Multi-source data alignment method and device of power system, medium and equipment

Abstract

The invention discloses an electric power system multi-source data alignment method and device, a medium and equipment, and the method comprises the steps: firstly, obtaining N pieces of time series data (N is greater than or equal to 2), and enabling the data to be from different data sources of an electric power system; then, calculating a window average value in each time sequence data by using a sliding window, and determining a sudden change starting point of each time sequence; then, aligning all time sequence data by taking the sudden change starting points as references; and finally, acquiring the sampling frequency of each piece of time sequence data, and determining and aligning other data based on the reference frequency by taking the data with the lowest sampling frequency as a reference. Through the method, the problem that the multi-source data of the power system is difficult to align can be effectively solved, the consistency of the data from different sources on the time axis is ensured, subsequent analysis and processing are facilitated, and the accuracy and efficiency of data processing are improved.

Description

Technical Field

[0001] The present invention relates to the technical field of power systems, and in particular to a method, device, medium and equipment for aligning multi-source data in a power system. Background Art

[0002] When power system data comes from multiple different data sources, there may be different sampling frequencies and data formats. For example, some data sources use absolute time records, while others use relative time records. This difference makes it difficult to correlate and analyze the operation of each device in the system after a disturbance or failure occurs in the power grid. Summary of the invention

[0003] Based on this, it is necessary to provide a method, device, medium and equipment for aligning multi-source data in a power system to solve the problem that multi-source data in a power system is difficult to align.

[0004] A method for aligning multi-source data in a power system, the method comprising:

[0005] Acquire N time series data; wherein N≥2, and the N time series data respectively belong to different data sources of the power system;

[0006] Use the nth sliding window to slide in the nth time series data and calculate the average value in the sliding window to obtain the average values ​​of multiple windows of the nth time series data; wherein 1≤n≤N;

[0007] In the nth time series data, the mutation starting point of the nth time series data is determined in sequence according to the change trend of the average value of each consecutive m windows to obtain N mutation starting points; where m≥2;

[0008] Aligning the N time series data based on the N mutation starting points;

[0009] Acquire the sampling frequencies of the N time series data, take the time series data with the lowest sampling frequency as the benchmark data, take the sampling frequency of the benchmark data as the benchmark frequency, determine the points to be aligned in the non-benchmark data based on the benchmark frequency, and align the non-benchmark data with the benchmark data based on the points to be aligned.

[0010] In one embodiment, before using the nth sliding window to slide in the nth time series data and calculating the average value in the sliding window to obtain the average values ​​of multiple windows of the nth time series data, the method includes:

[0011] The ratio between the unit data amount calculated each time by the nth sliding window and the total data amount of the nth time series data is set to a preset ratio.

[0012] In one embodiment, before using the nth sliding window to slide in the nth time series data and calculating the average value in the sliding window to obtain the average values ​​of multiple windows of the nth time series data, the method includes:

[0013] Set the unit data amount calculated by different sliding windows to be the same each time.

[0014] In one embodiment, determining the mutation starting point of the nth time series data according to the change trend of each consecutive m window average values ​​in the nth time series data in sequence includes:

[0015] In the nth time series data, the average value of the first m consecutive windows that meet any mutation condition in chronological order is taken as the mutation interval, and the last value in the first window that is most forward in time in the mutation interval is taken as the mutation starting point.

[0016] In one embodiment, the mutation condition includes that the average values ​​of m consecutive windows have an upward trend, and the ratio between the average value of the last m-1 window average values ​​in time and the average value of the first window average value in time is greater than a preset first ratio threshold; and,

[0017] The average values ​​of m consecutive windows have a downward trend, and the ratio of the average value of the last m-1 window average values ​​in time to the average value of the first window in time is less than a preset second ratio threshold.

[0018] In one embodiment, determining the point to be aligned in the non-reference data based on the reference frequency, and aligning the non-reference data with the reference data based on the point to be aligned includes:

[0019] Taking the corresponding mutation point as the starting point in each non-basic data, and resampling the points to be aligned based on the reference frequency, so as to obtain a plurality of points to be aligned for each non-basic data;

[0020] A plurality of to-be-aligned points in each non-basic data are aligned with the data in the reference data in sequence based on a time sequence.

[0021] A multi-source data alignment device for a power system, the multi-source data alignment device for a power system comprising:

[0022] A data acquisition module, used to acquire N time series data; wherein N≥2, and the N time series data respectively belong to different data sources of the power system;

[0023] A starting point alignment module is used to slide the nth sliding window in the nth time series data and calculate the average value in the sliding window to obtain multiple window average values ​​of the nth time series data; wherein 1≤n≤N; in the nth time series data, sequentially determine the mutation starting point of the nth time series data according to the change trend of each consecutive m window average values ​​to obtain N mutation starting points; wherein m≥2; align the N time series data with the N mutation starting points as the reference;

[0024] The overall alignment module is used to obtain the sampling frequencies of the N time series data, take the time series data with the lowest sampling frequency as the benchmark data, take the sampling frequency of the benchmark data as the benchmark frequency, determine the points to be aligned in the non-benchmark data based on the benchmark frequency, and align the non-benchmark data with the benchmark data based on the points to be aligned.

[0025] A computer-readable storage medium stores a computer program. When the computer program is executed by a processor, the processor executes the steps of the above-mentioned power system multi-source data alignment method.

[0026] A multi-source data alignment device for an electric power system comprises a memory and a processor, wherein the memory stores a computer program, and when the computer program is executed by the processor, the processor executes the steps of the multi-source data alignment method for an electric power system.

[0027] The present invention provides a method, device, medium and equipment for aligning multi-source data in a power system. First, N time series data (N≥2) are obtained, and these data come from different data sources of the power system; then, a sliding window is used to calculate the window average value in each time series data to determine the mutation starting point of each time series; then, all time series data are aligned based on these mutation starting points; finally, the sampling frequency of each time series data is obtained, and the data with the lowest sampling frequency is used as a benchmark to determine and align other data based on the benchmark frequency. Through this method, the problem of difficulty in aligning multi-source data in the power system can be effectively solved, the consistency of data from different sources on the time axis can be ensured, subsequent analysis and processing are facilitated, and the accuracy and efficiency of data processing are improved. BRIEF DESCRIPTION OF THE DRAWINGS

[0028] In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings required for use in the embodiments or the description of the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative work.

[0029] in:

[0030] Figure 1 A flowchart of a multi-source data alignment method for a power system is shown;

[0031] Figure 2a The time series data with a sampling frequency of 100 Hz obtained by monitoring device A;

[0032] Figure 2b The time series data with a sampling frequency of 50 Hz obtained by monitoring device B;

[0033] Figure 3 For the general Figure 2a and Figure 2b Schematic diagram of alignment based on the mutation starting point;

[0034] Figure 4 A schematic diagram of a process for determining points to be aligned and performing alignment;

[0035] Figure 5 It is a structural schematic diagram of a multi-source data alignment device for a power system;

[0036] Figure 6 Structural block diagram of the multi-source data alignment device for the power system. DETAILED DESCRIPTION

[0037] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0038] The terms "first", "second", etc. in the specification and claims of this application and the above-mentioned drawings are used to distinguish different objects, rather than to describe a specific order. In addition, the terms "including" and "having" and any variations thereof are intended to cover non-exclusive inclusions. For example, a process, method, system, product or device that includes a series of steps or units is not limited to the listed steps or units, but optionally includes steps or units that are not listed, or optionally includes other steps or units inherent to these processes, methods, products or devices.

[0039] Reference to "embodiments" herein means that a particular feature, structure, or characteristic described in conjunction with the embodiments may be included in at least one embodiment of the present application. The appearance of the phrase in various locations in the specification does not necessarily refer to the same embodiment, nor is it an independent or alternative embodiment that is mutually exclusive with other embodiments. It is explicitly and implicitly understood by those skilled in the art that the embodiments described herein may be combined with other embodiments.

[0040] like Figure 1 As shown, Figure 1 : is a flow chart of a method for aligning multi-source data in a power system in an embodiment. The steps provided by the method for aligning multi-source data in a power system in this embodiment include:

[0041] S101, obtain N time series data.

[0042] Among them, time series data refers to a series of data points arranged in chronological order. In the power system, time series data can include various measurements, such as voltage, current, frequency, etc., which are recorded regularly at time intervals. In this example, N≥2, and these N time series data belong to different data sources of the power system. Specifically, different data sources can refer to different devices, different monitoring points or different types of sensors in the power system. For example, one data source may record voltage changes, and another data source may record frequency changes. For example, when recording voltage changes, one data source comes from device A and the other data source comes from device B.

[0043] For example, Figure 2a and Figure 2b As shown in Figure 1, we collected time series data of two different units after a short circuit fault occurred in a power system. Figure 2a The time series data is monitored by device A, with a sampling frequency of 100 Hz, and each sampling point is marked with a “+”; Figure 2b The time series data is monitored by device B, with a sampling frequency of 50 Hz, and each sampling point is marked with a “*”.

[0044] S102, using the nth sliding window to slide within the nth time series data and calculating the average value within the sliding window to obtain multiple window average values ​​of the nth time series data.

[0045] Among them, 1≤n≤N, that is, this step performs the same average calculation operation on each time series data. Each time series data has a corresponding sliding window, and the size and step size of the sliding window can be set according to actual needs. The sliding window will slide on the time series data, and multiple sliding window positions can be obtained. At each window position, the average value of the data points in the window will be calculated, which will generate multiple window averages.

[0046] In a specific embodiment, the ratio between the unit data amount calculated each time by the nth sliding window and the total data amount of the nth time series data is set to a preset ratio.

[0047] For example, assuming that the preset ratio is 0.02, the first time series data has 1000 data points, and the second time series data has 500 data points. Then the unit data volume calculated by the first sliding window each time is 1000×0.02=20 data points, and the unit data volume calculated by the second sliding window each time is 500×0.02=10 data points. This ensures that the smoothing effect is consistent on data sets of different sizes, and improves the accuracy and consistency of data analysis.

[0048] In another specific embodiment, the unit data amounts calculated each time by different sliding windows are set to be the same.

[0049] For example, the unit data amount calculated by each sliding window each time is 20 data points. This method can ensure the uniformity of calculation and simplify the processing steps, and is suitable for scenarios that require a consistent window size for data smoothing or feature extraction.

[0050] In addition, when the data volume is greater than a preset quantity threshold, a smaller preset ratio threshold value may be selected in which the window data volume of the sliding window is not greater than the total data volume. When the data volume is less than a preset quantity threshold value, a larger preset ratio threshold value may be selected in which the window data volume is not greater than the total data volume.

[0051] For example, when the amount of data is greater than 1000, the amount of window data of the sliding window can be selected to be no greater than 1% of the total amount of data. When the amount of data is less than 1000, the amount of window data is selected to be no greater than 5% of the total amount of data. This method dynamically adjusts the amount of window data of the sliding window by setting a data amount threshold, ensuring that under different data amounts, the size of the sliding window can both reasonably cover the data and ensure the accuracy and efficiency of the calculation. This processing can adapt to data of different sizes and avoid calculation errors or efficiency problems caused by too large or too small sliding windows.

[0052] It is understandable that any of the above sliding window setting methods can be selected based on actual needs.

[0053] S103, in the nth time series data, determine the mutation starting point of the nth time series data according to the change trend of each continuous m window average value in order to obtain N mutation starting points.

[0054] Among them, m≥2, the change trend can be rising, falling or stable, and the mutation starting point refers to the location where a significant change occurs in the time series data. The determination method here may include: 1. First calculate the cumulative sum of the deviations between the data points and the target value (or mean), and when the cumulative sum exceeds the control limit, determine the mutation starting point; 2. Use Kalman filtering to filter the time series data and estimate the system state. When the filter residual (the difference between the actual value and the estimated value) increases significantly, determine the mutation starting point. 3. Use machine learning algorithms, such as cluster analysis, neural networks, etc., to automatically detect mutation points in time series data. These methods can be used alone or in combination in practical applications to improve the accuracy and robustness of mutation detection.

[0055] In a specific embodiment, the specific steps for determining the mutation starting point are: in the nth time series data, the average value of the first m consecutive windows that meet any mutation condition in chronological order is used as the mutation interval, and the last value in the first window that is closest in time to the mutation interval is used as the mutation starting point.

[0056] That is to say, when it is first found that the average values ​​of m consecutive windows meet the mutation condition, these window average values ​​are used as the mutation interval. For example, assuming that m=5, the consecutive window average values ​​include A1, A2, A3, A4, A5, A6, A7, etc. If A2, A3, A4, A5, A6 first meet any mutation condition in the time sequence, then [A2, A3, A4, A5, A6] is the mutation interval, and then in the nth time series data, the last value in the window corresponding to A2 is selected as the mutation starting point.

[0057] Optionally, the mutation conditions include that the average values ​​of m consecutive windows have an upward trend, and the ratio of the average value of the last m-1 window average values ​​in time to the average value of the first window average value in time is greater than a preset first ratio threshold; and, the average values ​​of m consecutive windows have a downward trend, and the ratio of the average value of the last m-1 window average values ​​in time to the average value of the first window average value in time is less than a preset second ratio threshold.

[0058] For example, m=5, the preset first ratio threshold is 1.02, and the second ratio threshold is 0.98. If the average value changes of five consecutive time windows have the same trend for the first time, for example, W k <W k+1 <W k+2 <W k+3 <W k+4 And 1 / 4(W k+1 +W k+2 +W k+3 +W k+4 )>1.02*W k, or W k >W k+1 >W k+2 >W k+3 >W k+4 And 1 / 4(W k+1 +W k+2 +W k+3 +W k+4 )<0.98*W k , then we think that W k , W k+1 , W k+2 , W k+3 , W k+4 If the mutation condition is met, then in the nth time series data, W k The last value in the corresponding window is used as the mutation starting point.

[0059] Optionally, the mutation condition can also be that the change in the window average exceeds a certain absolute numerical threshold. If the window average changes from 50 to 65, the absolute change is 15, which exceeds the set threshold of 10, then the mutation condition is considered to be met. Alternatively, the change rate of the window average exceeds a certain relative percentage threshold. If the window average changes from 50 to 70, the change rate is (70-50) / 50=40%, which exceeds the set threshold of 20%, then the mutation condition is met. Of course, other determination methods are also possible.

[0060] This step can more comprehensively identify mutation points in time series data and improve the accuracy and reliability of data analysis.

[0061] S104, aligning the N time series data based on the N mutation starting points.

[0062] like Figure 3 As shown, corresponding to Figure 2a and Figure 2b For the example, we get Figure 2a The mutation starting point is 3.05s, Figure 2b The mutation starting point is 13.04s, so we compare these two time points and get the following: Figure 3 Schematic diagram shown.

[0063] In this way, the time reference points when abnormal failures occur in different data can be unified, so that these data can be compared and analyzed in the same time frame, thereby more accurately reflecting the overall behavior of the system at the moment of mutation.

[0064] S105, obtaining the sampling frequencies of N time series data, taking the time series data with the lowest sampling frequency as the benchmark data, taking the sampling frequency of the benchmark data as the benchmark frequency, determining the points to be aligned in the non-benchmark data based on the benchmark frequency, and aligning the non-benchmark data with the benchmark data based on the points to be aligned.

[0065] Corresponds to Figure 2a and Figure 2b An embodiment of Figure 2a The sampling frequency is 100Hz, Figure 2b The sampling frequency is 50Hz, then Figure 2b The time series data of Figure 2a The time series data is used as non-reference data, 50Hz is used as the reference frequency, and then Figure 2a The points to be aligned are selected from the time series data, and the non-benchmark data are aligned with the benchmark data based on these points to be aligned.

[0066] In a specific embodiment, Figure 4 As shown, the specific steps of determining the points to be aligned and aligning them include:

[0067] S1051, taking the corresponding mutation point in each non-basic data as the starting point, and resampling the points to be aligned based on the reference frequency, so as to obtain a plurality of points to be aligned for each non-basic data.

[0068] For example, Figure 2a The sampling frequency is 100Hz, and the original data point sequence is: 1, 2, 3, 4, 5, 6, 7, 8..., Figure 2b The sampling frequency is 50Hz, then Figure 2a After secondary sampling, sampling points 1, 3, 5, and 7 will be used as points to be aligned.

[0069] S1051, aligning multiple points to be aligned in each non-basic data with the data in the benchmark data in chronological order.

[0070] That is to say, Figure 2a The data points 1, 3, 5, 7, etc. Figure 2b Data points 1, 2, 3, 4, etc. are aligned in chronological order:

[0071] Figure 2a Align the first point to be aligned (1) Figure 2b The first reference point (1)

[0072] Figure 2a Align the second point to be aligned (3) Figure 2b The second reference point (2)

[0073] Figure 2aAlign the third point to be aligned (5) Figure 2b The third reference point (3)

[0074] Figure 2a Align the fourth point to be aligned (7) Figure 2b The fourth reference point (4)

[0075] This ensures that data with different sampling frequencies can be compared on the same time basis, thus avoiding analysis errors caused by different sampling frequencies.

[0076] The above-mentioned method for aligning multi-source data in a power system includes the following steps: first, obtaining N time series data (N≥2), which are respectively from different data sources of the power system; then, using a sliding window to calculate the window average in each time series data to determine the mutation starting point of each time series; then, aligning all time series data based on these mutation starting points; finally, obtaining the sampling frequency of each time series data, taking the data with the lowest sampling frequency as the benchmark, and determining and aligning other data based on the benchmark frequency. Through this method, the problem of difficulty in aligning multi-source data in the power system can be effectively solved, ensuring the consistency of data from different sources on the time axis, facilitating subsequent analysis and processing, and improving the accuracy and efficiency of data processing.

[0077] In one embodiment, Figure 5 As shown, a multi-source data alignment device for a power system is proposed, the device comprising:

[0078] The data acquisition module 501 is used to acquire N time series data; wherein N≥2, and the N time series data respectively belong to different data sources of the power system;

[0079] A starting point alignment module is used to slide the nth sliding window in the nth time series data and calculate the average value 502 in the sliding window to obtain multiple window average values ​​of the nth time series data; wherein 1≤n≤N; in the nth time series data, sequentially determine the mutation starting point of the nth time series data according to the change trend of each consecutive m window average values ​​to obtain N mutation starting points; wherein m≥2; align the N time series data with the N mutation starting points as a reference;

[0080] The overall alignment module 503 is used to obtain the sampling frequencies of the N time series data, take the time series data with the lowest sampling frequency as the benchmark data, take the sampling frequency of the benchmark data as the benchmark frequency, determine the points to be aligned in the non-benchmark data based on the benchmark frequency, and align the non-benchmark data with the benchmark data based on the points to be aligned.

[0081] Figure 6FIG. 2 shows an internal structure diagram of a multi-source data alignment device for a power system in an embodiment. Figure 6 As shown, the power system multi-source data alignment device includes a processor, a memory and a network interface connected via a system bus. Among them, the memory includes a non-volatile storage medium and an internal memory. The non-volatile storage medium of the power system multi-source data alignment device stores an operating system and may also store a computer program. When the computer program is executed by the processor, the processor can implement the power system multi-source data alignment method. The internal memory may also store a computer program. When the computer program is executed by the processor, the processor can execute the power system multi-source data alignment method. Those skilled in the art can understand that Figure 6 The structure shown in the figure is only a block diagram of a partial structure related to the scheme of the present application, and does not constitute a limitation on the multi-source data alignment device of the power system to which the scheme of the present application is applied. The specific multi-source data alignment device of the power system may include more or fewer components than those shown in the figure, or combine certain components, or have a different arrangement of components.

[0082] A computer-readable storage medium stores a computer program, which implements the following steps when executed by a processor: obtaining N time series data; wherein N≥2, and the N time series data respectively belong to different data sources of an electric power system; using an nth sliding window to slide in the nth time series data and calculating the average value in the sliding window to obtain multiple window average values ​​of the nth time series data; wherein 1≤n≤N; in the nth time series data, sequentially determining the mutation starting point of the nth time series data according to the change trend of each consecutive m window average values ​​to obtain N mutation starting points; wherein m≥2; aligning the N time series data with the N mutation starting points as a benchmark; obtaining the sampling frequency of the N time series data, taking the time series data with the lowest sampling frequency as the benchmark data, taking the sampling frequency of the benchmark data as the benchmark frequency, determining the points to be aligned in the non-benchmark data based on the benchmark frequency, and aligning the non-benchmark data with the benchmark data based on the points to be aligned.

[0083] A multi-source data alignment device for an electric power system comprises a memory, a processor and a computer program stored in the memory and executable on the processor, wherein the processor implements the following steps when executing the computer program: obtaining N time series data; wherein N≥2, and the N time series data respectively belong to different data sources of the electric power system; using an nth sliding window to slide in the nth time series data and calculating the average value in the sliding window to obtain multiple window average values ​​of the nth time series data; wherein 1≤n≤N; in the nth time series data, sequentially determining the mutation starting point of the nth time series data according to the change trend of each continuous m window average value to obtain N mutation starting points; wherein m≥2; aligning the N time series data with the N mutation starting points as a benchmark; obtaining the sampling frequency of the N time series data, taking the time series data with the lowest sampling frequency as the benchmark data, taking the sampling frequency of the benchmark data as the benchmark frequency, determining the point to be aligned in the non-benchmark data based on the benchmark frequency, and aligning the non-benchmark data with the benchmark data based on the point to be aligned.

[0084] It should be noted that the above-mentioned power system multi-source data alignment method, device, equipment and computer-readable storage medium belong to a general inventive concept, and the contents in the embodiments of the power system multi-source data alignment method, device, equipment and computer-readable storage medium are applicable to each other.

[0085] Those skilled in the art can understand that all or part of the processes in the above-mentioned embodiment methods can be completed by instructing the relevant hardware through a computer program, and the program can be stored in a non-volatile computer-readable storage medium. When the program is executed, it can include the processes of the embodiments of the above-mentioned methods. Among them, any reference to memory, storage, database or other media used in the embodiments provided in this application can include non-volatile and / or volatile memory. Non-volatile memory may include read-only memory (ROM), programmable ROM (PROM), electrically programmable ROM (EPROM), electrically erasable programmable ROM (EEPROM) or flash memory. Volatile memory may include random access memory (RAM) or external cache memory. As an illustration and not limitation, RAM is available in many forms, such as static RAM (SRAM), dynamic RAM (DRAM), synchronous DRAM (SDRAM), double data rate SDRAM (DDRSDRAM), enhanced SDRAM (ESDRAM), synchronous link (Synchlink) DRAM (SLDRAM), memory bus (Rambus) direct RAM (RDRAM), direct memory bus dynamic RAM (DRDRAM), and memory bus dynamic RAM (RDRAM).

[0086] The technical features of the above embodiments may be combined arbitrarily. To make the description concise, not all possible combinations of the technical features in the above embodiments are described. However, as long as there is no contradiction in the combination of these technical features, they should be considered to be within the scope of this specification.

[0087] The above embodiments only express several implementation methods of the present application, and the descriptions thereof are relatively specific and detailed, but they cannot be understood as limiting the scope of the present application. It should be pointed out that, for a person of ordinary skill in the art, several modifications and improvements can be made without departing from the concept of the present application, and these all belong to the protection scope of the present application. Therefore, the protection scope of the present application shall be subject to the attached claims.

Claims

1. A method for aligning multi-source data in a power system, characterized in that: The method comprises: Acquire N time series data; wherein N≥2, and the N time series data respectively belong to different data sources of the power system; Use the nth sliding window to slide in the nth time series data and calculate the average value in the sliding window to obtain the average values ​​of multiple windows of the nth time series data; wherein 1≤n≤N; In the nth time series data, the mutation starting point of the nth time series data is determined in sequence according to the change trend of the average value of each consecutive m windows to obtain N mutation starting points; where m≥2; Aligning the N time series data based on the N mutation starting points; Acquire the sampling frequencies of the N time series data, take the time series data with the lowest sampling frequency as the benchmark data, take the sampling frequency of the benchmark data as the benchmark frequency, determine the points to be aligned in the non-benchmark data based on the benchmark frequency, and align the non-benchmark data with the benchmark data based on the points to be aligned.

2. The method according to claim 1, characterized in that Before the nth sliding window is used to slide in the nth time series data and the average value in the sliding window is calculated to obtain the average values ​​of multiple windows of the nth time series data, the method includes: The ratio between the unit data amount calculated each time by the nth sliding window and the total data amount of the nth time series data is set to a preset ratio.

3. The method according to claim 1, characterized in that Before the nth sliding window is used to slide in the nth time series data and the average value in the sliding window is calculated to obtain the average values ​​of multiple windows of the nth time series data, the method includes: Set the unit data amount calculated by different sliding windows to be the same each time.

4. The method according to claim 1, characterized in that: The step of determining the mutation starting point of the nth time series data according to the change trend of the average values ​​of each of the consecutive m windows in the nth time series data comprises: In the nth time series data, the average value of the first m consecutive windows that meet any mutation condition in chronological order is taken as the mutation interval, and the last value in the first window that is most forward in time in the mutation interval is taken as the mutation starting point.

5. The method according to claim 4, characterized in that The mutation condition includes that the average values ​​of m consecutive windows have an upward trend, and the ratio between the average value of the last m-1 window average values ​​in time and the average value of the first window average value in time is greater than a preset first ratio threshold; and, The average values ​​of the consecutive m windows have a downward trend, and the ratio of the average value of the last m-1 window average values ​​in time to the average value of the first window in time is less than a preset second ratio threshold.

6. The method according to claim 1, characterized in that The step of determining the point to be aligned in the non-reference data based on the reference frequency, and aligning the non-reference data with the reference data based on the point to be aligned includes: Taking the corresponding mutation point as the starting point in each non-basic data, and resampling the points to be aligned based on the reference frequency, so as to obtain a plurality of points to be aligned for each non-basic data; A plurality of to-be-aligned points in each non-basic data are aligned with the data in the reference data in sequence based on a time sequence.

7. A multi-source data alignment device for a power system, characterized in that: The power system multi-source data alignment device comprises: A data acquisition module, used to acquire N time series data; wherein N≥2, and the N time series data respectively belong to different data sources of the power system; A starting point alignment module is used to slide the nth sliding window in the nth time series data and calculate the average value in the sliding window to obtain multiple window average values ​​of the nth time series data; wherein 1≤n≤N; in the nth time series data, sequentially determine the mutation starting point of the nth time series data according to the change trend of each consecutive m window average values ​​to obtain N mutation starting points; wherein m≥2; align the N time series data with the N mutation starting points as the reference; The overall alignment module is used to obtain the sampling frequencies of the N time series data, take the time series data with the lowest sampling frequency as the benchmark data, take the sampling frequency of the benchmark data as the benchmark frequency, determine the points to be aligned in the non-benchmark data based on the benchmark frequency, and align the non-benchmark data with the benchmark data based on the points to be aligned.

8. A computer-readable storage medium, characterized in that: A computer program is stored, and when the computer program is executed by a processor, the processor is caused to execute the steps of the method according to any one of claims 1 to 6.

9. A multi-source data alignment device for a power system, characterized in that: The method comprises a memory and a processor, wherein the memory stores a computer program, and when the computer program is executed by the processor, the processor executes the steps of the method according to any one of claims 1 to 6.

Images