Self-monitoring method for intelligent box-type substation

Abstract

The present invention relates to the field of data processing technology, and in particular to a self-monitoring method for an intelligent box-type substation, comprising: obtaining a real-time output power curve of a substation and dividing it into several segments, identifying several sudden drop bands; obtaining the degree of recovery and low-level stability of each recovery band after each sudden drop band according to the performance of abnormal fluctuations on the curve, thereby obtaining the abnormal interval corresponding to each sudden drop band and the abnormal value of each abnormal interval; obtaining the weights of all data points on the real-time output power curve of the substation in an ARIMA model according to the data point distribution of each abnormal interval and its abnormal value and normal interval, and finally predicting the real-time output power curve of the substation. The present invention provides a more accurate load forecast by accurately identifying the abnormal intervals of sudden drop, low-level maintenance and recovery oscillation caused by sudden abnormalities, thereby ensuring the self-monitoring efficiency and accuracy of the intelligent box-type substation.

Description

Technical Field

[0001] The invention relates to the technical field of data processing, and in particular to a self-monitoring method for an intelligent box-type substation. Background Art

[0002] The box-type substation is a complete set of power distribution equipment with a compact structure. Each box constitutes an independent system. The combination method is flexible and can be freely combined according to the actual situation to meet the needs of safe operation. It is suitable for urban distribution networks, commercial centers, residential areas and industrial parks. With the continuous development of the power industry, the self-monitoring of intelligent box-type substations has become the mainstream choice for unmanned substation solutions. The traditional manual operation of substations has many problems, such as safety hazards, low efficiency, and high labor costs. The intelligent box-type substation monitoring system realizes real-time monitoring and intelligent control of substations through digital and intelligent means, providing new ideas and solutions for the development of the power industry.

[0003] In the wide coverage area of ​​power supply of substations, when an abnormal situation occurs in a certain area, the existing system is often difficult to reflect in time due to insufficient monitoring, which directly leads to significant time delays in problem discovery and processing. Manual intervention for abnormal investigation and confirmation is not only cumbersome and time-consuming, but also susceptible to human factors, increasing the risk of error, thereby further delaying the rapid recovery ability of the power system. In addition, in the complex total output power curve of the substation, when trying to screen fluctuations caused by abnormalities in a specific area, it is often interfered by natural fluctuations caused by the power consumption or natural conditions of the area, making it more difficult to identify abnormalities and difficult to ensure the accuracy and timeliness of the analysis. Summary of the invention

[0004] The self-monitoring method for intelligent box-type substation of the present invention adopts the following technical scheme:

[0005] An embodiment of the present invention provides a self-monitoring method for an intelligent box-type substation, the method comprising the following steps:

[0006] Obtain the real-time output power curve of the substation;

[0007] Obtain the principal component direction of the real-time output power curve of the substation; on the real-time output power curve of the substation, divide the curve into several segments according to all extreme value points, record the segment on the right side of each maximum value point as a falling band, and record the segment on the left side of each maximum value point as a rising band; obtain several sudden drop bands according to the output power value and principal component direction of each falling band of the real-time output power curve of the substation;

[0008] According to the output power values ​​of each sudden drop band and each rebound band after each sudden drop band and the output power values ​​of all segments of each rebound band within a preset reference range, the recovery degree of each rebound band after each sudden drop band is obtained; according to the output power values ​​between each sudden drop band and each rebound band after each sudden drop band and the direction of the main components, the low-level stability degree of each rebound band after each sudden drop band is obtained; according to the recovery degree and low-level stability degree of each rebound band after each sudden drop band, the abnormal interval corresponding to each sudden drop band and the abnormal value of each abnormal interval are obtained;

[0009] According to the abnormal value of each abnormal interval, the weight of the data points in each abnormal interval in the ARIMA model is determined; all intervals on the real-time output power curve of the substation except the abnormal interval are recorded as normal intervals; according to the data point distribution of each data point in each normal interval within a preset reference range, the weight of each data point in each normal interval in the ARIMA model is obtained; according to all data points of the real-time output power curve of the substation and their weights, the real-time output power curve of the substation is predicted to obtain the real-time predicted power of the substation.

[0010] Furthermore, the step of obtaining a plurality of sudden drop bands according to the output power value and the main component direction of each drop band of the real-time output power curve of the substation includes the following specific steps:

[0011] According to the output power value and the main component direction of each declining band of the real-time output power curve of the substation, the sudden drop degree of each declining band is obtained;

[0012] All the decreasing bands on the real-time output power curve of the substation with a sudden drop degree greater than a preset judgment threshold are recorded as sudden drop bands.

[0013] Furthermore, the output power value and the main component direction of each decreasing band of the real-time output power curve of the substation are used to obtain the sudden drop degree of each decreasing band. The specific calculation formula is:

[0014]

[0015] In the formula, Indicates The degree of sudden drop in each descending band; Indicates The first output power value of the descending band; Indicates The last output power value of a descending band; Indicates The number of data points in a descending band; Indicates The average value of the shortest distance between all data points in a descending band and the straight line where the principal component direction lies; is a linear normalization function.

[0016] Further, the recovery degree of each recovery band after each sudden drop band is obtained according to the output power value of each sudden drop band and each recovery band after each sudden drop band and the output power values ​​of all segments of each recovery band within a preset reference range, including the following specific steps:

[0017] According to the output power values ​​of each sudden drop band and each recovery band after each sudden drop band, the recovery evaluation of each recovery band after each sudden drop band is obtained;

[0018] According to the output power values ​​of all segments of each rebound band after each sudden drop band within a preset reference range, the oscillation possibility of each rebound band after each sudden drop band is obtained;

[0019] The product of the recovery evaluation of any recovery band after any sudden drop band and the oscillation possibility is recorded as the recovery degree of each recovery band after the sudden drop band.

[0020] Furthermore, according to the output power values ​​of each sudden drop band and each recovery band after each sudden drop band, the recovery evaluation of each recovery band after each sudden drop band is obtained, and the specific calculation formula is:

[0021]

[0022] In the formula, Indicates The first The rebound evaluation of each rebound band; Indicates The first output power value of a sudden drop band; Indicates The last output power value of a sudden drop band; Indicates The first The last output power value of a recovery band; Indicates The first The first output power value of the recovery band.

[0023] Furthermore, the oscillation possibility of each rebound band after each sudden drop band is obtained according to the output power values ​​of all segments within a preset reference range of each rebound band after each sudden drop band. The specific calculation formula is:

[0024]

[0025] In the formula, Indicates The first The possibility of oscillation in the rebound band; Indicates The first The number of segments within the adjacent reference range on both sides of a rebound band; Indicates The first Within the reference range of the rebound band The first output power value of each segment; Indicates The first Within the reference range of the rebound band The last output power value of a segment; Indicates The first Within the reference range of the rebound band The number of data points per segment; represents the absolute value function.

[0026] Furthermore, the low-level stability of each rebound band after each sudden drop band is obtained according to the output power value between each sudden drop band and each rebound band after each sudden drop band and the main component direction. The specific calculation formula is:

[0027]

[0028] In the formula, Indicates The first The stability of the low level of each rebound band; Indicated in The first dip band and the The first Among all the data points between the rebound bands, the output power value is greater than the The first dip and the next The number of data points with the maximum output power value in the rebound band; Indicates The first dip band and the The first The average value of the shortest distance between all data points and the straight line of the principal component direction between the rebound bands.

[0029] Furthermore, the abnormal interval corresponding to each sudden drop band and the abnormal value of each abnormal interval are obtained according to the recovery degree and low-level stability degree of each recovery band after each sudden drop band, and the specific steps include the following:

[0030] For All the rebound bands after the sudden drop band are normalized using the linear normalization algorithm to calculate the rebound degree of all the rebound bands. and low stability Normalize the The recovery band corresponding to the maximum value of the sum of the normalized value of the recovery degree of each recovery band after the sudden drop band and the normalized value of the low stability degree is taken as the first A rebound band after a sudden drop band;

[0031] The first The first dip band to the The interval formed by the rebound band of the sudden drop band is taken as the The abnormal interval corresponding to the sudden drop band;

[0032] The maximum value of the sum of the normalized values ​​is recorded as The abnormal value of the abnormal interval corresponding to the sudden drop band.

[0033] Furthermore, according to the outlier value of each outlier interval, the weight of the data point in each outlier interval in the ARIMA model is determined, and the specific calculation formula is:

[0034]

[0035] In the formula, Indicates The weight of the data points in the abnormal interval in the ARIMA model; Indicates The outliers in the outlier interval.

[0036] Furthermore, according to the data point distribution of each data point in each normal interval within a preset reference range, the weight of each data point in each normal interval in the ARIMA model is obtained, and the specific calculation formula is:

[0037]

[0038] In the formula, Indicates The normal interval The weight of the data point in the ARIMA model; Indicates the number of data points within the preset reference range; Indicates The normal interval Output power value of each data point; Indicates The normal interval The first data point within the preset reference range Output power value of each data point; represents the absolute value function; is an exponential function with a natural constant as its base.

[0039] The beneficial effects of the technical solution of the present invention are:

[0040] In the embodiment of the present invention, by segmenting the real-time output power curve of the substation according to the extreme value points, the rising and falling trends of the curve can be clearly identified, which facilitates the identification of subsequent sudden drop bands; according to the output power value and the main component direction of each falling band of the real-time output power curve of the substation, a number of sudden drop bands are obtained, and the abnormal intervals caused by certain abnormal conditions are accurately identified; according to the output power value of each sudden drop band and each rebound band after each sudden drop band, and the output power values ​​of all segments of each rebound band within a preset reference range, the recovery degree and low-level stability degree of each rebound band after each sudden drop band are obtained, the recovery ability of the power system after a fault and the stability of the recovery process are effectively evaluated, and the abnormal interval and the abnormal values ​​of the abnormal interval are further effectively screened; by adjusting the weights of the data points in the abnormal interval and the data points in the normal interval, the ARIMA model can more accurately reflect the normal operating state of the power system, thereby improving the prediction accuracy of the model. Thus, the present invention provides more accurate load forecasting by accurately identifying abnormal intervals of sudden drop, low-level maintenance and rebound oscillation caused by sudden abnormalities, thereby ensuring the self-monitoring efficiency and accuracy of the intelligent box-type substation. BRIEF DESCRIPTION OF THE DRAWINGS

[0041] In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings required for use in the embodiments or the description of the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative work.

[0042] Figure 1 A flowchart of the steps of the self-monitoring method for an intelligent box-type substation according to the present invention;

[0043] Figure 2 The natural fluctuation on the real-time output power curve of a substation provided in this embodiment;

[0044] Figure 3This embodiment provides an abnormal band on a real-time output power curve of a substation. DETAILED DESCRIPTION

[0045] In order to further explain the technical means and effects adopted by the present invention to achieve the predetermined invention purpose, the self-monitoring method for intelligent box-type substation proposed by the present invention, its specific implementation method, structure, characteristics and effects are described in detail as follows in combination with the accompanying drawings and preferred embodiments. In the following description, different "one embodiment" or "another embodiment" does not necessarily refer to the same embodiment. In addition, specific features, structures or characteristics in one or more embodiments may be combined in any suitable form.

[0046] Unless defined otherwise, all technical and scientific terms used herein have the same meaning as commonly understood by one of ordinary skill in the art to which this invention belongs.

[0047] The specific scheme of the self-monitoring method for an intelligent box-type substation provided by the present invention is described in detail below with reference to the accompanying drawings.

[0048] See also Figure 1 , which shows a flowchart of a self-monitoring method for an intelligent box-type substation provided by an embodiment of the present invention, the method comprising the following steps:

[0049] Step S001: Obtain the real-time output power curve of the substation.

[0050] This embodiment needs to realize self-monitoring of the box-type substation, so it is necessary to select a certain dimension of data that can reflect the working status of the substation. In the power system, the real-time output power of the substation will fluctuate with various factors such as time, load changes, and power grid dispatching. The real-time output power curve of the substation reflects the amount of power transmitted by the substation to the power grid at different time points, which can help power dispatchers understand the real-time operating status of the power grid, perform power dispatch and load management, and ensure the safe and stable operation of the power grid.

[0051] Specifically, monitoring starts at 0:00 every day, with each monitoring moment being every ten seconds. The real-time output power of the substation is monitored and obtained to form a real-time output power curve of the substation. The real-time output power curve of the substation is then transmitted to the cloud platform for intelligent analysis and prediction.

[0052] Step S002: Obtain the principal component direction of the real-time output power curve of the substation; on the real-time output power curve of the substation, divide the curve into several segments according to all extreme points, record the segment on the right side of each maximum point as a declining band, and record the segment on the left side of each maximum point as a recovery band; obtain several sudden drop bands according to the output power value and principal component direction of each declining band of the real-time output power curve of the substation.

[0053] It should be noted that the real-time output power curve of the substation comprehensively reflects the sum of the output power of the substation to all areas in its power supply network. The power demand of each area has natural volatility and uncertainty. Therefore, the real-time output power curve of the substation has become a comprehensive reflection of these complex power consumption trends. Therefore, it is impractical to simply attribute any significant changes in the output power of the substation to a single area. It requires more complex data analysis to widely collect and integrate real-time data from various areas. This process undoubtedly increases the complexity of substation self-monitoring and may reduce its operational efficiency to a certain extent. For the self-monitoring of the substation, the core goal is to analyze its ability to maintain normal and efficient operation in the future based on current and historical data. Once the data analysis results show potential abnormalities or signs of risk, the system immediately triggers the early warning mechanism, and the staff takes timely measures to confirm and repair on the spot, thereby effectively curbing the possible problems of grid voltage fluctuations, frequency offsets, etc. that affect the power consumption of the area, and even cause the collapse of the grid.

[0054] It should be further explained that the real-time output power curve of the substation is affected by multi-dimensional data. It contains multiple variables, which represent the power supply conditions of different areas or the power output in different time periods. There is a high probability of linear correlation between the dimensions, that is, the power consumption changes in some areas may be related to other areas or the overall power output trend. This correlation makes direct analysis complicated and difficult to accurately capture the essential characteristics of the data.

[0055] Specifically, according to the PCA algorithm, the principal component direction of the real-time output power curve of the substation is obtained, wherein PCA principal component analysis is a commonly used data analysis method. For a set of data that may have linear correlation between different dimensions, PCA can transform this set of data into linearly independent data between dimensions through orthogonal transformation.

[0056] It should be further explained that in the process of the substation supplying power to the grid, the real-time output power of the substation has natural fluctuations on the real-time output power curve itself as time changes. It is usually affected by the power consumption or natural conditions in the area. It is shown on the curve as continuous, with relatively small fluctuations and within a certain range. There will be no sudden jumps or faults. In sudden abnormal situations, for example, due to equipment failure in a certain area, sudden load changes, etc., the output power will suddenly drop. The amplitude of the sudden drop is usually large, far exceeding the range of natural fluctuations. After the sudden drop, it will remain at a low level for a period of time. During this period of time, the curve still has natural fluctuations, and the data points of the low-level curve segment after the sudden drop are relatively far away from the direction of the main component. In addition, before and after the sudden drop, the output power will have obvious jumps or faults. The natural fluctuations on the real-time output power curve of the substation are as follows: Figure 2 As shown, the abnormal bands are Figure 3 shown.

[0057] Specifically, all extreme value points are obtained on the real-time output power curve of the substation; the curve is divided into several segments according to all extreme value points, and the segment on the right side of each maximum value point is recorded as a descending band, and the segment on the left side of each maximum value point is recorded as a rising band; each extreme value point is the first output power in the segment, and the first output power in the real-time output power curve is also used as a segmentation point;

[0058] According to the output power value and main component direction of each declining band of the real-time output power curve of the substation, the sudden drop degree of each declining band is obtained. The specific calculation formula is:

[0059]

[0060] In the formula, Indicates The degree of sudden drop in each descending band; Indicates The first output power value of the descending band; Indicates The last output power value of a descending band; Indicates The number of data points in a descending band; Indicates The average value of the shortest distance between all data points in a descending band and the straight line where the principal component direction lies; is a linear normalization function, and the normalization object is all descending bands .

[0061] It should be noted that Indicates The greater the value of the descending speed of the descending band, the faster the The more drastic the drop in output power in the first decreasing band, the more it deviates from the natural fluctuation, that is, the greater the degree of sudden drop; The average value of the shortest distance between all data points in the descending band and the principal component direction The larger it is, the greater the degree of deviation from the overall fluctuation of the real-time output power curve, that is, the greater the degree of sudden drop.

[0062] Furthermore, the judgment threshold value preset in this embodiment is 0.7, which is used as an example for description. Other implementations may be set to other values, which are not limited in this embodiment;

[0063] All the decreasing bands on the real-time output power curve of the substation with a sudden drop degree greater than a preset judgment threshold are recorded as sudden drop bands; thereby, a plurality of sudden drop bands of the real-time output power curve of the substation are obtained.

[0064] Step S003: According to the output power values ​​of each sudden drop band and each rebound band after each sudden drop band and the output power values ​​of all segments of each rebound band within a preset reference range, the recovery degree of each rebound band after each sudden drop band is obtained; according to the output power values ​​between each sudden drop band and each rebound band after each sudden drop band and the direction of the main components, the low-level stability degree of each rebound band after each sudden drop band is obtained; according to the recovery degree and the low-level stability degree of each rebound band after each sudden drop band, the abnormal interval corresponding to each sudden drop band and the abnormal value of each abnormal interval are obtained.

[0065] It should be noted that when a sudden abnormality occurs in a certain area, maintenance personnel will repair it in time, and the power consumption in the area will return to normal after a period of abnormality. This is reflected in the real-time output power curve of the substation. When a sudden situation occurs in a certain area, causing a sudden drop in real-time output power, the real-time output power will drop to a low level after the sudden drop band, and then there will be a data point where it will start to rise again, and as the fault is cleared and the system is restored, the real-time output power curve may experience a period of oscillation. This is because the generators, loads, transmission lines and other components in the system will produce dynamic responses when responding to sudden abnormalities, and these responses may overlap and cause the real-time output power curve of the substation to oscillate.

[0066] Specifically, the preset reference range in this embodiment is two minutes, that is, the preset reference range of the recovery band is within one minute before the first output power in the recovery band and within one minute after the last output power; and the preset reference range of subsequent data points is the data points within one minute before and after the data point. This is used as an example for description, and other values ​​can be set in other implementation modes;

[0067] According to the output power values ​​of each sudden drop band and each recovery band after each sudden drop band and the output power values ​​of all segments of each recovery band within the preset reference range, the recovery degree of each recovery band after each sudden drop band is obtained. The specific calculation formula is:

[0068]

[0069] In the formula, Indicates The first The degree of recovery of each recovery band; Indicates The first The rebound evaluation of each rebound band; Indicates The first The possibility of oscillation in the rebound band; Indicates The first output power value of a sudden drop band; Indicates The last output power value of a sudden drop band; Indicates The first The last output power value of a recovery band; Indicates The first The first output power value of the rebound band; Indicates The first The number of segments within the adjacent reference range on both sides of a rebound band; Indicates The first Within the reference range of the rebound band The first output power value of each segment; Indicates The first Within the reference range of the rebound band The last output power value of a segment; Indicates The first Within the reference range of the rebound band The number of data points per segment; represents the absolute value function.

[0070] It should be noted that Reflect the The first The larger the value of the recovery amplitude of the recovery band, the more likely it is that the recovery of the entire area is caused by the completion of maintenance rather than natural fluctuations, and the The rebound amplitude of the first rebound band is similar to that of the The closer the reduction amplitudes of the first and second dip bands are, the more likely it is that this is the corresponding recovery band of the dip band. Since the recovery band corresponding to the output power dip is often accompanied by an oscillation process, The first The more segments there are within the reference range of the rebound band, and the more significant the changes in the segments, the higher the possibility of oscillation, that is, The bigger; when The higher the recovery evaluation of the recovery band after a sudden drop band, the higher the possibility of oscillation, and the higher the recovery degree of the recovery band.

[0071] It should be further explained that for Each rebound band after the first drop band is The output power curve between the sudden drop bands reflects the comprehensive superposition of the power loads of all areas powered by the substation except the area where the abnormal fault occurs, and should be presented as a curve that is lower than the initial output power value of the sudden drop band (that is, the output power value before the fault occurs) and the ending output power value of the recovery band (that is, the output power value after the fault is recovered); the output power curve between the sudden drop band and the recovery band should be in the normal operating state of all other areas of the substation, and the distance between the data points and the main component direction in this interval should be very small.

[0072] According to the output power value between each sudden drop band and each rebound band after each sudden drop band and the direction of the main component, the low-level stability of each rebound band after each sudden drop band is obtained. The specific calculation formula is as follows:

[0073]

[0074] In the formula, Indicates The first The stability of the low level of each rebound band; Indicated in The first dip band and the The first Among all the data points between the rebound bands, the output power value is greater than the The first dip and the next The number of data points with the maximum output power value in the rebound band; Indicates The first dip band and the The first The average value of the shortest distance between all data points and the straight line of the principal component direction between the rebound bands.

[0075] It should be noted that when The first dip band and the The first The output power value of all data points between the rebound bands is greater than The first dip band and the The first The smaller the number of maximum output power values ​​in the rebound band, and the smaller the average value of the shortest distance between the data point and the main component direction, the more it can be reflected in the The first dip band and the The first The low stability of the output power curve between the first rebound bands, that is, The more likely a rebound band is, the more likely it is the rebound band corresponding to a sudden drop in output power.

[0076] Furthermore, for the All the rebound bands after the sudden drop band are normalized using the linear normalization algorithm to calculate the rebound degree of all the rebound bands. and low stability Normalize the The recovery band corresponding to the maximum value of the sum of the normalized value of the recovery degree of each recovery band after the sudden drop band and the normalized value of the low stability degree is taken as the first The recovery band of the first falling band; The first dip band to the The interval formed by the rebound band of the sudden drop band is taken as the The abnormal interval corresponding to the sudden drop band; the maximum value of the sum of the normalized values ​​is recorded as The abnormal values ​​of the abnormal intervals corresponding to the sudden drop bands are obtained; thus, several abnormal intervals and their abnormal values ​​are obtained.

[0077] Step S004: according to the abnormal value of each abnormal interval, determine the weight of the data points in each abnormal interval in the ARIMA model; record all intervals on the real-time output power curve of the substation except the abnormal interval as normal intervals; according to the data point distribution of each data point in each normal interval within a preset reference range, obtain the weight of each data point in each normal interval in the ARIMA model; according to all data points of the real-time output power curve of the substation and their weights, predict the real-time output power curve of the substation to obtain the real-time predicted power of the substation.

[0078] Specifically, according to the outlier value of each outlier interval, the weight of the data point in each outlier interval in the ARIMA model is determined. The specific calculation formula is:

[0079]

[0080] In the formula, Indicates The weight of the data points in the abnormal interval in the ARIMA model; Indicates The outliers in the outlier interval.

[0081] It should be noted that Outliers in the outlier interval The value range is ,therefore The value range is , and when the outlier value in the outlier interval is larger, its weight in the ARIMA model is smaller.

[0082] Furthermore, all intervals except the abnormal interval on the real-time output power curve of the substation are recorded as normal intervals;

[0083] According to the distribution of each data point in each normal interval within the preset reference range, the weight of each data point in each normal interval in the ARIMA model is obtained. The specific calculation formula is:

[0084]

[0085] In the formula, Indicates The normal interval The weight of the data point in the ARIMA model; Indicates the number of data points within the preset reference range; Indicates The normal interval Output power value of each data point; Indicates The normal interval The first data point within the preset reference range Output power value of each data point; represents the absolute value function; is an exponential function with a natural constant as the base. Model to present inverse proportional relationship and normalization, As the input of the model, the implementer can set the inverse proportional function and normalization function according to the actual situation.

[0086] It should be noted that when The normal interval The greater the difference between a data point and all data points in the preset reference range, the higher the abnormality of the data point, and the smaller its weight in the ARIMA model.

[0087] According to all data points and weights of the real-time output power curve of the substation, the weighted ARIMA model is used to predict the real-time output power curve of the substation, and the power prediction value of the next moment is output in real time to obtain the real-time predicted power of the substation; the output power of the box-type substation is monitored in time according to the real-time predicted power, thereby realizing a self-monitoring method for the box-type substation.

[0088] At this point, this embodiment is completed.

[0089] The above description is only a preferred embodiment of the present invention and is not intended to limit the present invention. Any modifications, equivalent substitutions, improvements, etc. made within the principles of the present invention should be included in the protection scope of the present invention.

Claims

1. A self-monitoring method for an intelligent box-type substation, characterized in that: The method comprises the following steps: Obtain the real-time output power curve of the substation; Obtain the principal component direction of the real-time output power curve of the substation; on the real-time output power curve of the substation, divide the curve into several segments according to all extreme value points, record the segment on the right side of each maximum value point as a falling band, and record the segment on the left side of each maximum value point as a rising band; obtain several sudden drop bands according to the output power value and principal component direction of each falling band of the real-time output power curve of the substation; According to the output power values ​​of each sudden drop band and each rebound band after each sudden drop band and the output power values ​​of all segments of each rebound band within a preset reference range, the recovery degree of each rebound band after each sudden drop band is obtained; according to the output power values ​​between each sudden drop band and each rebound band after each sudden drop band and the direction of the main components, the low-level stability degree of each rebound band after each sudden drop band is obtained; according to the recovery degree and low-level stability degree of each rebound band after each sudden drop band, the abnormal interval corresponding to each sudden drop band and the abnormal value of each abnormal interval are obtained; According to the abnormal value of each abnormal interval, the weight of the data point in each abnormal interval in the ARIMA model is determined; all intervals on the real-time output power curve of the substation except the abnormal interval are recorded as normal intervals; according to the distribution of data points of each data point in each normal interval within a preset reference range, the weight of each data point in each normal interval in the ARIMA model is obtained; according to all data points of the real-time output power curve of the substation and their weights, the real-time output power curve of the substation is predicted to obtain the real-time predicted power of the substation; According to the outlier value of each outlier interval, the weight of the data point in each outlier interval in the ARIMA model is determined. The specific calculation formula is: In the formula, Indicates The weight of the data points in the abnormal interval in the ARIMA model; Indicates Outliers in the outlier interval; According to the data point distribution of each data point in each normal interval within the preset reference range, the weight of each data point in each normal interval in the ARIMA model is obtained, and the specific calculation formula is: In the formula, Indicates The normal interval The weight of the data point in the ARIMA model; Indicates the number of data points within the preset reference range; Indicates The normal interval Output power value of each data point; Indicates The normal interval The first data point within the preset reference range Output power value of each data point; represents the absolute value function; is an exponential function with a natural constant as its base.

2. The self-monitoring method for intelligent box-type substation according to claim 1 is characterized in that: The specific steps of obtaining a plurality of sudden drop bands according to the output power value and the main component direction of each drop band of the real-time output power curve of the substation are as follows: According to the output power value and the main component direction of each declining band of the real-time output power curve of the substation, the sudden drop degree of each declining band is obtained; All the decreasing bands on the real-time output power curve of the substation with a sudden drop degree greater than a preset judgment threshold are recorded as sudden drop bands.

3. The self-monitoring method for intelligent box-type substation according to claim 2 is characterized in that: According to the output power value and the main component direction of each decreasing band of the real-time output power curve of the substation, the sudden drop degree of each decreasing band is obtained, and the specific calculation formula is: In the formula, Indicates The degree of sudden drop in each descending band; Indicates The first output power value of the descending band; Indicates The last output power value of a descending band; Indicates The number of data points in a descending band; Indicates The average value of the shortest distance between all data points in a descending band and the straight line where the principal component direction lies; is a linear normalization function.

4. The self-monitoring method for intelligent box-type substation according to claim 1 is characterized in that: The method of obtaining the recovery degree of each recovery band after each sudden drop band according to the output power values ​​of each sudden drop band and each recovery band after each sudden drop band and the output power values ​​of all segments of each recovery band within a preset reference range includes the following specific steps: According to the output power values ​​of each sudden drop band and each recovery band after each sudden drop band, the recovery evaluation of each recovery band after each sudden drop band is obtained; According to the output power values ​​of all segments of each rebound band after each sudden drop band within a preset reference range, the oscillation possibility of each rebound band after each sudden drop band is obtained; The product of the recovery evaluation of any recovery band after any sudden drop band and the oscillation possibility is recorded as the recovery degree of each recovery band after the sudden drop band.

5. The self-monitoring method for intelligent box-type substation according to claim 4 is characterized in that: According to the output power values ​​of each sudden drop band and each recovery band after each sudden drop band, the recovery evaluation of each recovery band after each sudden drop band is obtained, and the specific calculation formula is: In the formula, Indicates The first The rebound evaluation of each rebound band; Indicates The first output power value of a sudden drop band; Indicates The last output power value of a sudden drop band; Indicates The first The last output power value of a recovery band; Indicates The first The first output power value of the recovery band.

6. The self-monitoring method for intelligent box-type substation according to claim 4 is characterized in that: According to the output power values ​​of all segments of each rebound band after each sudden drop band within a preset reference range, the oscillation possibility of each rebound band after each sudden drop band is obtained, and the specific calculation formula is: In the formula, Indicates The first The possibility of oscillation in the rebound band; Indicates The first The number of segments within the adjacent reference range on both sides of a rebound band; Indicates The first Within the reference range of the rebound band The first output power value of each segment; Indicates The first Within the reference range of the rebound band The last output power value of a segment; Indicates The first Within the reference range of the rebound band The number of data points per segment; represents the absolute value function.

7. The self-monitoring method for intelligent box-type substation according to claim 1 is characterized in that: According to the output power value between each sudden drop band and each rebound band after each sudden drop band and the direction of the main component, the low-level stability of each rebound band after each sudden drop band is obtained, and the specific calculation formula is: In the formula, Indicates The first The stability of the low level of each rebound band; Indicated in The first dip band and the The first Among all the data points between the rebound bands, the output power value is greater than the The first dip and the next The number of data points with the maximum output power value in the rebound band; Indicates The first dip band and the The first The average value of the shortest distance between all data points and the straight line of the principal component direction between the rebound bands.

8. The self-monitoring method for intelligent box-type substation according to claim 1 is characterized in that: The method of obtaining the abnormal interval corresponding to each sudden drop band and the abnormal value of each abnormal interval according to the recovery degree and low-level stability degree of each recovery band after each sudden drop band includes the following specific steps: For All the rebound bands after the sudden drop band are normalized using the linear normalization algorithm to calculate the rebound degree of all the rebound bands. and low stability Perform normalization and convert the The recovery band corresponding to the maximum value of the sum of the normalized value of the recovery degree of each recovery band after the sudden drop band and the normalized value of the low stability degree is taken as the first A rebound band after a sudden drop band; The first The first dip band to the The interval formed by the rebound band of the sudden drop band is taken as the The abnormal interval corresponding to the sudden drop band; The maximum value of the sum of the normalized values ​​is recorded as The abnormal value of the abnormal interval corresponding to the sudden drop band.

Images