Power load data processing method and related equipment

Abstract

The invention provides a power load data processing method and related equipment. Specifically, the processing method comprises the following steps: acquiring a first data set of a power load; preprocessing the first data set to obtain a second data set; determining a load image according to the second data set and a preset dimension rule; and processing the load image by using an image compression algorithm to obtain compressed data. By adopting the mode, the data of the power load is converted into the load image, so that regulation and control personnel can be assisted to quickly grasp huge or multi-dimensional data; and meanwhile, the load image is compressed by using an image compression algorithm to obtain compressed data, so that the technical effects of reducing the data storage requirement and improving the data transmission efficiency are achieved.

Description

Technical Field

[0001] The present disclosure relates to the field of data processing technology, and in particular to a method for processing power load data and related equipment. Background Art

[0002] The power industry is an important basic industry for national development and plays a vital role in the country's development. It is directly related to the lives of residents, industrial and agricultural production, and social development. With the continuous improvement of the level of intelligent power grids, the application of 5G communication technology in the power industry has gradually matured. The measurement data collected and transmitted by the power system has shown explosive growth, resulting in a large amount of storage space occupied by stored data, and a large amount of channel bandwidth occupied by transmitted data, with relatively high economic costs. The massive load data in the power industry comes from different scenarios and sources, including traditional power load data, renewable energy data, electric vehicle charging load data, distributed energy data, etc. These data are usually characterized by high dimensionality, high density, and high real-time performance, which poses challenges to traditional data storage and processing technologies. Summary of the invention

[0003] In view of this, the purpose of the present disclosure is to provide a method for processing power load data and related equipment.

[0004] Based on the above purpose, the present disclosure provides a method for processing power load data, comprising:

[0005] Acquiring a first data set of power load;

[0006] Preprocessing the first data set to obtain a second data set;

[0007] determining a load image according to the second data set and a preset dimensional rule;

[0008] The load image is processed using an image compression algorithm to obtain compressed data.

[0009] Based on the same inventive concept, an embodiment of the present disclosure also provides an electronic device, including a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein when the processor executes the program, the processing method described in any one of the above items is implemented.

[0010] Based on the same inventive concept, an embodiment of the present disclosure further provides a non-transitory computer-readable storage medium, wherein the non-transitory computer-readable storage medium stores computer instructions, and the computer instructions are used to enable a computer to execute any of the above-mentioned processing methods.

[0011] From the above, it can be seen that the present disclosure provides a method for processing power load data and related equipment, which obtains a first data set of power load; pre-processes the first data set to obtain a second data set; determines a load image based on the second data set and preset dimensional rules; and uses an image compression algorithm to process the load image to obtain compressed data. In this way, the power load data is converted into a load image, which can assist control personnel in quickly grasping large or multi-dimensional data; at the same time, the load image is compressed using an image compression algorithm to obtain compressed data, thereby achieving the technical effect of reducing data storage requirements and improving data transmission efficiency. BRIEF DESCRIPTION OF THE DRAWINGS

[0012] In order to more clearly illustrate the technical solutions in the present disclosure or related technologies, the drawings required for use in the embodiments or related technical descriptions are briefly introduced below. Obviously, the drawings described below are only embodiments of the present disclosure. For ordinary technicians in this field, other drawings can be obtained based on these drawings without creative work.

[0013] Figure 1 A schematic diagram showing a load image of a user provided by an embodiment of the present disclosure;

[0014] Figure 2 A schematic diagram showing another user load image provided by an embodiment of the present disclosure;

[0015] Figure 3 A schematic diagram showing a flow chart of a method for processing power load data provided by an embodiment of the present disclosure;

[0016] Figure 4 A schematic structural diagram of an electronic device provided by an embodiment of the present disclosure is shown. DETAILED DESCRIPTION

[0017] In order to make the objectives, technical solutions and advantages of the present disclosure more clearly understood, the present disclosure is further described in detail below in combination with specific embodiments and with reference to the accompanying drawings.

[0018] It should be noted that, unless otherwise defined, the technical terms or scientific terms used in the embodiments of the present disclosure should be understood by people with ordinary skills in the field to which the present disclosure belongs. The "first", "second" and similar words used in the embodiments of the present disclosure do not indicate any order, quantity or importance, but are only used to distinguish different components. "Including" or "comprising" and similar words mean that the elements or objects appearing before the word cover the elements or objects listed after the word and their equivalents, without excluding other elements or objects. "Connect" or "connected" and similar words are not limited to physical or mechanical connections, but can include electrical connections, whether direct or indirect. "Up", "down", "left", "right" and the like are only used to indicate relative positional relationships. When the absolute position of the described object changes, the relative positional relationship may also change accordingly.

[0019] In order to facilitate understanding of the technical solutions of the present disclosure, some technical terms involved in the present disclosure are introduced below.

[0020] Fast Fourier Transform (FFT) is an efficient algorithm for calculating Discrete Fourier Transform (DFT). The FFT algorithm is based on a divide-and-conquer strategy. By sorting the time domain signal sequence into even-odd numbers and using the characteristics of periodicity and symmetry, the calculation of the discrete Fourier transform is divided into several steps, thereby greatly improving the calculation efficiency. This algorithm is not only of great significance in theory, but also widely used in practice in various occasions that require frequency domain analysis, such as audio processing, image analysis, and signal processing in communication systems.

[0021] HSV (Hue, Saturation, Value) is a color space, also known as HSV color space or HSV color system. It decomposes color into three basic attributes: hue, saturation, and brightness (value) based on the way the human visual system perceives color. Hue represents the basic attribute of color, that is, the position of the spectral color, and the value range is 0° to 360°. Saturation represents the purity or vividness of the color, and the value range is 0.0 to 1.0. Brightness represents the lightness or darkness of the color, and the value range is 0.0 (black) to 1.0 (white).

[0022] The RGB color model is a color standard in the industry. It obtains a variety of colors by changing the three color channels of red (R), green (G), and blue (B) and superimposing them on each other. RGB represents the colors of the three channels of red, green, and blue. This standard covers almost all colors that can be perceived by human vision and is one of the most widely used color systems.

[0023] As described in the background technology section, the difficulty of transmitting and storing power load data is increasing. In the related art, the data acquisition unit obtains the data acquisition instruction; then, the data processing unit starts at a fixed time point according to the instruction and collects the power load data at a preset time interval; then, the encoding unit performs LZ encoding on the collected data, and stores the encoded data in a temporary database through the communication module; finally, when the preset upload conditions are met, the data storage unit uploads the power load data stored in the temporary database to the server. Such a data compression and storage method will have the risk of data loss. Although LZ encoding can compress data, it may cause information loss in some cases. Especially for data with strong randomness, the compression effect may not be ideal, and it may even happen that the amount of compressed data is larger than the original data.

[0024] In view of this, the disclosed embodiment provides a method for processing power load data and related equipment, which obtains a first data set of power load; pre-processes the first data set to obtain a second data set; determines a load image according to the second data set and preset dimensional rules; and processes the load image using an image compression algorithm to obtain compressed data. In this way, the power load data is converted into a load image, which can assist control personnel in quickly grasping large or multi-dimensional data; at the same time, the load image is compressed using an image compression algorithm to obtain compressed data, thereby achieving the technical effect of reducing data storage requirements and improving data transmission efficiency, and can provide strong support for the safe and stable operation of the power system.

[0025] In order to make the technical solution of the present disclosure clearer and easier to understand, a method for processing power load data provided by an embodiment of the present disclosure is described in detail below with reference to the accompanying drawings.

[0026] The method for processing power load data specifically includes:

[0027] First, the original data set of power load is obtained.

[0028] It should be noted that the original data set includes sampling time and load. The present disclosure does not limit the specific recording form of sampling time and load. For example, the sampling time can be expressed as 202X year X month X day X hour X minute, or it can be recorded as a string of numbers.

[0029] Taking into account the reasons such as the failure of the collection equipment and the negligence of the collection personnel, the original data set often has problems such as missing data, data anomalies, and data duplication. The quality of the data directly affects the effect of the subsequent data analysis, so it is necessary to improve the data quality by repairing or removing it. In other words, the original data set needs to be preprocessed to obtain data suitable for subsequent processing.

[0030] The original data set may have missing values. There are many reasons for missing values, mainly mechanical reasons and human reasons. Mechanical reasons are data missing due to failure in data collection or storage caused by mechanical reasons.

[0031] In some embodiments, based on the sampling times corresponding to the missing values, a plurality of reference times are determined;

[0032] The missing values ​​are corrected according to a prediction formula and the loads corresponding to the multiple reference times; wherein the prediction formula is determined by performing a linear regression analysis based on historical load data.

[0033] For example, if the sampling time is a certain time on a certain day, the reference time can be the time before and after the time, or the same time before and after the time. The corresponding prediction formula is as follows:

[0034] x d，t =α1x d-1，t +α2x d+1，t +α3x d，t-1 +α4x d，t+1 (1)

[0035] In the formula, x d，t is the load data at time t on the dth day, and α is the weight coefficient.

[0036] Among them, α1, α2, α3, and α4 can be determined based on historical data analysis, such as power load data for the same period in history, by performing linear regression calculations on the historical data.

[0037] It should be noted that the historical data here is not limited to the data of the same historical period, but can also be data with the same climatic conditions, which is not limited in the present disclosure.

[0038] Electricity load data usually has a certain fluctuation range, even if some users have some small changes in electricity consumption. Occasionally, data that exceeds the fluctuation range may be an outlier caused by a malfunction in the measuring equipment. In order to avoid outliers interfering with the normalized data, outliers need to be processed. If outliers are not processed, they may cause the range of normalized data to become very small, making the color range of the data on the image also very small. Therefore, in order to ensure that the color range of the data on the image is large enough, outliers need to be processed.

[0039] In some embodiments, median absolute difference (MAD) is used to identify outliers. Here, MAD is a method used to measure data variation. It is defined as follows: MAD = median absolute difference (MAD) i (|x i -median j (x j )|). Where x i 、x j For load data.

[0040] Specifically, first find the median of the data, then calculate the absolute difference between each data point and the median, and then find the median of these differences. If the absolute difference between a data point and the median exceeds a preset multiple of the median difference, such as 2.5 times, 3 times, etc., then this data point can be considered an outlier.

[0041] Compared with the method of using the mean and standard deviation to judge outliers, the MAD method is more robust because the mean and standard deviation are easily affected by outliers, while the median is less affected. Therefore, the MAD method based on the median works better when there are individual extreme outliers in the data set.

[0042] For outliers, a similar method to missing values ​​can be used to process them, using formula (1) to calculate the value to replace the outliers. Of course, the average value, median value, etc. can also be used to replace the outliers, and this disclosure does not limit this.

[0043] In the original data set, if a user has multiple load data at a certain moment, it is necessary to perform a deduplication operation during preprocessing. In some embodiments, the average value of the load data at the same moment is calculated as the load data of the user at that moment.

[0044] Considering that the load fluctuation ranges of users with different voltage levels, types, and sizes in the distribution network are different, in order to map the load data of different users into the same color range, it is necessary to normalize the load data of each user so that the color variation range of the load data of different users on the image is relatively consistent, which is easier to compare and analyze.

[0045] In some embodiments, a min-max normalization method is used, which can map the load data of each user to a range from 0 to 1:

[0046] In the formula, x * are normalized values, x and x min 、x max They are respectively the load data before normalization, the maximum value, and the minimum value in the user load data.

[0047] It should be noted that the above-mentioned normalization processing is an indispensable step, and the processing of missing values, outliers and duplicate values ​​is an optional step in the preprocessing process. If there are missing values, outliers and duplicate values ​​in the original data set, corresponding processing will be performed. If not, no processing is required.

[0048] Next, the load image can be determined based on the normalized raw data and the pre-set dimensionality rules.

[0049] Since HSV is more consistent with visual perception of color attributes, the preprocessed data values ​​are converted to HSV color values.

[0050] In some embodiments, the normalized load data θ i Convert to HSV color value (h i ,s i , v i ) is:

[0051]

[0052] Among them, θ i =0.04 is the break point, at which the color can quickly change from blue to black, θ i =0 is black, θ i =1 is red.

[0053] The HSV value (h i ,s i , v i ) to RGB value (r i , g i , b i ); finally, the RGB values ​​are used to generate a load image according to a preset dimension rule. Here, the preset dimension rule is the correspondence between time and dimension.

[0054] The following is an exemplary description of the method of generating a load image in combination with specific dimensional rules.

[0055] The dimension rule may be that the first dimension (eg, the horizontal axis) of the plane image represents the first to last day of the year, and the second dimension (eg, the vertical axis) represents the sampling time of each day. Figure 1 A schematic diagram of a user load image provided by an embodiment of the present disclosure is shown. Figure 1 Taking the metal products industry as an example, the horizontal axis represents 365 days a year, and the vertical axis represents 96 sampling points every day. Figure 1 The pixel color in the middle corresponds to the normalized load data of a sampling point on a certain day.

[0056] The dimensional rule may also be that the first dimension (eg, horizontal axis) of the plane image represents the sampling time of each day of the seven days of the week, and the second dimension (eg, vertical axis) represents the first to last week of the year. Figure 2 A schematic diagram of another user load image provided by an embodiment of the present disclosure is shown. Figure 2 As described above, taking the metal products industry user as an example, the horizontal axis represents the sampling points of seven days a week (a total of 96×7), and the vertical axis represents 52 weeks a year (the last day can be discarded for ease of display), and the pixel color corresponds to the size of the normalized load data value.

[0057] The load image can quickly understand the distribution of data, which helps to discover the periodicity and regularity in the load data, such as daily characteristics, weekly characteristics and annual characteristics, so as to better understand the operation of the power system, and facilitate timely detection of problems and optimization of scheduling. For example, the morning peak and evening peak occur at 10 am and 8 pm, and the load at night is much smaller than the load during the day, which is closely related to the laws of social production and life; weekly load characteristics: Monday to Friday are working days, Saturday to Sunday are rest days, and the load changes significantly within a week, which cycles in weeks; annual load characteristics: the load is highest in winter, followed by summer, and lowest in spring and autumn. In addition, holidays (such as Spring Festival, National Day, etc.) will have a significant impact on power load.

[0058] It should be noted that the load image can be displayed in a variety of common image formats such as PNG, JPG, TIF, BMP, etc., and the present disclosure does not limit this.

[0059] The present disclosure also provides a method for reconstructing power load data based on a load image. For example, the RGB value of a pixel in the load image is obtained; the RGB value of the pixel is converted into an HSV value; the HSV value is converted into a normalized load; and the load data value is obtained by denormalizing it.

[0060]

[0061] Due to the massive amount of power load data, it is urgent to improve the storage capacity of power system data. In the exemplary embodiment of the present disclosure, a compression algorithm based on fast Fourier transform and improved PCA technology is proposed to achieve compression of power data load images and reduce data storage pressure. It should be noted that PCA, or principal component analysis, is a statistical method used for dimensionality reduction of data. It transforms linearly correlated variables in the original feature space into new linearly independent variables through orthogonal transformation. These new variables are called principal components, which are linear combinations of the original variables. The main goal of principal component analysis is to use fewer variables to explain most of the variation in the original data, simplify complex data sets, and retain important features in the data sets.

[0062] The compression method specifically comprises:

[0063] First, the original load image is grayed and transformed into the frequency domain through FFT to obtain a frequency domain image. Here, a single-channel grayscale image is conducive to frequency domain processing.

[0064] For example, assuming that the grayscale image is f(x,y), its two-dimensional FFT transform is F(u,v), where u and v are coordinates in the frequency domain, and x and y are coordinates in the spatial domain. The expression is:

[0065]

[0066] Where M and N are the width (number of columns) and height (number of rows) of the image, respectively. F(u,v) is a complex-valued function consisting of a real part and an imaginary part. Where i represents the imaginary unit.

[0067] Next, the frequency domain image is divided into g m×m matrices to obtain a data set containing g m×m matrix blocks. Each matrix block will be regarded as an m 2 dimensional data samples, such as small matrices M1, M2, …, M g It should be noted that these small matrices are extracted from the original frequency domain images, and their sizes are all m×m.

[0068] Then, each small matrix M i is expanded into a vector of m rows, containing m elements. These elements are obtained from M i According to the preset rules (such as taking the mean of each row, or selecting the elements at a specific position in each row), select the elements from m 2 Select m elements from the elements to form a new vector; arrange the new vector in columns to construct a frequency domain matrix. Here, the frequency domain matrix is ​​an m-row and g-column matrix.

[0069] Here, block processing can improve efficiency, reduce computational complexity and better capture local features of the image.

[0070] Next, the block-processed frequency domain image (corresponding to the frequency domain matrix) is subjected to PCA transformation.

[0071] The specific steps of PCA transformation are described in detail below.

[0072] In some embodiments, the PCA transformation includes:

[0073] (1) The frequency domain matrix containing g samples is centralized to obtain a central matrix S, for example:

[0074] Among them, x ij Represents the element in the i-th row and j-th column.

[0075] (2) Calculate the sample covariance matrix of the centered matrix S:

[0076]

[0077] In the formula, i, j = 1, 2, 3, ..., g; i and j represent the index of the row and column in the covariance matrix, m is the number of features (dimensions) in the sample data; g represents the number of sample data.

[0078] (3) Based on the sample covariance matrix, calculate the eigenvalues ​​and eigenvectors of the covariance matrix Σ by solving the following characteristic equation:

[0079] det(∑-λI)=0 (7)

[0080] Where I is the unit matrix and λ is the eigenvalue. Thus, the eigenvalue λi and eigenvector of the covariance matrix can be obtained, and the eigenvector corresponding to the largest eigenvalue is selected to form a new eigenvector matrix. The unit eigenvector corresponding to each eigenvalue λi is a i , then the i-th principal component F i For: F i =Sa i .

[0081] (4) Calculate the cumulative contribution rate, where the cumulative contribution rate α i for:

[0082] Optionally, a principal component with a cumulative contribution rate of 0.999 is selected to form a transformation matrix. Here, if the dimension of the matrix after block division is relatively large, the amount of principal component information contained is relatively large, and discarding a block will lose more information, and the restored image will be more distorted. Therefore, the cumulative contribution rate should be selected with a larger value, otherwise, the cumulative contribution rate can be selected with a smaller value.

[0083] (5) The frequency domain matrix is ​​transformed into the principal component projection space using the transformation matrix to obtain the image data after dimensionality reduction.

[0084] It can be seen that the image is transformed to the frequency domain through the fast Fourier transform (FFT) algorithm; then the frequency domain image is reduced in dimension using PCA to achieve image compression. If Size represents the size of the image, FD is the data after dimensionality reduction, UL is the transformation matrix, Mean is the mean matrix, and End is the end sign symbol, the data format between PCA and PCA inverse transformation is composed of Size+FD+UL+Mean+End. In other words, the compressed data can be stored in such a data format, which can effectively reduce the pressure of data storage and save the space required for data storage, thereby improving the efficiency of data management and reducing storage costs.

[0085] In some alternative embodiments, a wavelet transform-based image compression algorithm may be used to process the load image to achieve compression of the load image.

[0086] Finally, the embodiment of the present disclosure also provides a method for restoring compressed image data, and an example is described as follows:

[0087] In some embodiments, the compressed image data is subjected to an inverse PCA transform, and the block matrix (i.e., the frequency domain matrix) is merged into a frequency domain image. Optionally, in the frequency domain image, the principal component positions based on the cumulative contribution rate are filled with the principal component data, and other positions are replaced with O to represent the data in these areas, so that the information content is 0.

[0088] The frequency domain image is decentralized, and the inverse FFT transform is applied to the decentralized frequency domain signal to obtain the restored load image.

[0089] Alternatively, the mathematical expression for the inverse FFT is as follows:

[0090] Among them, M and N are the width and height of the image respectively, F(u,v) is the value in the frequency domain, (x,y) is the coordinate in the spatial domain, and the restored image is obtained.

[0091] Optionally, the frequency domain signal is decentralized by subtracting a mean value of the frequency domain signal so that its mean value in the frequency domain is zero.

[0092] In summary, the method for processing power load data provided by the embodiments of the present disclosure can not only perform data visualization processing to help control personnel quickly understand the characteristic distribution of data, user power consumption behavior and changes in the operating status of the power system, but also compress and store data based on image compression and decompression algorithms to reduce the space required for data storage, thereby improving data management efficiency and reducing storage costs.

[0093] It should be noted that the processing method of the embodiment of the present disclosure can be executed by a single device, such as a computer or a server. The processing method of this embodiment can also be applied to a distributed scenario and completed by multiple devices cooperating with each other. In the case of such a distributed scenario, one of the multiple devices can only execute one or more steps in the processing method of the embodiment of the present disclosure, and the multiple devices will interact with each other to complete the processing method.

[0094] It should be noted that the above describes some embodiments of the present disclosure. Other embodiments are within the scope of the appended claims. In some cases, the actions or steps recorded in the claims can be performed in an order different from that in the above embodiments and still achieve the desired results. In addition, the processes depicted in the accompanying drawings do not necessarily require the specific order or continuous order shown to achieve the desired results. In some embodiments, multitasking and parallel processing are also possible or may be advantageous.

[0095] Based on the same inventive concept, corresponding to the processing method of any of the above embodiments, the present disclosure also provides a method for processing power load data.

[0096] Figure 3 A schematic diagram of a process flow of a method for processing power load data provided by an embodiment of the present disclosure is shown. Figure 3 As shown, the processing method includes:

[0097] S301: Acquire a first data set of power load; here, the first data set corresponds to the original data;

[0098] S303: preprocessing the first data set to obtain a second data set;

[0099] S305: determining a load image according to the second data set and a preset dimension rule; the preset dimension rule may refer to the above and will not be described in detail;

[0100] S307: Process the load image using an image compression algorithm to obtain compressed data. Here, the image compression algorithm may be a wavelet transform algorithm or a FFT combined with PCA algorithm.

[0101] Such a technical solution can not only perform data visualization to help control personnel quickly understand the characteristic distribution of data, user electricity consumption behavior and changes in the operating status of the power system, but also compress and store data based on image compression and decompression algorithms to reduce the space required for data storage, thereby improving data management efficiency and reducing storage costs.

[0102] In some embodiments, the preprocessing of the first data set to obtain the second data set in S303 specifically includes:

[0103] In response to determining that the first data set includes missing values, correcting the missing values; for example, correcting based on formula (1);

[0104] In response to determining that the first data set includes an outlier, correcting the outlier;

[0105] In response to determining that the first data set includes a repeated value, correcting the repeated value; here, the repeated value refers to different load data corresponding to the same sampling time; the correction method may be to take an average of multiple load data;

[0106] The corrected first data set is normalized to obtain a second data set. Here, the normalization formula is shown in formula (2).

[0107] In some embodiments, the first data set includes sampling time and load;

[0108] In response to determining that the first data set includes missing values, correcting the missing values ​​specifically includes:

[0109] Determining multiple reference times according to the sampling time of the missing value;

[0110] The missing values ​​are corrected according to a prediction formula and the loads corresponding to the multiple reference times; wherein the prediction formula is determined by performing a linear regression analysis based on historical load data.

[0111] In some embodiments, outliers in the first data set are determined based on an absolute median difference method.

[0112] In some embodiments, the first data set includes sampling time and load; the repetition value is that the same sampling time corresponds to multiple loads; and modifying the repetition value specifically includes:

[0113] An average value of the plurality of loads is calculated, and the average value is used as the load at the sampling time.

[0114] In some embodiments, determining the load image according to the second data set and the preset dimension rule in S305 specifically includes:

[0115] Convert normalized loads to HSV color values;

[0116] Convert the HSV color value to an RGB color value;

[0117] refer to Figure 1 and Figure 2, arranging the RGB color values ​​according to a preset dimensionality rule to obtain the load image; wherein the dimensionality rule includes the correspondence between sampling time and dimension.

[0118] In some embodiments, the process of processing the load image using an image compression algorithm to obtain compressed data in S307 specifically includes:

[0119] Perform fast Fourier transform on the grayscaled load image to obtain a frequency domain image;

[0120] Performing block processing on the frequency domain image to obtain a plurality of block data;

[0121] Based on the plurality of block data, construct a frequency domain matrix including a plurality of samples; wherein each block data corresponds to a sample data of the frequency domain matrix;

[0122] The frequency domain matrix is ​​processed by principal component analysis to obtain the compressed data.

[0123] In some embodiments, the processing the frequency domain matrix using principal component analysis to obtain the compressed data specifically includes:

[0124] Centralizing the frequency domain matrix to obtain a centralized matrix;

[0125] Based on the centering matrix, a sample covariance matrix is ​​determined, as shown in formula (6);

[0126] Determine a transformation matrix based on the sample covariance matrix and a preset cumulative contribution rate condition;

[0127] The compressed data is determined based on the frequency domain matrix and the transformation matrix.

[0128] In some embodiments, it also includes:

[0129] Performing principal component analysis inverse transformation on the compressed data and merging the data to obtain a restored frequency domain image;

[0130] Based on the restored frequency domain image, the restored load image is obtained through decentering and inverse fast Fourier transform, refer to formula (8).

[0131] The processing method of the above embodiment has the beneficial effects of the aforementioned processing method embodiments, which will not be described in detail here.

[0132] Based on the same inventive concept, corresponding to the processing method of any of the above embodiments, the present disclosure also provides an electronic device, including a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor implements the processing method described in any of the above embodiments when executing the program.

[0133] Figure 4 A more specific schematic diagram of the hardware structure of an electronic device provided in this embodiment is shown, and the device may include: a processor 1010, a memory 1020, an input / output interface 1030, a communication interface 1040, and a bus 1050. The processor 1010, the memory 1020, the input / output interface 1030, and the communication interface 1040 are connected to each other through the bus 1050 in the device.

[0134] The processor 1010 can be implemented by a general-purpose CPU (Central Processing Unit), a microprocessor, an application-specific integrated circuit (ASIC), or one or more integrated circuits, and is used to execute relevant programs to implement the technical solutions provided in the embodiments of this specification.

[0135] The memory 1020 may be implemented in the form of ROM (Read Only Memory), RAM (Random Access Memory), static storage device, dynamic storage device, etc. The memory 1020 may store an operating system and other application programs. When the technical solutions provided in the embodiments of this specification are implemented by software or firmware, the relevant program codes are stored in the memory 1020 and are called and executed by the processor 1010.

[0136] The input / output interface 1030 is used to connect the input / output module to realize information input and output. The input / output module can be configured in the device as a component (not shown in the figure), or it can be externally connected to the device to provide corresponding functions. The input device may include a keyboard, a mouse, a touch screen, a microphone, various sensors, etc., and the output device may include a display, a speaker, a vibrator, an indicator light, etc.

[0137] The communication interface 1040 is used to connect a communication module (not shown) to realize communication interaction between the device and other devices. The communication module can realize communication through a wired mode (such as USB, network cable, etc.) or a wireless mode (such as mobile network, WIFI, Bluetooth, etc.).

[0138] The bus 1050 includes a path that transmits information between the various components of the device (eg, the processor 1010, the memory 1020, the input / output interface 1030, and the communication interface 1040).

[0139] It should be noted that, although the above device only shows the processor 1010, the memory 1020, the input / output interface 1030, the communication interface 1040 and the bus 1050, in the specific implementation process, the device may also include other components necessary for normal operation. In addition, it can be understood by those skilled in the art that the above device may also only include the components necessary for implementing the embodiments of the present specification, and does not necessarily include all the components shown in the figure.

[0140] The electronic device of the above embodiment is used to implement the corresponding processing method in any of the above embodiments, and has the beneficial effects of the corresponding processing method embodiment, which will not be described in detail here.

[0141] Based on the same inventive concept, corresponding to the processing method of any of the above embodiments, the present disclosure also provides a non-transitory computer-readable storage medium, wherein the non-transitory computer-readable storage medium stores computer instructions, and the computer instructions are used to enable the computer to execute the processing method described in any of the above embodiments.

[0142] The computer-readable medium of this embodiment includes permanent and non-permanent, removable and non-removable media, and information storage can be achieved by any processing method or technology. Information can be computer-readable instructions, data structures, modules of programs, or other data. Examples of computer storage media include, but are not limited to, phase change memory (PRAM), static random access memory (SRAM), dynamic random access memory (DRAM), other types of random access memory (RAM), read-only memory (ROM), electrically erasable programmable read-only memory (EEPROM), flash memory or other memory technology, read-only compact disk read-only memory (CD-ROM), digital versatile disk (DVD) or other optical storage, magnetic cassettes, magnetic tape magnetic disk storage or other magnetic storage devices or any other non-transmission media that can be used to store information that can be accessed by a computing device.

[0143] The computer instructions stored in the storage medium of the above embodiments are used to enable the computer to execute the processing method described in any of the above embodiments, and have the beneficial effects of the corresponding processing method embodiments, which will not be repeated here.

[0144] Based on the same inventive concept, corresponding to the processing method described in any of the above embodiments, the present disclosure also provides a computer program product, which includes computer program instructions. In some embodiments, the computer program instructions can be executed by one or more processors of a computer so that the computer and / or the processor executes the processing method. Corresponding to the execution subject corresponding to each step in each embodiment of the processing method, the processor that executes the corresponding step may belong to the corresponding execution subject.

[0145] The computer program product of the above embodiment is used to enable the computer and / or the processor to execute the processing method described in any of the above embodiments, and has the beneficial effects of the corresponding processing method embodiments, which will not be repeated here.

[0146] Those skilled in the art should understand that the discussion of any of the above embodiments is merely illustrative and is not intended to imply that the scope of the present disclosure (including the claims) is limited to these examples. Based on the concept of the present disclosure, the technical features in the above embodiments or different embodiments may be combined, the steps may be implemented in any order, and there are many other variations of the different aspects of the embodiments of the present disclosure as described above, which are not provided in detail for the sake of simplicity.

[0147] In addition, to simplify the description and discussion, and in order not to make the embodiments of the present disclosure difficult to understand, the known power / ground connections to the integrated circuit (IC) chips and other components may or may not be shown in the provided figures. In addition, the device can be shown in the form of a block diagram to avoid making the embodiments of the present disclosure difficult to understand, and this also takes into account the fact that the details of the implementation of these block diagram devices are highly dependent on the platform on which the embodiments of the present disclosure will be implemented (that is, these details should be fully within the scope of understanding of those skilled in the art). Where specific details (e.g., circuits) are set forth to describe exemplary embodiments of the present disclosure, it is apparent to those skilled in the art that the embodiments of the present disclosure can be implemented without these specific details or with changes in these specific details. Therefore, these descriptions should be considered illustrative rather than restrictive.

[0148] Although the present disclosure has been described in conjunction with specific embodiments of the present disclosure, many replacements, modifications and variations of these embodiments will be apparent to those skilled in the art from the foregoing description. For example, other memory architectures (e.g., dynamic RAM (DRAM)) may use the embodiments discussed.

[0149] The embodiments of the present disclosure are intended to cover all such substitutions, modifications and variations that fall within the broad scope of the appended claims. Therefore, any omissions, modifications, equivalent substitutions, improvements, etc. made within the spirit and principles of the embodiments of the present disclosure should be included in the scope of protection of the present disclosure.

Claims

1. A method for processing power load data, characterized in that: include: Acquiring a first data set of power load; Preprocessing the first data set to obtain a second data set; determining a load image according to the second data set and a preset dimensional rule; The load image is processed using an image compression algorithm to obtain compressed data.

2. The processing method according to claim 1, characterized in that: in, The preprocessing of the first data set to obtain the second data set specifically includes: In response to determining that the first data set includes missing values, correcting the missing values; In response to determining that the first data set includes an outlier, correcting the outlier; In response to determining that the first data set includes duplicate values, correcting the duplicate values; The corrected first data set is normalized to obtain a second data set.

3. The processing method according to claim 2, characterized in that: The first data set includes sampling time and load; In response to determining that the first data set includes missing values, correcting the missing values ​​specifically includes: Determining multiple reference times according to the sampling time of the missing value; The missing values ​​are corrected according to a prediction formula and the loads corresponding to the multiple reference times; wherein the prediction formula is determined by performing a linear regression analysis based on historical load data.

4. The processing method according to claim 2, characterized in that: Determining outliers in the first data set based on the absolute median difference method; and / or The first data set includes sampling time and load; The repetition value corresponds to multiple loads at the same sampling time; Correcting the duplicate value includes: An average value of the plurality of loads is calculated, and the average value is used as the load at the sampling time.

5. The processing method according to claim 1, characterized in that: Determining the load image according to the second data set and the preset dimension rule specifically includes: Convert normalized loads to HSV color values; Convert the HSV color value to an RGB color value; The RGB color values ​​are arranged according to a preset dimension rule to obtain the load image; wherein the dimension rule includes a corresponding relationship between sampling time and dimension.

6. The processing method according to claim 1, characterized in that: The process of processing the load image using an image compression algorithm to obtain compressed data specifically includes: Perform fast Fourier transform on the grayscaled load image to obtain a frequency domain image; Performing block processing on the frequency domain image to obtain a plurality of block data; Based on the plurality of block data, construct a frequency domain matrix including a plurality of samples; wherein each block data corresponds to a sample data of the frequency domain matrix; The frequency domain matrix is ​​processed by principal component analysis to obtain the compressed data.

7. The processing method according to claim 6, characterized in that: The method of processing the frequency domain matrix by principal component analysis to obtain the compressed data specifically includes: Centralizing the frequency domain matrix to obtain a centralized matrix; Based on the centering matrix, determining a sample covariance matrix; Determine a transformation matrix based on the sample covariance matrix and a preset cumulative contribution rate condition; The compressed data is determined based on the frequency domain matrix and the transformation matrix.

8. The processing method according to claim 6, characterized in that: Also includes: Performing principal component analysis inverse transformation on the compressed data and merging the data to obtain a restored frequency domain image; Based on the restored frequency domain image, the restored load image is obtained through decentering and inverse fast Fourier transform.

9. An electronic device comprising a memory, a processor, and a computer program stored in the memory and executable by the processor, characterized in that: The processor implements the processing method according to any one of claims 1 to 8 when executing the computer program.

10. A non-transitory computer-readable storage medium, characterized in that: The non-transitory computer-readable storage medium stores computer instructions, and the computer instructions are used to enable a computer to execute the processing method according to any one of claims 1 to 8.

Images