A statistical analysis system based on intelligent substation operation data management and control

Abstract

The present invention relates to the field of data processing, and more specifically, to a statistical analysis system based on intelligent substation operation data management and control, comprising: collecting various parameter data of the substation regularly or on demand through a data acquisition module, in a data segmentation module, using a sliding window trajectory algorithm to perform fine segmentation on each parameter, screening out target segments with statistical analysis value, a data calculation module comprehensively evaluating the comprehensive abnormality degree and fluctuation coefficient of each target segment and assigning a preferred degree value to each target segment, a data compression and storage module using the preferred degree of each target segment as a screening basis, and selectively selecting target segments for compression and storage, thereby improving the accuracy of data compression, saving storage space, and being able to achieve effective statistics and analysis of the operation data of the intelligent substation.

Description

Technical Field

[0001] The present invention relates to the field of data processing, and in particular to a statistical analysis system based on intelligent substation operation data management and control. Background Art

[0002] Statistical analysis of smart substation operation data is of great significance for monitoring the operating status of smart substations. By real-time monitoring of key parameters such as voltage, current, temperature, humidity, vibration, and comprehensive statistical analysis of these parameter data, potential problems in substation operation can be discovered in a timely manner, which plays a key role in ensuring the safe and efficient operation of smart substations, as well as troubleshooting and maintenance.

[0003] In view of the massive amount of real-time monitoring data generated by smart substations, traditional storage and transmission methods face the challenges of high storage costs and low transmission efficiency. In response, the industry generally uses sliding window trajectory compression algorithms to compress these monitoring data in segments before storing them, in order to save storage space and improve data transmission efficiency.

[0004] However, the current sliding window trajectory compression algorithm has certain limitations. The algorithm only segments the data based on whether there are points in the monitoring data that exceed the error threshold, and does not fully consider the overall error accumulation effect of the entire segment of data. In other words, even if the errors of each point in a certain segment of data do not exceed the threshold, the overall error may be too large, resulting in these data segments with large overall errors being regarded as the same compressed object during the compression process, which may cause a certain degree of deviation in the overall data, resulting in distortion, affecting the accuracy of subsequent statistical analysis. Summary of the invention

[0005] In order to solve one or more of the above technical problems, the present invention proposes a statistical analysis system based on the operation data management of smart substations, which improves the accuracy of data compression and enables effective statistics and analysis based on accurate compression. The following technical solutions are specifically adopted: A statistical analysis system based on the operation data management of smart substations, comprising:

[0006] The data acquisition module is used to collect various parameter data of the smart substation regularly or multiple times as needed, wherein the parameter data covers multiple different types of parameter items, and each parameter item is composed of a series of continuous data points;

[0007] A data segmentation module obtains the error threshold of each parameter item, performs segmentation processing on each collected parameter item using a sliding window trajectory algorithm to obtain multiple segments of each parameter item, and determines a target segment of each parameter item based on the data points contained in each segment and the error threshold of the parameter item;

[0008] A data calculation module, including an abnormality degree calculation unit, a fluctuation coefficient calculation unit and a preference degree calculation unit;

[0009] The abnormality degree calculation unit determines the comprehensive abnormality degree of each target segment according to the actual value of each data point in each target segment and its predicted value obtained by the sliding window trajectory algorithm, as well as the error threshold;

[0010] The fluctuation coefficient calculation unit determines the fluctuation coefficient of each target segment based on the actual numerical difference between adjacent data points in each target segment and the error threshold;

[0011] The preference calculation unit determines the preference value of each target segment based on the fluctuation coefficient and the comprehensive abnormality degree of each target segment;

[0012] The data compression and storage module uses the preference level of each target segment to screen the target segment and compress and store it.

[0013] Furthermore, the fluctuation coefficient calculation unit determines the fluctuation coefficient of each target segment according to the following relationship:

[0014]

[0015] In the formula, is the jth parameter item The fluctuation coefficient of the target segment, For the The first parameter The number of data points contained in each target segment. is the sequence number of the data point, The value of Any integer in the range, For the The first parameter Target segment The actual value of the data point, For the The first parameter Target segment The actual value of the data point, The natural constant is the exponential function of the base, is the preset acceptable overall error ratio, For the The error threshold of the parameter item.

[0016] Furthermore, the abnormality degree calculation unit determines the comprehensive abnormality degree of each target segment according to the following relationship:

[0017]

[0018] In the formula, For the The first parameter The comprehensive difference of target segments, For the The first parameter The number of data points contained in each target segment. is the sequence number of the data point, The value of Any integer in the range, For the The first parameter Target segment The actual value of the data point, For the The first parameter Target segment The predicted value of the data point, is the preset acceptable overall error ratio, For the The error threshold of the parameter item.

[0019] Furthermore, the preference calculation unit determines the preference value of each target segment, including:

[0020] The inverse of the product of the comprehensive difference and the fluctuation coefficient of each target segment is used as the preference value of the target segment.

[0021] Furthermore, the data segmentation module determines the target segmentation of each parameter item, including:

[0022] The data points in each segment are compared with a preset error threshold. If the values ​​of the data points in the segment do not exceed the abnormal threshold, the segment is the target segment of the parameter item.

[0023] Furthermore, the data compression and storage module uses the preference level of each target segment to screen and compress and store the target segment, including:

[0024] Set a preference threshold for each target segment;

[0025] If the preference value of the target segment is greater than the preference threshold of the target segment, the target segment is a preferred data segment and is compressed and stored; otherwise, the target segment does not belong to the preferred data segment and is directly stored without compression.

[0026] Furthermore, the data segmentation module obtains the error threshold of each parameter item, including:

[0027] The upper quartile and lower quartile of all data points contained in each parameter item are obtained through the box plot algorithm, the difference between the upper quartile and the lower quartile is calculated, and half of the absolute value of the difference is used as the error threshold of the parameter item.

[0028] The present invention has the following effects:

[0029] The present invention designs a system containing multiple modules, collects monitoring data during the operation of the smart substation, and uses a sliding trajectory algorithm to divide multiple data points contained in each parameter into multiple segments. On the basis of the traditional sliding window trajectory algorithm that performs segmented compression based on the error between the data points and the fitting straight line, it adds dual considerations of the overall comprehensive difference degree and fluctuation coefficient of the data points contained in the segment. First, the comprehensive difference degree is calculated for each segment, and the comprehensive difference degree reveals the overall situation of the variation degree of all data points in the segment. Secondly, the actual fluctuation of the data points in the segment and the fitting straight line is examined, and the data characteristics in the segment are comprehensively evaluated, and then the compression decision of the segment is determined. In this way, the present invention can effectively reduce the deviation between the data points contained in the segment and the original trend fitting straight line of these data points during the data compression process, thereby avoiding the problem of data distortion, enhancing the accuracy of the compressed data, and thus improving the accuracy and reliability of subsequent statistical analysis. BRIEF DESCRIPTION OF THE DRAWINGS

[0030] By reading the following detailed description with reference to the accompanying drawings, the above and other objects, features and advantages of the exemplary embodiments of the present invention will become readily understood. In the accompanying drawings, several embodiments of the present invention are shown in an exemplary and non-limiting manner, and the same or corresponding reference numerals represent the same or corresponding parts, wherein:

[0031] Figure 1 It is a schematic diagram of the method flow of the present invention. DETAILED DESCRIPTION

[0032] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative work are within the scope of protection of the present invention.

[0033] The specific embodiments of the present invention are described in detail below with reference to the accompanying drawings.

[0034] Reference Figure 1, a statistical analysis system based on intelligent substation operation data management and control, including a data acquisition module S100, a data segmentation module S200, a data calculation module S300, and a data compression and storage module S400;

[0035] The data collection module S100 is used to collect various parameter data of the smart substation regularly or multiple times as needed, wherein the parameter data covers multiple different types of parameter items, and each parameter item is composed of a series of continuous data points.

[0036] Collecting the operation data of smart substations is an important part of the operation and maintenance control of smart substations. It helps to achieve the goals of real-time monitoring of equipment status, fault diagnosis and prediction, operation optimization, energy saving and consumption reduction. The parameter data collected in this embodiment include: equipment current, voltage, equipment casing and key component temperature, transformer oil temperature, oil level and other parameter items. Through these monitoring data, the operation status of various equipment in the smart substation can be accurately evaluated. For example, through current and voltage data, it can be judged whether the equipment is operating within the normal load range, whether there are potential faults such as overload and short circuit. By monitoring the oil temperature and oil level of equipment such as transformers, it is possible to detect early whether the equipment has oil leakage, cooling system failure or overheating and other safety hazards. Monitoring equipment temperature data can timely detect whether the equipment is overheating, help judge whether the equipment heat dissipation performance is normal and other potential problems.

[0037] In the specific implementation, data acquisition devices such as ammeters, voltmeters, and temperature sensors are deployed at key locations of equipment in the smart substation. The monitoring data are systematically collected at a sampling frequency of once an hour or more frequently (the specific sampling frequency can be set or adjusted according to demand). Each collection will obtain data of the above-mentioned multiple parameter items. In this sampling process, the values ​​of each parameter item sampled are arranged in order of sampling time. Therefore, each parameter item is composed of a series of continuous data points. The data collection of this module can provide a data basis for the operation and maintenance and safety of the smart substation for subsequent analysis.

[0038] The data segmentation module S200 obtains the error threshold of each parameter item, performs segmentation processing on each parameter item collected by using a sliding window trajectory algorithm to obtain multiple segments of each parameter item, and determines a target segment of each parameter item according to the data points included in each segment and the error threshold of the parameter item;

[0039] The process of obtaining the error threshold of each parameter item includes:

[0040] Get all the data points included in the parameter item, and use the box plot algorithm to get the upper quartile and lower quartile of all the data points included in the parameter item;

[0041] Calculate the difference between the upper quartile and the lower quartile, and use half of the absolute value of the difference as the error threshold of the parameter item. Indicates the serial number of the parameter item, then The error threshold of the parameter item is recorded as .

[0042] The sliding window trajectory algorithm is used to segment each parameter item collected to obtain multiple segments of each parameter item. The process includes:

[0043] The sliding window trajectory compression algorithm is a common data compression algorithm, and its general process is as follows:

[0044] Determine a sliding window of fixed size, and the number of data points in the window can be flexibly set according to actual conditions;

[0045] Data point selection: As the window slides over the time series, the data points in the window are updated accordingly;

[0046] Linear regression analysis: Apply the linear regression algorithm to each set of data points in the window to find a straight line that best represents the trend of these data points. The equation of this line is usually in the form of ,in is the slope, is the intercept, and are the independent and dependent variables respectively;

[0047] Fitting straight line calculation: Use the least squares method or other suitable optimization algorithms to calculate the straight line equation parameters that minimize the sum of the squares of the vertical distances from all data points in the window to the straight line. and ;

[0048] Error judgment: Compare the fitted straight line with each data point in the window, calculate the error between the data point and the straight line, and use this as the basis for judging whether to compress or truncate the data;

[0049] Compression segmentation: When compressing data, data truncation or compression is performed only when the error between the data points in the window and the fitting line exceeds the preset error threshold. Otherwise, the original data is retained to maintain data accuracy.

[0050] In this embodiment, a sliding window trajectory algorithm is used to divide all data points included in each parameter item into multiple segments, that is, each parameter item is divided into multiple segments, and each segment includes a number of data points.

[0051] Among them, the specific method for determining the target segment of each parameter item based on the data points contained in each segment and the error threshold of the parameter item is: comparing the values ​​of all data points contained in each segment with the error threshold of the parameter item corresponding to the segment; if the values ​​of the data points in the segment do not exceed the error threshold, then the segment is the target segment of the parameter item; otherwise, the segment is not the target segment of the parameter item.

[0052] The data calculation module S300 includes an abnormality degree calculation unit S301, a fluctuation coefficient calculation unit S302, and a preference degree calculation unit S303;

[0053] The abnormality degree calculation unit S301 determines the comprehensive abnormality degree of each target segment according to the actual value of each data point in each target segment and its predicted value obtained by the sliding window trajectory algorithm, as well as the error threshold;

[0054] In this unit, for each target segment of each parameter item, the difference between the value of each data point in the target segment and the fitted straight line does not exceed the error threshold, but the difference between the segment as a whole and the fit may be large. Direct compression will cause large data distortion. Therefore, this module determines the difference between each data point in each segment and the fitted straight line to determine the overall difference between the segment and the fitted straight line.

[0055] The calculation method of the comprehensive abnormality degree of each target segment is as follows:

[0056]

[0057] In the formula, For the The first parameter The comprehensive difference of target segments, For the The first parameter The number of data points contained in each target segment. is the sequence number of the data point, The value of Any integer in the range, For the The first parameter Target segment The actual value of the data point, For the The first parameter Target segment The predicted value of the data point, is the preset acceptable overall error ratio, and the empirical value is This embodiment does not limit the acceptable ratio of the overall error, and the implementer can set the acceptable ratio of the overall error according to the actual implementation situation.

[0058] This formula measures the degree of deviation of the data point based on the gap between each data point in the segment and the fitted straight line. The product of the acceptable overall error ratio and the error threshold is used as the acceptable error range of the segment (the denominator in the formula). The ratio of the deviation of the data point to the acceptable error range is used as the error degree corresponding to each data point in the segment. The error degrees of all data points in the segment are then comprehensively calculated to obtain the comprehensive difference of the segment.

[0059] The fluctuation coefficient calculation unit S302 determines the fluctuation coefficient of each target segment based on the actual numerical difference between adjacent data points in each target segment and the error threshold;

[0060] In this unit, for each target segment of each parameter item, the data points in the segment may have more fluctuations. Although the gap between each data point that produces fluctuations and the fitting straight line does not exceed the error threshold, if such a data segment is compressed, the original fluctuation characteristics of the data will be lost, resulting in a large degree of data distortion. Therefore, the fluctuation characteristics of the data points in the segment should also be analyzed.

[0061] The fluctuation coefficient of each target segment satisfies the following relationship:

[0062]

[0063] In the formula, For the The first parameter The fluctuation coefficient of the target segment, For the The first parameter The number of data points contained in each target segment. is the sequence number of the data point, The value of Any integer in the range, For the The first parameter Target segment The actual value of the data point, For the The first parameter Target segment The actual value of the data point, The natural constant is the exponential function of the base, is the preset acceptable overall error ratio, For the The error threshold of the parameter item.

[0064] The numerator of the formula reflects the numerical difference between adjacent data points in the target segment. The ratio of this numerical difference to the acceptable error range of the segment (the denominator in the formula) is calculated, and the data fluctuation in the segment is judged based on the ratio. If the ratio is greater than 1, it means that the data points in the segment are likely to fluctuate to a large extent relative to the fitted line. If the ratio is less than or equal to 1, it is considered that the data points in the segment are likely to fluctuate to a small extent relative to the fitted line, which is considered to be a negligible fluctuation. The exponential function with base will convert ratios less than 1 to a range less than 1, and ratios greater than 1 to a range greater than 1. Moreover, a slight increase in the ratio of the range greater than 1 will cause a sharp increase in the result of the exponential function, thereby distinguishing the parts of the data segment with greater fluctuations.

[0065] The numerator of the formula reflects the numerical difference between adjacent data points in the target segment, which reflects the fluctuation of data points in the local range. The denominator is the product of the preset acceptable overall error ratio and the error threshold of the corresponding parameter item of the segment, which represents the acceptable relative threshold for the data fluctuation of the parameter item in the segment. When the ratio of the numerator to the denominator is greater than 1, it means that the difference between adjacent data points exceeds the acceptable fluctuation range of the data segment based on the error threshold, suggesting that the data segment may have a large degree of fluctuation. On the contrary, if the ratio is less than or equal to 1, it is considered that the adjacent data points in the data segment have a small fluctuation, close to fluctuating within the acceptable error range, and can be regarded as having no significant fluctuation or belonging to a negligible fluctuation range. The formula is expressed by the natural exponential The ratio is operated by an exponential function with base 1, so that ratios less than 1 are mapped to a result interval less than 1, while ratios greater than 1 are mapped to an exponential result interval greater than 1 and growing rapidly, thereby enhancing the ability to distinguish differences in segmented fluctuations, especially emphasizing the part where the data fluctuates violently. In summary, this formula effectively quantifies the degree of fluctuation within the target segment and distinguishes between small and large data fluctuations by performing exponential conversion on the ratio of the difference between adjacent data points in the segment and the preset error range.

[0066] The preference calculation unit S303 determines the preference value of each target segment based on the fluctuation coefficient and the comprehensive abnormality degree of each target segment;

[0067] In this unit, the comprehensive difference and fluctuation coefficient of the data segment can be used to obtain the degree of deviation between the actual data in the data segment and the fitted straight line, as well as the smoothed fluctuations, so as to judge the preferred degree of the data segment.

[0068] The preferred value of each target segment is calculated by taking the inverse of the product of the comprehensive difference and the fluctuation coefficient of each target segment as the preferred value of the target segment, and specifically calculating according to the following formula:

[0069]

[0070] In the formula, Indicates the preference value of the jth parameter item corresponding to the sth target segment, Represents the comprehensive difference of the target segment, reflecting the overall deviation between the actual value of the segment and its fitting trend. is the fluctuation coefficient of the target segment, which expresses the intensity of the fluctuation between adjacent data points in the segment and its relationship with the preset error range. and volatility coefficient , it can comprehensively evaluate the fluctuation of the internal data points of the segment and the degree of deviation from the theoretical expectation while maintaining the original trend fitting straight line. This means that the data of the target segment is closer to the ideal fitting state while maintaining a low comprehensive difference and moderate fluctuation, and is therefore more suitable for subsequent data compression and storage processing.

[0071] The data compression and storage module S400 selects the target segments and compresses and stores them according to the preference level of each target segment.

[0072] The specific screening process of this module is as follows:

[0073] First, a preference threshold is set for each target segment. During this process, if the preference value of a target segment is higher than the preference threshold set for it, the target segment is identified as a high-quality preferred data segment and is suitable for compression and storage. On the contrary, if the preference value of the target segment fails to reach the set threshold, the segment is considered not to be a preferred data segment and is not suitable for compression. The original data is directly stored during storage.

[0074] In specific operation, the system will compare the comprehensive preference level of each data segment with the preset preference level threshold. In this embodiment, a data segment preference level threshold is preset. , the experience value is , which can be set by yourself. When the preference degree of a data segment is higher than the threshold, the data segment is considered to be a preferred data segment and is suitable for compression using the sliding window trajectory algorithm. For those segments whose preference degree does not meet the standard, even if the data points therein do not exceed the error threshold, they will still be considered to have no compression value, and thus will be reasonably truncated or re-divided to optimize the data segment structure.

[0075] Next, the system performs compression operations on the selected preferred data segments and stores the compression results. By using the sliding window trajectory algorithm to compress the smart substation operation data, it can not only significantly reduce the demand for data storage space, but more importantly, this method fully considers the degree of preference of the data segments during the compression process, so that it can more effectively maintain the original appearance of the data, minimize data distortion, and ensure that the subsequent statistical analysis results are more accurate and reliable. Finally, the system obtains and stores the compressed data for subsequent in-depth analysis.

[0076] In the description of this specification, "plurality" or "several" means at least two, such as two, three or more, etc., unless otherwise clearly and specifically defined.

[0077] Although this specification has shown and described a number of embodiments of the present invention, it will be apparent to those skilled in the art that such embodiments are provided by way of example only. Those skilled in the art will conceive of many modifications, changes and alternatives without departing from the ideas and spirit of the present invention. It should be understood that in the practice of the present invention, various alternatives to the embodiments of the present invention described herein may be employed.

Claims

1. A statistical analysis system based on intelligent substation operation data management and control, characterized in that: include: The data acquisition module is used to collect various parameter data of the smart substation regularly or multiple times as needed, wherein the parameter data covers multiple different types of parameter items, and each parameter item is composed of a series of continuous data points; A data segmentation module obtains the error threshold of each parameter item, performs segmentation processing on each collected parameter item using a sliding window trajectory algorithm to obtain multiple segments of each parameter item, and determines a target segment of each parameter item based on the data points contained in each segment and the error threshold of the parameter item; Obtaining the error threshold of each parameter item in the data segmentation module includes: obtaining the upper quartile and the lower quartile of all data points included in each parameter item through a box plot algorithm, calculating the difference between the upper quartile and the lower quartile, and taking half of the absolute value of the difference as the error threshold of the parameter item; A data calculation module, including an abnormality degree calculation unit, a fluctuation coefficient calculation unit and a preference degree calculation unit; The abnormality degree calculation unit determines the comprehensive abnormality degree of each target segment according to the actual value of each data point in each target segment and its predicted value obtained by the sliding window trajectory algorithm, as well as the error threshold; The fluctuation coefficient calculation unit determines the fluctuation coefficient of each target segment based on the actual numerical difference between adjacent data points in each target segment and the error threshold; The preference calculation unit determines the preference value of each target segment based on the fluctuation coefficient and the comprehensive abnormality degree of each target segment; The data compression and storage module uses the preference level of each target segment to screen the target segment and compress and store it.

2. A statistical analysis system based on intelligent substation operation data management and control according to claim 1, characterized in that: The fluctuation coefficient calculation unit determines the fluctuation coefficient of each target segment according to the following relationship: In the formula, is the jth parameter item The fluctuation coefficient of the target segment, For the The first parameter The number of data points contained in the target segment, is the sequence number of the data point, The value of Any integer in the range, For the The first parameter Target segment The actual value of the data point, For the The first parameter Target segment The actual value of the data point, The natural constant is the exponential function of the base, is the preset acceptable overall error ratio, For the The error threshold of the parameter item.

3. A statistical analysis system based on intelligent substation operation data management and control according to claim 2, characterized in that: The abnormality degree calculation unit determines the comprehensive abnormality degree of each target segment according to the following relationship: In the formula, For the The first parameter The comprehensive difference of target segments, For the The first parameter The number of data points contained in the target segment, is the sequence number of the data point, The value of Any integer in the range, For the The first parameter Target segment The actual value of the data point, For the The first parameter Target segment The predicted value of the data point, is the preset acceptable overall error ratio, For the The error threshold of the parameter item.

4. A statistical analysis system based on intelligent substation operation data management and control according to claim 3, characterized in that: The preference calculation unit determines the preference value of each target segment, including: The inverse of the product of the comprehensive difference and the fluctuation coefficient of each target segment is used as the preference value of the target segment.

5. A statistical analysis system based on intelligent substation operation data management and control according to claim 1, characterized in that: Determining the target segmentation of each parameter item in the data segmentation module includes: The values ​​of all data points contained in each segment are compared with the error threshold of the parameter item corresponding to the segment. If the values ​​of the data points in the segment do not exceed the error threshold, the segment is the target segment of the parameter item, otherwise the segment is not the target segment of the parameter item.

6. A statistical analysis system based on intelligent substation operation data management and control according to claim 1, characterized in that: The data compression and storage module uses the preference level of each target segment to screen and compress and store the target segment, including: Set a preference threshold for each target segment; If the preference value of the target segment is greater than the preference threshold of the target segment, the target segment is a preferred data segment and is compressed and stored; otherwise, the target segment does not belong to the preferred data segment and is directly stored without compression.

Images