Device and method for analyzing power grid information

Abstract

The invention discloses an analysis device and method for power grid information, and the method comprises the steps: obtaining the prominent credible amount of each information point according to the fluctuation condition of the prominent amount of the change of a my information point in a group where each information point is located, thereby achieving the separate extraction of an outlier and a reasonable value in a power grid information queue of a power grid; and the outliers are directly stored, so that real-time feeding and discharging of the outliers during later power grid information storage are facilitated, the accuracy and correctness of power grid information storage are improved, reasonable values are stored in a reduced manner, and the storage low-efficiency degree of the power grid information is reduced.

Description

Technical Field

[0001] The present invention belongs to the technical field of power grid information analysis, and in particular relates to a device and method for analyzing power grid information. Background Art

[0002] Informatization is an important content and means of the intelligent and digital construction of power grids, and occupies an important position in the construction of smart grids. The construction of smart grids provides a strong impetus for the development of power grid informatization. Strengthening the informatization transformation of power grids is an important task for power grid companies to build new power systems. It will promote power grid companies to accelerate the construction of information acquisition, perception, processing, and application, and build a smart Internet of Things system that connects users and equipment in all links of society, improve holographic perception and ubiquitous interconnection capabilities, and realize dynamic acquisition, real-time perception, and online monitoring of power grids, equipment, and customer status, thereby promoting the investment of power grid companies in the field of power informatization.

[0003] Therefore, during the period of power grid informatization, it is often as mentioned in the prior art solution with patent publication number "CN117277566B": the power grid information obtained in real time needs to be stored, and in actual applications, the storage efficiency needs to be improved and the power grid information needs to be distinguished into outliers and reasonable values ​​by grouping methods (reasonable values ​​are the power grid information remaining after outliers are removed from the power grid information). In addition, the reduced storage of power grid information is a key link in the execution of power grid information storage. When the K-means method is used to process the power grid information, when executing the outlier and reasonable value grouping and reduced storage analysis rules of the power grid information, the outlier attributes are often not prominent in the group due to the huge amount of information.

[0004] Therefore, when using the K-means method to process power grid information, outliers are often merged into the group during the group merging process, making it difficult to accurately extract the outliers of the power grid information. Finally, it is difficult to achieve the accuracy and correctness of the power grid information storage, increasing the inefficiency of power grid information storage. Summary of the invention

[0005] In order to solve the defects in the prior art, the present invention proposes an analysis device and method for power grid information, which obtains the outstanding reliability of each information point according to the fluctuation of the change outstanding amount of each information point in the group to which it belongs, thereby achieving the separate extraction of outliers and reasonable values ​​in the power grid information queue of the power grid; directly storing the outliers is convenient for the real-time input and output of the outliers when the power grid information is stored later, thereby improving the accuracy and correctness of the power grid information storage, and performing reduced storage of reasonable values, thereby reducing the storage efficiency of power grid information.

[0006] The present invention uses the following technical solutions.

[0007] A method for analyzing power grid information, comprising: Step 1, obtain the power grid information queue of the power grid; Step 2, construct a use case interval according to the power grid information queue of the power grid, group all information points in the use case interval, obtain multiple groups, and obtain the change prominence of each information point in each group; Step 3, obtain the outstanding credibility of each information point in each group; Step 4, use the prominent credibility of the information point to distinguish the outliers and reasonable values ​​in the power grid information queue of the power grid, so as to achieve the analyzed storage of the power grid information.

[0008] Preferably, Step 1 specifically includes: A sensor group connected to the analysis terminal is used to obtain the grid information of the power grid, which includes the voltage value, current value or power value of the power grid; each sensor in the sensor group takes a sampling point every 2s, and the grid information of the power grid is sampled each time and transmitted to the analysis terminal in real time, with a total sampling time of 12h; the analysis terminal arranges the grid information of the power grid at each sampling point obtained in the order of the sampling points to form a grid information queue of the power grid.

[0009] Preferably, Step 2 specifically includes: setting an indicator in advance , construct a use case interval according to the power grid information queue of the power grid, take the sampling time point as the horizontal coordinate value of the use case interval, take the power grid information of the power grid as the vertical coordinate value of the use case interval, and each information point in the use case interval represents the power grid information of the power grid at each sampling time point.

[0010] Preferably, The value is twenty.

[0011] Preferably, Step 2 specifically includes: performing agglomeration grouping on all information points in the use case interval to obtain multiple groups; In the group information point, Any information point in a group and The modulus of the quantity obtained by subtracting the information points is taken as the The interval between each information point is one, The total interval of each information point is arranged in order from low to high, and the queue formed is taken as the first The interval between each information point is one queue, and the first The interval between the first and the last information point in the queue The corresponding information point is regarded as The information point of the information point is The information point queue formed by all information points of information points is taken as the first The information point queue near the information point; then the In the group The calculation equation of the change prominence of each information point is: ; In the equation, Representative In the group The change prominence of each information point; Representative In the group In the queue of nearby information points of information points The numerical difference of each information point; Representative In the group Information points; Representative In the group In the queue of nearby information points of information points Information points; Representative In the group In the queue of nearby information points of information points Information points and The modulus of the quantity obtained by subtracting information points; It is a pre-set indicator.

[0012] Preferably, Step 3 specifically includes: setting three indicators in advance , for the first Group the first The total number of all information points in the group and The multiplication is defined as , put the The total number of all information points in the group and The multiplication is defined as , put the first The total number of all information points in the group and The multiplication result is defined as ; The first time in Use any of the groups The information point is taken as the initial center for the The groups are grouped by k-means method, and the groups and defined as The subgroup of the group, the second time in Use any of the following in a group The information point is taken as the initial center for the The groups are grouped by k-means method, and the The group is defined as The third time in the subgroup of the group Randomly select within each group The information point is taken as the initial center for the The groups are grouped by k-means method, and the The group is defined as The subgroup of the group, the 4th Randomly select within each group The information point is taken as the initial center for the The groups are grouped by k-means method, and the The group is defined as The sub-groups of the group can be obtained in the same way until the last time in The number of initial centers selected in each group is not less than When , the grouping is terminated; and then the first All subgroups of a group.

[0013] Preferably, The value is , The value is , The value is .

[0014] Preferably, Step 3 specifically further includes: for the In the group information point, The subgroup where the information point is located is defined as The target subgroups of the information points are grouped and the linear discriminant LDA method is used to Perform analysis on any subgroup of information points to obtain a linear combination of the subgroup, and take the average of the values ​​obtained after performing orthogonal transformation on all information points in the subgroup and the linear combination of the subgroup as the arrangement balance attribute value of the subgroup; then In the group The calculation equation of the outstanding credibility of each information point is: ; In the equation, Representative In the group The outstanding credibility of each information point; Representative In the group The total number of all purpose subgroups of information points; Representative In the group Information point The average of the change prominence of all information points in the subgroups of the target; Representative In the group Information point The variance of the change prominence of all information points in each subgroup; Representative In the group Information point The total number of all information points in the sub-groups of the target; Representative In the group Information point The distribution balance attribute value of each sub-group; Represents the use of Z-score method to Implement standardization.

[0015] Preferably, Step 4 specifically includes: setting a critical value in advance For any group within the use case interval, if the outstanding credibility of any information point in the group is not lower than the critical value, the information point is defined as an outlier point, and all outlier points of all groups are regarded as outliers of the power grid information queue of the power grid, and the power grid information outside the outliers in the power grid information queue of the power grid is regarded as reasonable values. Then, the LZ4 method is used to perform information reduction on the reasonable values ​​of the power grid information queue of the power grid and store them in the hard disk of the analysis terminal. The outliers in the power grid information queue of the power grid are not reduced and are directly stored in the hard disk of the analysis terminal.

[0016] A device for analyzing power grid information, comprising: A queue module, which is used to obtain a grid information queue of a grid; A grouping module, which is used to construct a use case interval according to the power grid information queue of the power grid, perform grouping on all information points in the use case interval, obtain multiple groups, and obtain the change prominence of each information point in each group; A highlight module, which is used to obtain the highlight credibility of each information point in each group; The analysis module is used to use the prominent credibility of the information point to distinguish the outliers and reasonable values ​​in the power grid information queue of the power grid, so as to achieve the analyzed storage of the power grid information.

[0017] The beneficial effects of the present invention are that, compared with the prior art, the technical effects of the present invention include: According to the fluctuation of the change of the outstanding value of each information point in the group, the outstanding reliability of each information point is obtained, so as to achieve the separate extraction of outliers and reasonable values ​​in the power grid information queue of the power grid; the outliers are directly stored, which is convenient for the real-time input and output of the outliers in the subsequent power grid information storage, thereby improving the accuracy and correctness of the power grid information storage, and the reasonable values ​​are stored in a reduced amount, thereby reducing the storage efficiency of the power grid information. BRIEF DESCRIPTION OF THE DRAWINGS

[0018] Figure 1 is a flow chart of the method for analyzing power grid information described in the present invention; Figure 2 It is a partial structural schematic diagram of the analysis device for power grid information described in the present invention. DETAILED DESCRIPTION

[0019] In order to make the purpose, technical solution and advantages of the present invention clearer, the technical solution of the present invention will be clearly and completely expressed in combination with the drawings in the embodiments of the present invention. The embodiments expressed in this application are only part of the embodiments of the present invention, not all of them. According to the spirit of the present invention, other embodiments obtained by ordinary technicians in this field without creative work are all within the protection scope of the present invention.

[0020] like Figure 1 As shown, a method for analyzing power grid information according to the present invention is run on an analysis terminal (the analysis terminal may be an industrial computer, a computer or a server), and includes: Step 1, obtain the power grid information queue of the power grid; Under the analysis of power grid information, corresponding methods can be used to achieve the improvement of reduced storage and grouped storage of power grid information. The key is to effectively improve the accuracy and correctness of power grid information storage through the acquisition and analysis of power grid information.

[0021] In a preferred but non-limiting embodiment of the present invention, to achieve a method for analyzing power grid information provided in this application, it is necessary to initially obtain a power grid information queue of the power grid, wherein Step 1 specifically includes: A sensor group connected to the analysis terminal for communication is used (the sensor group includes a voltage sensor, a current sensor and an electric quantity sensor, and the voltage sensor, the current sensor and the electric quantity sensor are used to sample the voltage value, the current value and the electric quantity value of the power grid respectively and transmit them to the analysis terminal) to obtain the power grid information of the power grid, and the power grid information of the power grid includes the voltage value, the current value or the electric quantity value of the power grid; each sensor in the sensor group takes a sampling time point every 2s, samples the power grid information of the power grid each time and transmits it to the analysis terminal in real time, and the sampling time is 12h in total; the analysis terminal arranges the power grid information of the power grid at each sampling time point obtained according to the sequence of the sampling time points to form a power grid information queue of the power grid.

[0022] Therefore, the power grid information queue of the power grid is obtained through the above method.

[0023] Step 2, construct a use case interval according to the power grid information queue of the power grid, group all information points in the use case interval, obtain multiple groups, and obtain the change prominence of each information point in each group; Because the obtained power grid information of the power grid is closely related to the sampling time point, it also corresponds to the power-on and power-off operation periods of the power grid and all its power grid components. The obtained power grid information has a chain structure in the order of time points, and the information points in different time periods have different arrangement attributes in the use case interval of the two types of values. Based on this, the change attributes of the information points in a single time period are analyzed to obtain the change prominence of each information point.

[0024] In a preferred but non-limiting embodiment of the present invention, Step 2 specifically includes: setting an index in advance , construct a use case interval according to the power grid information queue of the power grid, take the sampling time point as the horizontal coordinate value of the use case interval, take the power grid information of the power grid as the vertical coordinate value of the use case interval, and each information point in the use case interval represents the power grid information of the power grid at each sampling time point.

[0025] In a preferred but non-limiting embodiment of the present invention, The value is twenty.

[0026] In a preferred but non-limiting embodiment of the present invention, Step 2 specifically includes: performing agglomeration grouping on all information points in the use case interval to obtain multiple groups; In the group information point, Any information point in a group and The modulus of the quantity obtained by subtracting the information points is taken as the The interval between each information point is one, The total interval of each information point is arranged in order from low to high, and the queue formed is taken as the first The interval between each information point is one queue, and the first The interval between the first and the last information point in the queue The corresponding information point is regarded as The information point of the information point is The information point queue formed by all information points of information points is taken as the first The information point queue near the information point; then the In the group The calculation equation of the change prominence of each information point is: ; In the equation, Representative In the group The change prominence of each information point; Representative In the group In the queue of nearby information points of information points The numerical difference of each information point; Representative In the group Information points; Representative In the group In the queue of nearby information points of information points Information points; Representative In the group In the queue of nearby information points of information points Information points and The modulus of the quantity obtained by subtracting information points; It is a pre-set indicator.

[0027] Therefore, the above method is used to obtain the change prominence of each information point in each group.

[0028] Step 3, obtain the outstanding credibility of each information point in each group; According to the change prominence of all information points within the obtained use case interval, it can be used to characterize the probability that the information point belongs to an outlier point. However, due to the huge amount of information of the power grid information obtained, when the K-means method is used to obtain the grouping, the real outlier points will form different grouping distribution attributes due to the ratio of reasonable and outlier points in the selected information point group. Therefore, according to the prominence attributes of information points under different use case specifications under the distribution attributes of different groups, the number of information points can be adjusted to obtain the prominence reliability of the information points.

[0029] In a preferred but non-limiting embodiment of the present invention, Step 3 specifically includes: setting three indicators in advance: , for the first Group the first The total number of all information points in the group and The multiplication result is defined as , put the first The total number of all information points in the group and The multiplication result is defined as , put the first The total number of all information points in the group and The multiplication result is defined as ; The first time in Use any of the following in a group The information point is taken as the initial center for the The groups are grouped by k-means method, and the groups and defined as The subgroup of the group, the second time in Use any of the following in a group The information point is taken as the initial center for the The groups are grouped by k-means method, and the The group is defined as The third time in the subgroup of the group Use any of the following in a group The information point is taken as the initial center for the The groups are grouped by k-means method, and the The group is defined as The fourth subgroup of the group is Use any of the following in a group The information point is taken as the initial center for the The groups are grouped by k-means method, and the The group is defined as The sub-groups of the group can be obtained in the same way until the last time in The number of initial centers selected in each group is not less than When , the grouping is terminated; and then the first All subgroups of a group.

[0030] In a preferred but non-limiting embodiment of the present invention, The value is , The value is , The value is .

[0031] In a preferred but non-limiting embodiment of the present invention, Step 3 specifically further includes: In the group information point, The subgroup where the information point is located is defined as The target subgroups of the information points are grouped and the linear discriminant LDA method is used to Perform analysis on any subgroup of information points to obtain a linear combination of the subgroup, and take the average of the values ​​obtained after performing orthogonal transformation on all information points in the subgroup and the linear combination of the subgroup as the arrangement balance attribute value of the subgroup; then In the group The calculation equation of the outstanding credibility of each information point is: ; In the equation, Representative In the group The outstanding credibility of each information point; Representative In the group The total number of all purpose subgroups of information points; Representative In the group Information point The average of the change prominence of all information points in the subgroups of the target; Representative In the group Information point The variance of the change prominence of all information points in each subgroup; Representative In the group Information point The total number of all information points in the sub-groups of the target; Representative In the group Information point The distribution balance attribute value of each sub-group; Represents the use of Z-score method to Implement standardization.

[0032] According to the fluctuation property of the prominence of information points under the current grouping specifications, as well as the distribution properties of multiple sub-groups, that is, the balance between the number of information points contained in multiple groups and their distribution, the distribution balance attribute value of the target sub-group can be obtained through linear discriminant LDA analysis, and the prominence reliability of the information points can be obtained accordingly.

[0033] Therefore, the outstanding credibility of each information point in each group is obtained through the above method.

[0034] Step 4, use the prominent credibility of the information point to distinguish the outliers and reasonable values ​​in the power grid information queue of the power grid, so as to achieve the analyzed storage of the power grid information.

[0035] In a preferred but non-limiting embodiment of the present invention, Step 4 specifically includes: setting a critical value in advance For any group within the use case interval, if the outstanding credibility of any information point in the group is not lower than the critical value, the information point is defined as an outlier point, and all outlier points of all groups are regarded as outliers of the power grid information queue of the power grid, and the power grid information outside the outliers in the power grid information queue of the power grid is regarded as reasonable values. Then, the LZ4 method is used to perform information reduction on the reasonable values ​​of the power grid information queue of the power grid and store them in the hard disk of the analysis terminal. The outliers in the power grid information queue of the power grid are not reduced and are directly stored in the hard disk of the analysis terminal.

[0036] Since reasonable values ​​have high redundancy, reasonable values ​​are stored in reduced quantities to reduce the inefficiency of grid information storage. Outliers play an important role in analyzing grid operation status. Direct storage of outliers is convenient for real-time input and output of outliers in subsequent grid information storage, thereby improving the accuracy and correctness of grid information storage.

[0037] like Figure 2 As shown, the present invention provides an analysis device for power grid information, comprising: A queue module, which is used to obtain a grid information queue of a grid; A grouping module, which is used to construct a use case interval according to the power grid information queue of the power grid, perform grouping on all information points in the use case interval, obtain multiple groups, and obtain the change prominence of each information point in each group; A highlight module, which is used to obtain the highlight credibility of each information point in each group; The analysis module is used to use the prominent credibility of the information point to distinguish the outliers and reasonable values ​​in the power grid information queue of the power grid, so as to achieve the analyzed storage of the power grid information.

[0038] The beneficial effects of the present invention are that, compared with the prior art, the technical effects of the present invention include: According to the fluctuation of the change of the outstanding value of each information point in the group, the outstanding reliability of each information point is obtained, so as to achieve the separate extraction of outliers and reasonable values ​​in the power grid information queue of the power grid; the outliers are directly stored, which is convenient for the real-time input and output of the outliers in the subsequent power grid information storage, thereby improving the accuracy and correctness of the power grid information storage, and the reasonable values ​​are stored in a reduced amount, thereby reducing the storage efficiency of the power grid information.

[0039] Finally, it should be noted that the above embodiments are only used to illustrate the technical solution of the present invention rather than to limit it. Although the present invention has been described in detail with reference to the above embodiments, ordinary technicians in the relevant field should understand that the specific implementation methods of the present invention can still be modified or replaced by equivalents, and any modifications or equivalent replacements that do not deviate from the spirit and scope of the present invention should be covered within the protection space of the claims of the present invention.

Claims

1. A method for analyzing power grid information, characterized in that: include: Step 1, obtain the power grid information queue of the power grid; Step 2, construct a use case interval according to the power grid information queue of the power grid, group all information points in the use case interval, obtain multiple groups, and obtain the change prominence of each information point in each group; Step 3, obtain the outstanding credibility of each information point in each group; Step 4, use the prominent credibility of the information point to distinguish the outliers and reasonable values ​​in the power grid information queue of the power grid, so as to achieve the analyzed storage of the power grid information.

2. The method for analyzing power grid information according to claim 1, characterized in that: Step 1 specifically includes: A sensor group connected to the analysis terminal is used to obtain the grid information of the power grid, which includes the voltage value, current value or power value of the power grid; each sensor in the sensor group takes a sampling point every 2s, and the grid information of the power grid is sampled each time and transmitted to the analysis terminal in real time, with a total sampling time of 12h; the analysis terminal arranges the grid information of the power grid at each sampling point obtained in the order of the sampling points to form a grid information queue of the power grid.

3. The method for analyzing power grid information according to claim 2, characterized in that: Step 2 specifically includes: setting an indicator in advance , construct a use case interval according to the power grid information queue of the power grid, take the sampling time point as the horizontal coordinate value of the use case interval, take the power grid information of the power grid as the vertical coordinate value of the use case interval, and each information point in the use case interval represents the power grid information of the power grid at each sampling time point.

4. The method for analyzing power grid information according to claim 3, characterized in that: The value is twenty.

5. The method for analyzing power grid information according to claim 4, characterized in that: Step 2 specifically includes: performing agglomeration grouping on all information points in the use case interval to obtain multiple groups; In the group information point, Any information point in a group and The modulus of the quantity obtained by subtracting the information points is taken as the The interval between each information point is one, The total interval of each information point is arranged in order from low to high, and the queue formed is taken as the first The interval between each information point is one queue, and the first The interval between the first and the last information point in the queue The corresponding information point is regarded as The information point of the information point is one. The information point queue formed by all information points of information points is taken as the first The information point queue near the information point; then the In the group The calculation equation of the change prominence of each information point is: ； In the equation, Representative In the group The change prominence of each information point; Representative In the group In the queue of nearby information points of information points The numerical difference of each information point; Representative In the group Information points; Representative In the group In the queue of nearby information points of information points Information points; Representative In the group In the queue of nearby information points of information points Information points and The modulus of the quantity obtained by subtracting information points; It is a pre-set indicator.

6. The method for analyzing power grid information according to claim 5, characterized in that: Step 3 specifically includes: setting 3 indicators in advance , for the first Group the first The total number of all information points in the group and The multiplication result is defined as , put the first The total number of all information points in the group and The multiplication result is defined as , put the first The total number of all information points in the group and The multiplication result is defined as ; The first time in Use any of the following in a group The information point is taken as the initial center for the The groups are grouped by k-means method, and the groups and defined as The subgroup of the group, the second time in Use any of the following in a group The information point is taken as the initial center for the The groups are grouped by k-means method, and the The group is defined as The third time in the subgroup of the group Use any of the following in a group The information point is taken as the initial center for the The groups are grouped by k-means method, and the The group is defined as The subgroup of the group, the 4th Use any of the following in a group The information point is taken as the initial center for the The groups are grouped by k-means method, and the The group is defined as The sub-groups of the group can be obtained in the same way until the last time in The number of initial centers selected in each group is not less than When , the grouping is terminated; and then the first All subgroups of a group.

7. The method for analyzing power grid information according to claim 6, characterized in that: The value is , The value is , The value is .

8. The method for analyzing power grid information according to claim 7, characterized in that: Step 3 specifically includes: In the group information point, The subgroup where the information point is located is defined as The target subgroups of the information points are grouped and the linear discriminant LDA method is used to Perform analysis on any subgroup of information points to obtain a linear combination of the subgroup, and take the average of the values ​​obtained after performing orthogonal transformation on all information points in the subgroup and the linear combination of the subgroup as the arrangement balance attribute value of the subgroup; then In the group The calculation equation of the outstanding credibility of each information point is: ； In the equation, Representative In the group The outstanding credibility of each information point; Representative In the group The total number of all purpose subgroups of information points; Representative In the group Information point The average of the change prominence of all information points in the subgroups of the target; Representative In the group Information point The variance of the change prominence of all information points in each subgroup; Representative In the group Information point The total number of all information points in the sub-groups of the target; Representative In the group Information point The distribution balance attribute value of each sub-group; Represents the use of Z-score method to Implement standardization.

9. The method for analyzing power grid information according to claim 8, characterized in that: Step 4 specifically includes: setting a critical value in advance For any group within the use case interval, if the outstanding credibility of any information point in the group is not lower than the critical value, the information point is defined as an outlier point, and all outlier points of all groups are regarded as outliers of the power grid information queue of the power grid, and the power grid information outside the outliers in the power grid information queue of the power grid is regarded as reasonable values. Then, the LZ4 method is used to perform information reduction on the reasonable values ​​of the power grid information queue of the power grid and store them in the hard disk of the analysis terminal. The outliers in the power grid information queue of the power grid are not reduced and are directly stored in the hard disk of the analysis terminal.

10. A device for analyzing power grid information, characterized in that: include: A queue module, which is used to obtain a grid information queue of a grid; A grouping module, which is used to construct a use case interval according to the power grid information queue of the power grid, perform grouping on all information points in the use case interval, obtain multiple groups, and obtain the change prominence of each information point in each group; A highlight module, which is used to obtain the highlight credibility of each information point in each group; The analysis module is used to use the prominent credibility of the information point to distinguish the outliers and reasonable values ​​in the power grid information queue of the power grid, so as to achieve the analyzed storage of the power grid information.

Images