Substation secondary equipment data synchronization method based on data sampling
By using data sampling and seasonal decomposition algorithms in the substation, the trend and periodic data of secondary equipment of the substation are synchronized, and the problem of excessive load on the control center of the substation data synchronization is solved, and the system performance and data comprehensiveness and reliability are improved.
Patent Information
- Application Number
- CN202411846548.8
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2024-12-16
- Publication Date
- 2025-04-15
AI Technical Summary
When data synchronization is achieved in the substation, huge data volume and complex data processing and analysis bring huge loads to the control center, resulting in a degradation of system performance.
The data synchronization method of secondary equipment of substations based on data sampling is adopted. By acquiring historical time series data, data preprocessing, continuous sampling and interval sampling are performed, the data is decomposed into trendy and periodic data using a seasonal decomposition algorithm, and only these data are synchronized to the control center.
It reduces the load of the control center, avoids processing and analysis of complex data, improves system performance, and ensures the comprehensiveness and reliability of data from a global and local perspective through continuous and interval sampling.
Smart Images
Figure CN119834451A_ABST
Abstract
Description
Technical Field
[0001] The present invention relates to the technical field of power data processing, and in particular to a data synchronization method for secondary equipment in a substation based on data sampling. Background Art
[0002] Substations are the most basic, critical and large-scale important links and components in power supply. Accelerating the transformation of substations to digitalization and intelligence is a necessary way to realize the political, economic and social responsibilities of the national energy strategy, and is the only way to build a solid energy Internet service.
[0003] In recent years, with the development of information technologies such as the Internet of Things, big data, and artificial intelligence, it has become a common understanding among enterprises and society to realize digital transformation of enterprises and improve their operational management capabilities by relying on the digital economy. In the future, the power industry will realize "safe equipment operation", "efficient professional management", and "high-quality operation and maintenance" through digitalization, aiming to ultimately provide users with "reliable power supply services", which is also the goal of future digital intelligent operation and maintenance of substations. When building digital power, data synchronization between substations is an important link and component.
[0004] At present, when realizing data synchronization in substations, all data are usually synchronized to the control center through clock synchronization. However, in reality, the amount of data in substations is huge. When the data inside each substation is synchronized to the control center, it will bring a huge load to the control center. Moreover, after all the data are synchronized to the control center, the control center needs to further analyze, store and analyze the characteristics of the complex data, which further increases the load of the control center. Summary of the invention
[0005] In view of this, in order to address the above shortcomings, it is necessary to propose a substation secondary equipment data synchronization method based on data sampling to reduce the load of the control center.
[0006] In a first aspect, the present invention provides a method for synchronizing substation secondary equipment data based on data sampling, comprising:
[0007] Obtain the historical time series data collected by the target monitoring point of the substation secondary equipment, as well as the historical time series data of other monitoring points related to the current target monitoring point, and together form a historical operation time series data set;
[0008] Performing data preprocessing on the historical operation time series data set to obtain preprocessed operation time series data;
[0009] The pre-processing operation time series data is respectively subjected to continuous sampling and interval sampling, and the continuous sampling data obtained by continuous sampling and the interval sampling data obtained by interval sampling constitute sampling data;
[0010] Decomposing the sampled data into trend data and periodic data based on a seasonal decomposition algorithm;
[0011] The trend data and periodic data are synchronized to a control center.
[0012] Preferably, the historical operation time series data set is expressed as:
[0013] X′ t ={x t-T+1 ,…,x t-1 ,x t |x i ∈R N}
[0014] Where T is the length of the historical running time series data set, N is the number of time series, and for any x i ={x i1 ,x i2 ,…,x iN}, {x i1 ,x i2 ,…,x iN-1} is the historical time series data of the relevant monitoring points, x iN It is the historical time series data of the target monitoring point.
[0015] Preferably, the data preprocessing of the historical operation time series data set includes:
[0016] The missing values of the historical operation time series data set are filled, and the data after the missing values are filled are subjected to Box-Cox transformation.
[0017] Preferably, the filling of missing values in the historical operation time series data set includes:
[0018] For the historical operation time series data set X′={x (1) ,x (2) ,…,x (N)} for each historical time series data x (j) , perform the following operations:
[0019] If the historical time series data x (j) If there are no missing values, no processing is required;
[0020] If the historical time series data x (j) If there are missing values at time t, then intercept where [a,b] contains missing time points t; and, in time series data The cubic function y(t) = a + bt + ct is fitted on 2 +dt 3 , and bring the missing time point into the fitted cubic function to obtain the filling data of the missing time point t
[0021] Preferably, the Box-Cox transformation of the data after missing value filling includes:
[0022] For each historical time series data x after missing values are filled (j) , transformed by the following formula:
[0023]
[0024] Among them, λ is a variable parameter that determines the transformation form.
[0025] Preferably, the continuous sampling of the preprocessing operation time series data comprises:
[0026] Convert each historical time series data with a time length of T into several non-overlapping subsequences with a length of C;
[0027] For historical time series data, a two-dimensional matrix with a shape of H×W is used to downsample the inverted subsequence to obtain a two-dimensional matrix The j-th column element is represented by
[0028] Preferably, the interval sampling of the preprocessing operation time series data includes:
[0029] Sampling is performed at a fixed time interval C, and the matrix obtained by sampling is The j-th column element of in, Indicates rounding down.
[0030] Preferably, the decomposing the sampled data into trend data and periodic data based on a seasonal decomposition algorithm comprises:
[0031] The sample data is decomposed as follows:
[0032]
[0033] Among them, X(t) represents the sampling data, g(t) is the trend data, s(t) is the periodic data, ε t represents the noise component, A represents the maximum asymptotic value, the maximum value of g(t) is infinitely close to A, k and m represent the slope and bias of the linear part respectively, P is the period, an and b n is the fitting parameter.
[0034] In a second aspect, the present invention provides a computer-readable storage medium having a computer program stored thereon, which, when executed in a computer, causes the computer to execute any of the methods described in the first aspect.
[0035] In a third aspect, the present invention provides a computing device, including a memory and a processor, wherein executable code is stored in the memory, and when the processor executes the executable code, any method described in the first aspect is implemented.
[0036] It can be seen from the above technical scheme that in the data synchronization method for substation secondary equipment based on data sampling provided by the embodiment of the present invention, firstly, the historical time series data collected by the target monitoring point of the substation secondary equipment and the historical time series data of other monitoring points related to the current target monitoring point are obtained to form a historical operation time series data set; then the historical operation time series data set is preprocessed to obtain preprocessed operation time series data; further, the preprocessed operation time series data are respectively sampled continuously and sampled at intervals to obtain sampling data; then, the sampling data is decomposed based on the seasonal decomposition algorithm to obtain trend data and periodic data, and then the trend data and periodic data are synchronized to the control center. It can be seen that this scheme considers synchronizing the substation secondary equipment data to the control center after sampling, not synchronizing all the data to the control center, which greatly reduces the load of the control center. Moreover, the sampling data is decomposed into trend data and periodic data based on the seasonal decomposition algorithm, so that the control center can store them separately, and the corresponding data can be used when performing predictive analysis, avoiding the processing and analysis of complex data, and further reducing the load of the control center. In addition, this scheme samples the data using continuous sampling and interval sampling respectively. Continuous sampling can capture the local time pattern of the time series, while interval sampling can capture the global time pattern of the time series. Therefore, this scheme can not only reduce the amount of data that needs to be synchronized, but also ensure the comprehensiveness and reliability of the data both globally and locally. BRIEF DESCRIPTION OF THE DRAWINGS
[0037] Figure 1 A flow chart of a substation secondary equipment data synchronization method based on data sampling provided by an embodiment of the present invention. DETAILED DESCRIPTION
[0038] In order to more clearly illustrate the technical solutions of the embodiments of the present invention, the following briefly introduces the drawings required for use in the embodiments. Obviously, the drawings described below are some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without creative work.
[0039] like Figure 1 As shown, the present invention provides a substation secondary equipment data synchronization method based on data sampling, and the method may include the following steps:
[0040] Step 101: Acquire the historical time series data collected by the target monitoring point of the substation secondary equipment, and the historical time series data of other monitoring points related to the current target monitoring point, and together form a historical operation time series data set;
[0041] Step 102: preprocess the historical operation time series data set to obtain preprocessed operation time series data;
[0042] Step 103: performing continuous sampling and interval sampling on the preprocessing operation time series data respectively, and forming sampling data from the continuous sampling data obtained by the continuous sampling and the interval sampling data obtained by the interval sampling;
[0043] Step 104: Decomposing the sampled data into trend data and periodic data based on a seasonal decomposition algorithm;
[0044] Step 105: Synchronize the trend data and periodic data to the control center.
[0045] In this embodiment, it is considered to synchronize the substation secondary equipment data to the control center after sampling, instead of synchronizing all the data to the control center, which greatly reduces the load of the control center. Moreover, based on the seasonal decomposition algorithm, the sampled data is decomposed into trend data and periodic data, so that the control center can store them separately, and use the corresponding data when performing forecast analysis, avoiding the processing and analysis of complex data, further reducing the load of the control center. In addition, this scheme samples the data by continuous sampling and interval sampling respectively. Continuous sampling can capture the local time pattern of the time series, while interval sampling can capture the global time pattern of the time series. Therefore, this scheme can not only reduce the amount of data that needs to be synchronized, but also ensure the comprehensiveness and reliability of the synchronized data from a global and local perspective.
[0046] For step 101, the historical time series data collected by the target monitoring point of the substation secondary equipment and the historical time series data of other monitoring points related to the current target monitoring point are obtained, and together they form a historical operation time series data set;
[0047] In this step, consider acquiring the historical time series data of the target monitoring point of the power secondary equipment and the historical time series data of other influencing factor monitoring points that may be related to the target monitoring point, and combine the two types of data to form a historical operation time series data set for synchronization to the control center. The composed historical operation time series data set can be expressed as: X′ t ={x t-T+1 ,…,x t-1 ,x t |x i ∈R N}
[0048] Where T is the length of the historical running time series data set, N is the number of time series, and for any x i ={x i1 ,x i2 ,…,x iN}, {x i1 ,x i2 ,…,x iN-1} is the historical time series data of the relevant monitoring points, x iN It is the historical time series data of the target monitoring point.
[0049] In addition, when obtaining historical time series data, the scope of data collection should be clarified, and the requirements such as the data content and data format to be collected should be determined. Based on the collected data content and format, different data proofreading methods, as well as data proofreading precautions or key points should be given to ensure the integrity and standardization of the data.
[0050] For step 102, data preprocessing is performed on the historical operation time series data set to obtain preprocessed operation time series data;
[0051] In this step, in order to improve the quality of the data, consider filling the missing values of the historical operation time series data set, and perform Box-Cox transformation on the data after missing value filling. Specifically, when filling the missing values of the historical operation time series data set, it can be achieved in the following ways:
[0052] For the historical operation time series data set X′={x (1) ,x (2) ,…,x (N) Each historical time series data x in (j) , j=1,2,…,N, the following operations are performed:
[0053] If the historical time series data x (j) If there are no missing values, no processing is required;
[0054] If the historical time series data x(j) If there are missing values at time t, then intercept where [a,b] contains missing time points t; and, in time series data The cubic function y(t) = a + bt + ct is fitted on 2 +dt 3 , where a, b, c, d are the coefficients of the cubic function to be fitted. Then the missing time point is substituted into the fitted cubic function to obtain the filling data of the missing time point t In this way, by filling in missing values, the integrity and accuracy of the data can be guaranteed to a great extent.
[0055] When performing Box-Cox transformation on the data after missing value filling, it can be achieved in the following ways:
[0056] For each historical time series data x after missing values are filled (j) , transformed by the following formula:
[0057]
[0058] Here, λ is a variable parameter that determines the transformation form. For example, when λ is 0, the transformation is a logarithmic transformation, and when λ is not 0, the transformation is an exponential transformation.
[0059] In this embodiment, for each x (j) After the Box-Cox transformation, the final preprocessed running time series data set can be obtained. By performing the Box-Cox transformation on the data, the unobservable errors can be reduced to a certain extent, and the problem of data variance homogeneity can be solved, which helps to improve the quality of the data.
[0060] For step 103, continuous sampling and interval sampling are performed on the preprocessing operation time series data, and the continuous sampling data obtained by continuous sampling and the interval sampling data obtained by interval sampling constitute the sampling data;
[0061] In this step, consider performing continuous sampling and interval sampling on the preprocessing running time series data respectively, so as to capture the local time pattern of the time series through continuous sampling and the global time pattern of the time series through interval sampling. Specifically, when performing continuous sampling on the preprocessing running time series data, each historical time series data with a time length of T can be first converted into several non-overlapping subsequences with a length of C; then for the historical time series data, a two-dimensional matrix with a shape of H×W is used to downsample to the subsequence, and a two-dimensional matrix is obtained The j-th column element is represented by H is the time dimension and W is the channel dimension.
[0062] When performing interval sampling on the preprocessing running time series data, sampling can be performed at a fixed time interval C, and the matrix obtained by sampling is The j-th column element of in, Indicates rounding down.
[0063] Further, the continuous sampling data obtained by continuous sampling and the interval sampling data obtained by interval sampling are combined to obtain sampling data.
[0064] For step 104, the sampled data is decomposed into trend data and periodic data based on a seasonal decomposition algorithm;
[0065] Considering that the data may contain different components, and the control center often performs different types of analysis when performing data analysis, such as the prediction of future equipment operation status, the periodic change law of equipment data, etc. Therefore, this step considers decomposing the sampled data into trend data and periodic data. Specifically, the sampled data can be decomposed as follows:
[0066]
[0067] Among them, X(t) represents the sampling data, g(t) is the trend data, s(t) is the periodic data, ε t represents the noise component, A represents the maximum asymptotic value, the maximum value of g(t) is infinitely close to A, k and m represent the slope and bias of the linear part respectively, P is the period, a n and b n is the fitting parameter.
[0068] In this embodiment, P is the period of the data, and N is determined according to the selection of the period P. For example, for a year as the period, the value of P is 365, and the value of N is 10; and for a week as the period, the value of P is 7, and the value of N is 3.
[0069] For step 105 , the trend data and the periodic data are synchronized to the control center.
[0070] In this embodiment, the sampled data is decomposed into trend data and periodic data based on the seasonal decomposition algorithm, so that the control center can store them separately and use the corresponding data when performing predictive analysis, avoiding the processing and analysis of complex data and further reducing the load on the control center.
[0071] The data synchronization method for substation secondary equipment based on data sampling provided by the present invention has at least the following beneficial effects:
[0072] (1) The substation secondary equipment data is sampled and synchronized to the control center, but not all the data is synchronized to the control center, which greatly reduces the load of the control center.
[0073] (2) Based on the seasonal decomposition algorithm, the sampled data is decomposed into trend data and periodic data, so that the control center can store them separately and use the corresponding data when conducting forecast analysis, avoiding the processing and analysis of complex data and further reducing the load of the control center.
[0074] (3) The data were sampled using continuous sampling and interval sampling respectively. Continuous sampling can capture the local temporal pattern of the time series, while interval sampling can capture the global temporal pattern of the time series. This not only reduces the amount of data that needs to be synchronized, but also ensures the comprehensiveness and reliability of the data both globally and locally.
[0075] (4) Data preprocessing includes missing value filling and Box-Cox transformation. Missing value filling can avoid missing collected data, thus making the data more complete. Box-Cox transformation can significantly improve the normality, symmetry and equality of variance of the data, which is very suitable for time series data collected by power equipment sensors.
[0076] The present specification also provides a computer-readable storage medium on which a computer program is stored. When the computer program is executed in a computer, the computer is caused to execute a method in any one of the embodiments in the specification.
[0077] The present specification also provides a computing device, including a memory and a processor, wherein executable codes are stored in the memory, and when the processor executes the executable codes, a method in any embodiment of the present specification is implemented.
[0078] Since the device embodiments provided by the present invention are based on the same inventive concept as the method embodiments of this specification, the specific contents can be found in the description of the method embodiments of this specification and will not be repeated here.
[0079] The modules or units in the device of the embodiment of the present invention can be combined, divided and deleted according to actual needs. The above disclosure is only the preferred embodiment of the present invention, and of course it cannot be used to limit the scope of the rights of the present invention. Those skilled in the art can understand that all or part of the processes of the above embodiment are implemented, and the equivalent changes made according to the claims of the present invention still fall within the scope of the invention.
Claims
1. A data synchronization method for substation secondary equipment based on data sampling, characterized in that: include: Obtain the historical time series data collected by the target monitoring point of the substation secondary equipment, as well as the historical time series data of other monitoring points related to the current target monitoring point, and together form a historical operation time series data set; Performing data preprocessing on the historical operation time series data set to obtain preprocessed operation time series data; The pre-processing operation time series data is respectively subjected to continuous sampling and interval sampling, and the continuous sampling data obtained by continuous sampling and the interval sampling data obtained by interval sampling constitute sampling data; Decomposing the sampled data into trend data and periodic data based on a seasonal decomposition algorithm; The trend data and periodic data are synchronized to a control center.
2. The data synchronization method for substation secondary equipment based on data sampling according to claim 1 is characterized in that: The historical operation time series data set is expressed as: X t ′={x t-T+1 ,…,x t-1 ,x t x i ∈R N } Where T is the length of the historical running time series data set, N is the number of time series, and for any x i ={x i1 ,x i2 ,…,x iN }, {x i1 ,x i2 ,…,x iN-1 } is the historical time series data of the relevant monitoring points, x iN It is the historical time series data of the target monitoring point.
3. The data synchronization method for substation secondary equipment based on data sampling according to claim 1 is characterized in that: The data preprocessing of the historical operation time series data set includes: The missing values of the historical operation time series data set are filled, and the data after the missing values are filled are subjected to Box-Cox transformation.
4. The data synchronization method for substation secondary equipment based on data sampling according to claim 3 is characterized in that: The filling of missing values for the historical operation time series data set includes: For the historical operation time series data set X′={x (1) ,x (2) ,…,x (N) Each historical time series data x in (j) , perform the following operations: If the historical time series data x (j) If there are no missing values, no processing is required; If the historical time series data x (j) If there are missing values at time t, then intercept where [a,b] contains missing time points t; and, in time series data The cubic function y(t) = a + bt + ct is fitted on 2 +dt 3 , and bring the missing time point into the fitted cubic function to obtain the filling data of the missing time point t 5. The data synchronization method for substation secondary equipment based on data sampling according to claim 4 is characterized in that: The Box-Cox transformation of the data after missing value filling includes: For each historical time series data x after missing values are filled (j) , transformed by the following formula: Among them, λ is a variable parameter that determines the transformation form.
6. The method for data synchronization of substation secondary equipment based on data sampling according to claim 2 is characterized in that: The continuously sampling the preprocessing operation time series data comprises: Convert each historical time series data with a time length of T into several non-overlapping subsequences with a length of C; For historical time series data, a two-dimensional matrix with a shape of H×W is used to downsample to a subsequence to obtain a two-dimensional matrix The j-th column element is represented by 7. The method for data synchronization of substation secondary equipment based on data sampling according to claim 2, characterized in that: The step of performing interval sampling on the preprocessing operation time series data comprises: Sampling is performed at a fixed time interval C, and the matrix obtained by sampling is The j-th column element of in, Indicates rounding down.
8. The method for data synchronization of substation secondary equipment based on data sampling according to claim 1, characterized in that: The seasonal decomposition algorithm is used to decompose the sampled data into trend data and periodic data, including: The sample data is decomposed as follows: Among them, X(t) represents the sampling data, g(t) is the trend data, s(t) is the periodic data, ε t represents the noise component, A represents the maximum asymptotic value, the maximum value of g(t) is infinitely close to A, k and m represent the slope and bias of the linear part respectively, P is the period, a n and b n is the fitting parameter.
9. A computer-readable storage medium having a computer program stored thereon, which, when executed in a computer, causes the computer to execute the method according to any one of claims 1 to 8.
10. A computing device, comprising a memory and a processor, wherein the memory stores executable codes, and when the processor executes the executable codes, the method according to any one of claims 1 to 8 is implemented.