Figure 4As shown, this embodiment also provides a frequency modulation instruction prediction system for a sodium ion supercapacitor energy storage system, including: a sequence acquisition module, used to acquire an original frequency modulation instruction sequence of the sodium ion supercapacitor energy storage system; an integrable processing module, used to perform approximate integrable processing on the original frequency modulation instruction sequence to obtain an approximate integrable quantity; a periodic processing module, used to perform periodic processing on the approximate integrable quantity to obtain a periodic frequency modulation instruction sequence; a prediction module, used to input the periodic frequency modulation instruction sequence into a pre-trained prediction model to obtain a frequency modulation instruction prediction result.

[0063] The present invention also provides a device, comprising: a memory for storing a computer program; and a processor for implementing the steps of the frequency modulation instruction prediction method of the sodium ion supercapacitor energy storage system when executing the computer program.

[0064] When the processor executes the computer program, the steps of predicting the frequency modulation instructions of the above-mentioned sodium ion supercapacitor energy storage system are implemented, for example: collecting the original frequency modulation instruction sequence of the sodium ion supercapacitor energy storage system; performing approximate integrability processing on the original frequency modulation instruction sequence to obtain an approximate integrable quantity; performing periodic processing on the approximate integrable quantity to obtain a periodic frequency modulation instruction sequence; inputting the periodic frequency modulation instruction sequence into a pre-trained prediction model to obtain a frequency modulation instruction prediction result.

[0065] Alternatively, the processor implements the functions of each module in the above system when executing the computer program.

[0066] Exemplarily, the computer program can be divided into one or more modules / units, and the one or more modules / units are stored in the memory and executed by the processor to complete the present invention. The one or more modules / units can be a series of computer program instruction segments that can complete preset functions, and the instruction segments are used to describe the execution process of the computer program in the frequency modulation instruction prediction device of the sodium ion supercapacity energy storage system. For example, the computer program can be divided into a sequence acquisition module, an integrable processing module, a periodic processing module and a prediction module; the specific functions of each module are as follows: a sequence acquisition module, which is used to collect the original frequency modulation instruction sequence of the sodium ion supercapacity energy storage system; an integrable processing module, which is used to perform approximate integrable processing on the original frequency modulation instruction sequence to obtain an approximate integrable quantity; a periodic processing module, which is used to perform periodic processing on the approximate integrable quantity to obtain a periodic frequency modulation instruction sequence; a prediction module, which is used to input the periodic frequency modulation instruction sequence into a pre-trained prediction model to obtain a frequency modulation instruction prediction result.

[0067] The frequency modulation instruction prediction device of the sodium ion supercapacity energy storage system can be a computing device such as a desktop computer, a notebook, a PDA, and a cloud server. The frequency modulation instruction prediction device of the sodium ion supercapacity energy storage system may include, but is not limited to, a processor and a memory. Those skilled in the art will understand that the above is an example of a frequency modulation instruction prediction device of a sodium ion supercapacity energy storage system, and does not constitute a limitation on the frequency modulation instruction prediction device of a sodium ion supercapacity energy storage system. It may include more components than the above, or a combination of certain components, or different components. For example, the frequency modulation instruction prediction device of the sodium ion supercapacity energy storage system may also include input and output devices, network access devices, buses, etc.

[0068] The processor may be a central processing unit (CPU), or other general-purpose processors, digital signal processors (DSP), application-specific integrated circuits (ASIC), field-programmable gate arrays (FPGA) or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components, etc. The general-purpose processor may be a microprocessor or the processor may be any conventional processor, etc. The processor is the control center of the frequency modulation instruction prediction of the sodium-ion supercapacitor energy storage system, and uses various interfaces and lines to connect various parts of the frequency modulation instruction prediction device of the entire sodium-ion supercapacitor energy storage system.

[0069] The memory can be used to store the computer program and / or module, and the processor realizes various functions of the frequency modulation instruction prediction device of the sodium ion supercapacitor energy storage system by running or executing the computer program and / or module stored in the memory and calling the data stored in the memory.

[0070] The memory may mainly include a program storage area and a data storage area, wherein the program storage area may store an operating system, an application required for at least one function (such as a sound playback function, an image playback function, etc.), etc.; the data storage area may store data created according to the use of the mobile phone (such as audio data, a phone book, etc.), etc. In addition, the memory may include a high-speed random access memory, and may also include a non-volatile memory, such as a hard disk, a memory, a plug-in hard disk, a smart memory card (Smart Media Card, SMC), a secure digital (Secure Digital, SD) card, a flash card (Flash Card), at least one disk storage device, a flash memory device, or other volatile solid-state storage devices.

[0071] The present invention also provides a computer-readable storage medium, which stores a computer program. When the computer program is executed by a processor, the steps of the frequency modulation instruction prediction method of a sodium ion supercapacitor energy storage system are implemented.

[0072] If the module / unit integrated in the frequency modulation instruction prediction system of the sodium ion supercapacitor energy storage system is implemented in the form of a software functional unit and sold or used as an independent product, it can be stored in a computer-readable storage medium.

[0073] Based on such understanding, the present invention implements all or part of the process in the frequency modulation instruction prediction method of the above-mentioned sodium ion supercapacitor energy storage system, and can also be completed by instructing related hardware through a computer program, and the computer program can be stored in a computer-readable storage medium, and the computer program can be executed by a processor to implement the steps of the frequency modulation instruction prediction method of the above-mentioned sodium ion supercapacitor energy storage system. Wherein, the computer program includes computer program code, and the computer program code can be in source code form, object code form, executable file or preset intermediate form, etc.

[0074] The computer-readable storage medium may include: any entity or device capable of carrying the computer program code, recording medium, USB flash drive, mobile hard disk, magnetic disk, optical disk, computer memory, read-only memory (ROM), random access memory (RAM), electrical carrier signal, telecommunication signal and software distribution medium, etc.

[0075] It should be noted that the content contained in the computer-readable storage medium can be appropriately increased or decreased according to the requirements of legislation and patent practice in the jurisdiction. For example, in some jurisdictions, according to legislation and patent practice, computer-readable storage media do not include electrical carrier signals and telecommunication signals.

[0076] The present invention provides a method for predicting frequency modulation instructions of a sodium ion supercapacitor energy storage system, which has the following advantages: First, improving prediction accuracy: through periodic processing and linear merging transformation, the present invention can significantly reduce the prediction error of the frequency modulation instruction and improve the prediction accuracy.

[0077] Second, optimize power distribution: The power distribution strategy formulated according to the prediction results can optimize the power distribution of supercapacitors and lithium batteries during the power frequency modulation process, and improve the overall efficiency and reliability of the energy storage system.

[0078] Third, reduce operating costs: This method can reduce the frequency regulation burden of thermal power units, extend their operating life, reduce coal consumption and operating costs. At the same time, by improving the efficiency and reliability of the energy storage system, it helps thermal power plants obtain more grid subsidy benefits.

[0079] Fourth, promote the grid connection of new energy: The adoption of this method can enhance the frequency stability of the power grid, provide strong support for the large-scale grid connection of new energy, and promote the optimization and transformation of the energy structure.

[0080] The above embodiment is only one of the implementation methods that can realize the technical solution of the present invention. The scope of protection claimed by the present invention is not limited only to this embodiment, but also includes changes, replacements and other implementation methods that can be easily thought of by any technician familiar with the technical field within the technical scope disclosed by the present invention.

[0081] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention rather than to limit it. Although the present invention has been described in detail with reference to the above embodiments, ordinary technicians in the relevant field should understand that the specific implementation methods of the present invention can still be modified or replaced by equivalents, and any modifications or equivalent replacements that do not depart from the spirit and scope of the present invention should be covered within the scope of protection of the present invention.

Claims

1. A method for predicting frequency modulation instructions for a sodium ion supercapacitor energy storage system, characterized in that: include: Collect the original frequency modulation instruction sequence of the sodium ion supercapacitor energy storage system; The original frequency modulation instruction sequence is processed approximately as integrable to obtain an approximately integrable quantity; The approximately integrable quantity is processed periodically to obtain a periodic frequency modulation instruction sequence; The periodic frequency modulation instruction sequence is input into the pre-trained prediction model to obtain the frequency modulation instruction prediction result.

2. The frequency modulation instruction prediction method of the sodium ion supercapacitor energy storage system according to claim 1 is characterized in that: The process of performing approximate integrability processing on the original frequency modulation instruction sequence to obtain an approximate integrable quantity includes: The smooth function is used to smooth the collected original frequency modulation instruction sequence to obtain an approximate integrable quantity.

3. The frequency modulation instruction prediction method of the sodium ion supercapacitor energy storage system according to claim 2, characterized in that: The smooth function is used to smooth the collected original frequency modulation instruction sequence, and an incommensurable quantity is also obtained; the incommensurable quantity is used to be input into the pre-trained prediction model simultaneously with the periodic frequency modulation instruction sequence to predict the frequency modulation instruction; wherein the incommensurable quantity is expressed as: X2(t)=X(t)-X1(t) Where X(t) represents the original frequency modulation instruction sequence; X1(t) represents the approximately integrable quantity; X2(t) represents the non-integrable quantity.

4. The frequency modulation instruction prediction method of the sodium ion supercapacitor energy storage system according to claim 1 is characterized in that: The periodic processing of the approximate integrable quantity to obtain the periodic frequency modulation instruction sequence, the specific formula of the periodic processing is as follows: In the formula, X1(t) represents the approximately integrable quantity; represents the first coefficient associated with the approximately integrable quantity; represents the second coefficient related to the approximately integrable quantity; k represents the frequency component; t represents the time variable; j represents the length of the original frequency modulation instruction sequence.

5. The frequency modulation instruction prediction method of the sodium ion supercapacitor energy storage system according to claim 1 is characterized in that: Before inputting the periodic frequency modulation instruction sequence into the pre-trained prediction model to obtain the frequency modulation instruction prediction result, the method further includes: Performing linear merging transformation on the periodic frequency modulation instruction sequence to obtain merging item data; Input the merged item data into the pre-trained prediction model and output the frequency modulation instruction prediction result.

6. The frequency modulation instruction prediction method of the sodium ion supercapacitor energy storage system according to claim 5 is characterized in that: The specific formula for performing linear merging transformation on the periodic frequency modulation instruction sequence is as follows: Where X3(t) represents the merged data; θ represents the angle variable; j represents the length of the original frequency modulation instruction sequence; t represents the time variable; a j represents the first coefficient related to frequency; b j Represents the second frequency-dependent coefficient.

7. The frequency modulation instruction prediction method of the sodium ion supercapacitor energy storage system according to claim 1 is characterized in that: The prediction model is trained based on historical frequency modulation instruction sequences and historical frequency modulation instruction prediction results; the basic model of the prediction model is one or more combinations of a neural network model and a support vector machine model.

8. A frequency modulation instruction prediction system for a sodium ion supercapacitor energy storage system, characterized in that: include: A sequence acquisition module, used to acquire the original frequency modulation instruction sequence of the sodium ion supercapacitor energy storage system; An integrable processing module is used to perform approximate integrable processing on the original frequency modulation instruction sequence to obtain an approximate integrable quantity; A periodic processing module is used to perform periodic processing on the approximately integrable quantity to obtain a periodic frequency modulation instruction sequence; The prediction module is used to input the periodic frequency modulation instruction sequence into the pre-trained prediction model to obtain the frequency modulation instruction prediction result.

9. A device, characterized in that: include: Memory for storing computer programs; A processor, used to implement the steps of the frequency modulation instruction prediction method of the sodium ion supercapacitor energy storage system according to any one of claims 1 to 7 when executing the computer program.

10. A computer-readable storage medium storing a computer program, characterized in that: When the computer program is executed by a processor, it is used to implement the steps of the frequency modulation instruction prediction method of the sodium ion supercapacitor energy storage system according to any one of claims 1 to 7.

Images