A circuit breaker operating status monitoring method and system

Abstract

The present invention relates to the field of data processing technology, and in particular to a circuit breaker operation status monitoring method and system, comprising: obtaining a circuit breaker operation status training set; obtaining the possibility of any two state data points constituting a disconnected state interval; screening the disconnected state interval according to the possibility of constituting the disconnected state interval; obtaining the possibility of each state data point belonging to normal data; clustering all state data points to obtain a number of feature groups; obtaining the operation status of each feature group according to the possibility of belonging to normal data; constructing an optimized decision tree according to the operation status of the feature group; and classifying the operation status of the circuit breaker through the optimized decision tree. The present invention improves the accuracy of the monitoring results of the circuit breaker operation status.

Description

Technical Field

[0001] The present invention relates to the technical field of data processing, and in particular to a circuit breaker operating status monitoring method and system. Background Art

[0002] SF6 high-voltage circuit breaker is a common electrical equipment used to control and protect circuits in power systems. Accurately monitoring the operating status of SF6 circuit breakers can diagnose and repair circuit faults more timely. When a training set is used to construct a decision tree to perform binary classification on the circuit breaker operating status, the authenticity of each group of data points belonging to the category is inaccurate, resulting in inaccurate establishment of information entropy, which in turn affects information gain, and finally affects the construction of the decision tree in the information gain ID3 decision tree algorithm. This makes the circuit breaker operating status obtained through the decision tree inaccurate. Summary of the invention

[0003] In order to solve the above problems, the present invention provides a circuit breaker operating status monitoring method and system.

[0004] An embodiment of the present invention provides a method for monitoring the operating status of a circuit breaker, the method comprising the following steps:

[0005] Acquire a circuit breaker operation status training set; the circuit breaker operation status training set includes a plurality of status data points; the status data points refer to temperature data, pressure data and vibration data of the circuit breaker at each sampling time;

[0006] According to the temperature data fluctuation of the state data points in the surrounding area of ​​any two state data points, the possibility of forming a disconnected state interval of any two state data points is obtained; and the disconnected state interval is obtained by screening according to the possibility of forming a disconnected state interval;

[0007] According to the temperature data difference and pressure data difference between each state data point and the surrounding state data points, and whether it is in the disconnected state interval, the possibility of each state data point belonging to normal data is obtained; all state data points are clustered to obtain several feature groups; according to the possibility of belonging to normal data, the operating state of each feature group is obtained; and the optimized decision tree is constructed according to the operating state of the feature group;

[0008] The operating status of the circuit breaker is classified through the optimized decision tree.

[0009] Preferably, the method of obtaining the possibility of forming a disconnected state interval of any two state data points according to the temperature data fluctuation of the state data points in the surrounding area of ​​the any two state data points includes the following specific methods:

[0010] For any two state data points in the circuit breaker operation state training set, one state data point is recorded as the first state data point; the other state data point is recorded as the second state data point;

[0011] Obtain the temperature fluctuation within the left neighborhood of the first state data point and the temperature fluctuation within the right neighborhood of the second state data point;

[0012] Obtaining a temperature fluctuation degree between a first state data point and a second state data point;

[0013] The sum of the temperature fluctuation in the left neighborhood range of the first state data point and the temperature fluctuation in the right neighborhood range of the second state data point is recorded as the first sum value; the inverse proportional value of the first sum value and the product of the temperature fluctuation degree between the first state data point and the second state data point are taken as the possibility of the first state data point and the second state data point constituting a disconnected state interval.

[0014] Preferably, the method of obtaining the temperature fluctuation within the left neighborhood of the first state data point and the temperature fluctuation within the right neighborhood of the second state data point includes:

[0015] Preset a range parameter , the left side of the first state data point state data points constitute a data point sequence as the left neighborhood range of the first state data point; for any state data point in the left neighborhood range of the first state data point, the absolute value of the difference between the temperature data of the state data point and the first state data point is recorded as the first difference absolute value of the state data point; the cumulative sum of the first difference absolute values ​​of all state data points in the left neighborhood range of the first state data point is taken as the temperature volatility in the left neighborhood range of the first state data point;

[0016] The right side of the second state data point state data points constitute a data point sequence as the right neighborhood range of the second state data point; for any state data point in the left neighborhood range of the second state data point, the absolute value of the difference between the temperature data of the state data point and the second state data point is recorded as the second difference absolute value of the state data point; the cumulative sum of the second difference absolute values ​​of all state data points in the right neighborhood range of the second state data point is taken as the temperature volatility in the right neighborhood range of the second state data point.

[0017] Preferably, the step of obtaining the temperature fluctuation degree between the first state data point and the second state data point comprises the following specific methods:

[0018] All state data points between the first state data point and the second state data point are fitted by the least square method to obtain a fitting curve; the integral result of the fitting curve is used as the temperature fluctuation degree between the first state data point and the second state data point.

[0019] Preferably, the possibility of each state data point belonging to normal data is obtained according to the temperature data difference and pressure data difference between each state data point and the surrounding state data points, and whether it is in the disconnected state interval, including the specific method of:

[0020] Preset a range parameter , for any state data point, the left side of the state data point state data points and the right state data points, and the data point sequence formed by the data points is used as the surrounding area sequence of the state data points;

[0021] According to whether the state data point is in a disconnected state interval, obtaining a disconnection characteristic value of the state data point;

[0022] Obtaining the temperature and pressure data volatility of the state data point according to the temperature data difference and the pressure data difference between the state data point and the state data points in the surrounding area sequence;

[0023] Obtaining the temperature and pressure data correlation of the state data point;

[0024] The product of the disconnection characteristic value, the temperature and pressure data fluctuation and the temperature and pressure data correlation is used as the possibility that the state data point belongs to normal data.

[0025] Preferably, the specific method of obtaining the disconnection characteristic value of the state data point includes:

[0026] For any state data point, if the state data point is in the disconnected state interval, 1 is used as the disconnected characteristic value of the state data point; if the state data point is not in the disconnected state interval, 0 is used as the disconnected characteristic value of the state data point.

[0027] Preferably, the obtaining of the temperature and pressure data volatility of the state data point comprises the following specific methods:

[0028] Record any state data point as a reference state data point, and for any state data point in a surrounding region sequence of the reference state data point, record the absolute value of the difference in temperature data between the state data point and the reference state data point as a first temperature difference of the state data point; record the cumulative sum of the first temperature differences of all state data points in the surrounding region sequence of the reference state data point as a first cumulative sum;

[0029] The absolute value of the difference between the pressure data of the state data point and the reference state data point is recorded as the first pressure difference of the state data point; the cumulative sum of the first pressure differences of all state data points in the surrounding area sequence of the reference state data point is recorded as the second cumulative sum;

[0030] The product of the first accumulated sum and the second accumulated sum is taken as the temperature and pressure data fluctuation of the reference state data point.

[0031] Preferably, the specific method of obtaining the temperature and pressure data correlation of the state data point includes:

[0032] For any state data point, the data set consisting of the temperature data of all state data points in the surrounding area sequence of the state data point is recorded as the temperature data set; the data set consisting of the pressure data of all state data points in the surrounding area sequence of the state data point is recorded as the pressure data set; the Pearson correlation coefficient between the temperature data set and the pressure data set is taken as the correlation between the temperature and pressure data of the state data point.

[0033] Preferably, the operation status of each feature group is obtained according to the possibility of belonging to normal data, including the specific method of:

[0034] Preset a threshold parameter For any feature group, the mean value of the probability of all state data points in the feature group belonging to normal data is taken as the probability of the feature group belonging to the disconnected circuit; if the probability of the feature group belonging to the disconnected circuit is greater than or equal to the threshold parameter , the running state of the feature group is recorded as the disconnected state; if the probability of the feature group belonging to the disconnected circuit is less than the threshold parameter , and record the operating status of the feature group as other status.

[0035] The present invention also proposes a circuit breaker operating status monitoring system, comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor implements any one of the steps of the circuit breaker operating status monitoring method when executing the computer program.

[0036] The beneficial effects of the technical solution of the present invention are as follows: the present invention obtains the disconnected state interval according to the possibility of constituting the disconnected state interval; obtains the possibility of each state data point belonging to normal data according to the temperature data difference and pressure data difference between each state data point and the surrounding state data points, and whether it is in the disconnected state interval; clusters all state data points to obtain a number of feature groups; obtains the operating state of each feature group according to the possibility of belonging to normal data; constructs an optimized decision tree according to the operating state of the feature group; based on the volatility of the degree data, divides the training set into normal data and disconnected data, and then corrects the information entropy according to the authenticity of the disconnected data to construct a decision tree that can more accurately predict the operating state, so that the circuit breaker operating state obtained through the decision tree is more accurate, thereby improving the accuracy of the monitoring results of the circuit breaker operating state. BRIEF DESCRIPTION OF THE DRAWINGS

[0037] 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.

[0038] Figure 1 A flow chart of the steps of a circuit breaker operating status monitoring method according to the present invention;

[0039] Figure 2 The present invention is a characteristic relationship flow chart of circuit breaker operation status monitoring. DETAILED DESCRIPTION

[0040] In order to further explain the technical means and effects adopted by the present invention to achieve the predetermined invention purpose, the following is a detailed description of a circuit breaker operation status monitoring method and system proposed by the present invention, its specific implementation method, structure, features and effects, 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.

[0041] 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.

[0042] The specific scheme of a circuit breaker operating status monitoring method and system provided by the present invention is described in detail below with reference to the accompanying drawings.

[0043] See also Figure 1, which shows a flow chart of a method for monitoring the operating status of a circuit breaker provided by an embodiment of the present invention, the method comprising the following steps:

[0044] Step S001: Obtain a circuit breaker operation status training set.

[0045] Specifically, we first need to collect the circuit breaker operation status training set. The specific process is as follows:

[0046] Pressure sensors, temperature sensors and sound sensors are respectively placed in the SF6 high-voltage circuit breaker. The sampling moment is 1 second, and the three dimensional data types of temperature data, pressure data and vibration data are collected in turn each time, and the collection time is 1 hour in total. The three dimensional data of temperature data, pressure data and vibration data at each sampling moment are used as the operating status data sequence of the circuit breaker.

[0047] A multidimensional space is constructed according to the sampling time, temperature data, pressure data and vibration data, and the operating status data sequence of the circuit breaker is input into the multidimensional space to obtain a number of status data points; the data sequence composed of all status data points is recorded as the circuit breaker operating status training set.

[0048] The circuit breaker operation status training set includes a number of status data points; the status data points refer to the temperature data, pressure data and vibration data of the circuit breaker at each sampling moment.

[0049] So far, the circuit breaker operation status training set is obtained through the above method.

[0050] Step S002: according to the temperature data fluctuation of the state data points in the surrounding area of ​​the two state data points, the possibility of forming a disconnected state interval of any two state data points is obtained; and the disconnected state interval is obtained by screening according to the possibility of forming a disconnected state interval.

[0051] It should be noted that for SF6 circuit breakers, when the circuit is cut off, an arc extinction stage and an arc continuation stage will occur; in the arc continuation stage of cutting off the circuit, the gas pressure and the temperature inside the circuit breaker will gradually increase; in the arc extinction stage of cutting off the circuit, the gas pressure and the temperature of the circuit breaker will gradually decrease; because the temperature change lasts for a long time and the pressure change lasts for a short time, the left and right endpoints of the disconnection state interval can be found according to the temperature change, and all state data points in the circuit breaker operation state training set are traversed pairwise to find the two state data points that are most likely to constitute the disconnection state interval. For the left endpoint, the temperature volatility on the left side is weaker, and the temperature volatility on the right side is stronger; for the right endpoint, the temperature volatility on the left side is stronger, and the temperature volatility on the right side is weaker.

[0052] Therefore, based on the data point range formed by the left part of one state data point and the data point range formed by the right part of another state data point, the temperature fluctuation of the two data ranges, and the temperature fluctuation of all state data points between the two state data points, the possibility of the two state data points constituting a disconnected state interval can be quantified; that is, based on the temperature data fluctuation of the state data points in the surrounding area of ​​any two state data points, the possibility of any two state data points constituting a disconnected state interval can be obtained.

[0053] Preferably, in some implementations of the embodiments of the present invention, the calculation method for obtaining the possibility of any two state data points constituting a disconnected state interval is:

[0054] For any two state data points in the circuit breaker operation state training set, one state data point is recorded as the first state data point; the other state data point is recorded as the second state data point;

[0055] Obtain the temperature fluctuation within the left neighborhood of the first state data point and the temperature fluctuation within the right neighborhood of the second state data point;

[0056] Obtaining a temperature fluctuation degree between a first state data point and a second state data point;

[0057] The sum of the temperature fluctuation in the left neighborhood range of the first state data point and the temperature fluctuation in the right neighborhood range of the second state data point is recorded as the first sum value; the inverse proportional value of the first sum value and the product of the temperature fluctuation degree between the first state data point and the second state data point are taken as the possibility of the first state data point and the second state data point constituting a disconnected state interval.

[0058] The specific formula is:

[0059]

[0060] In the formula, Indicates the possibility of the first state data point and the second state data point constituting a disconnected state interval; Indicates the temperature volatility within the left neighborhood of the first state data point; Indicates the temperature fluctuation within the right neighborhood of the second state data point; Indicates the degree of temperature fluctuation between the first state data point and the second state data point; represents an exponential function with a natural constant as the base, and the embodiment adopts The model shows an inverse proportional relationship. As the input of the model, implementers can choose the inverse proportional function according to actual conditions.

[0061] It should be noted that, the smaller the temperature fluctuation in the left neighborhood range of the first state data point and the temperature fluctuation in the right neighborhood range of the second state data point, it means that for the first state data point, the temperature fluctuation on the left is weaker; for the second state data point, the weaker the temperature fluctuation on the right; the greater the possibility that the first state data point and the second state data point constitute a disconnected state interval, thus forming an inverse proportional relationship, which is quantified by an inverse proportional function; the greater the temperature fluctuation between the first state data point and the second state data point, it means that for the first state data point, the temperature fluctuation on the right is stronger; for the second state data point, the stronger the temperature fluctuation on the left; the greater the possibility that the first state data point and the second state data point constitute a disconnected state interval.

[0062] Preferably, in some implementations of the embodiments of the present invention, since the absolute value of the difference in temperature data between the first state data point and the second state data point and the state data point within the neighborhood range can represent its temperature volatility; then the method for obtaining the temperature volatility within the left neighborhood range of the first state data point and the temperature volatility within the right neighborhood range of the second state data point includes:

[0063] Preset a range parameter , wherein this embodiment is based on The present embodiment is not specifically limited. It depends on the specific implementation situation.

[0064] The left side of the first state data point state data points constitute a data point sequence as the left neighborhood range of the first state data point; for any state data point in the left neighborhood range of the first state data point, the absolute value of the difference between the temperature data of the state data point and the first state data point is recorded as the first difference absolute value of the state data point; the cumulative sum of the first difference absolute values ​​of all state data points in the left neighborhood range of the first state data point is taken as the temperature volatility in the left neighborhood range of the first state data point;

[0065] The right side of the second state data point state data points constitute a data point sequence as the right neighborhood range of the second state data point; for any state data point in the left neighborhood range of the second state data point, the absolute value of the difference between the temperature data of the state data point and the second state data point is recorded as the second difference absolute value of the state data point; the cumulative sum of the second difference absolute values ​​of all state data points in the right neighborhood range of the second state data point is taken as the temperature volatility in the right neighborhood range of the second state data point.

[0066] Preferably, in some implementations of the embodiments of the present invention, the method for acquiring the temperature fluctuation degree between the first state data point and the second state data point includes:

[0067] All state data points between the first state data point and the second state data point are fitted by the least square method to obtain a fitting curve; the integral result of the fitting curve is used as the temperature fluctuation degree between the first state data point and the second state data point.

[0068] Among them, the least square method is an existing technology and will not be described in detail in this embodiment.

[0069] At this point, the possibility of any two state data points constituting a disconnected state interval is obtained.

[0070] Specifically, the data interval formed by the two state data points that are most likely to form a disconnected state interval is used as the disconnected state interval.

[0071] So far, the disconnection state interval is obtained through the above method.

[0072] Step S003: According to the temperature data difference and pressure data difference between each state data point and the surrounding state data points, and whether it is in the disconnected state interval, obtain the possibility of each state data point belonging to normal data; cluster all state data points to obtain several feature groups; according to the possibility of belonging to normal data, obtain the operating state of each feature group; construct an optimized decision tree according to the operating state of the feature group.

[0073] It should be noted that since the state data points in the disconnected state interval are not necessarily normal data but may also be noise data, they cannot be generally assigned to the feature matrix in the information gain ID3 decision tree algorithm; therefore, it is necessary to judge the possibility that each state data point belongs to normal data when the circuit is disconnected, and then construct a feature matrix based on the normal data, so that the information entropy can be improved, and then the information gain ID3 decision tree algorithm can be improved, so that the constructed decision tree is more accurate.

[0074] It should be noted that for normal data when the circuit is disconnected, they are all within the disconnected state range, and the temperature fluctuations and pressure fluctuations are large, that is, the differences between the temperature data and the pressure data with the surrounding state data points are large, because the temperature trend and the pressure trend increase during arc formation and continuation; and the arc extinction decreases, so the correlation between the temperature data and the pressure data is high; therefore, the possibility of each state data point being normal data can be obtained based on the temperature data difference and pressure data difference between each state data point and the surrounding state data points, and whether it is in the disconnected state range.

[0075] Preferably, in some implementations of the embodiments of the present invention, the calculation method for obtaining the probability of each state data point belonging to normal data is:

[0076] Preset a range parameter , wherein this embodiment is based on The present embodiment is not specifically limited. It depends on the specific implementation situation.

[0077] For any state data point, state data points and the right The data point sequence formed by the state data points is used as the surrounding area sequence of the state data points.

[0078] According to whether the state data point is in a disconnected state interval, obtaining a disconnection characteristic value of the state data point;

[0079] Obtaining the temperature and pressure data volatility of the state data point according to the temperature data difference and the pressure data difference between the state data point and the state data points in the surrounding area sequence;

[0080] Obtaining the temperature and pressure data correlation of the state data point;

[0081] The product of the disconnection characteristic value, the temperature and pressure data fluctuation and the temperature and pressure data correlation is used as the possibility that the state data point belongs to normal data.

[0082] The specific formula is:

[0083]

[0084] In the formula, Indicates the possibility that the state data point belongs to normal data; Represents the disconnected characteristic value of the state data point; Indicates the volatility of temperature and pressure data of state data points; Represents the temperature-pressure data dependency of a state data point.

[0085] It should be noted that, since normal data are all within the disconnected state interval when the circuit is disconnected, the disconnection characteristic value can directly determine whether the state data point is normal; the stronger the volatility of the temperature and pressure data, the greater the difference in temperature data and pressure data between the state data point and the surrounding state data points, and the temperature trend and pressure trend increase during arc formation and continuation, so the state data point is more likely to be normal data; the greater the correlation between the temperature and pressure data, the higher the correlation between the temperature data and pressure data between the state data point and the surrounding state data points, and the greater the possibility that the state data point is normal data.

[0086] Preferably, in some implementations of the embodiments of the present invention, the method for obtaining the disconnection characteristic value of each state data point includes:

[0087] For any state data point, if the state data point is in the disconnected state interval, 1 is used as the disconnected characteristic value of the state data point; if the state data point is not in the disconnected state interval, 0 is used as the disconnected characteristic value of the state data point.

[0088] Preferably, in some implementations of the embodiments of the present invention, the method for obtaining the temperature and pressure data volatility of each state data point includes:

[0089] Record any state data point as a reference state data point, and for any state data point in a surrounding region sequence of the reference state data point, record the absolute value of the difference in temperature data between the state data point and the reference state data point as a first temperature difference of the state data point; record the cumulative sum of the first temperature differences of all state data points in the surrounding region sequence of the reference state data point as a first cumulative sum;

[0090] The absolute value of the difference between the pressure data of the state data point and the reference state data point is recorded as the first pressure difference of the state data point; the cumulative sum of the first pressure differences of all state data points in the surrounding area sequence of the reference state data point is recorded as the second cumulative sum;

[0091] The product of the first accumulated sum and the second accumulated sum is taken as the temperature and pressure data fluctuation of the reference state data point.

[0092] Preferably, in some implementations of the embodiments of the present invention, the method for obtaining the temperature and pressure data correlation of each state data point includes:

[0093] For any state data point, the data set consisting of the temperature data of all state data points in the surrounding area sequence of the state data point is recorded as the temperature data set; the data set consisting of the pressure data of all state data points in the surrounding area sequence of the state data point is recorded as the pressure data set; the Pearson correlation coefficient between the temperature data set and the pressure data set is taken as the correlation between the temperature and pressure data of the state data point.

[0094] Wherein, obtaining the Pearson correlation coefficient between two sets is a prior art, and this embodiment will not be described in detail here.

[0095] At this point, the possibility of each state data point belonging to normal data is obtained.

[0096] It should be noted that when constructing a decision tree using the information gain ID3 decision tree algorithm, all data points need to be divided into features, and it is necessary to try to select a number of feature groups that can divide all data points; grouping all data points helps to improve the generalization ability and robustness of the decision tree and avoid over-matching, so that the optimized decision tree can more accurately judge the operating status of the circuit breaker through the operating status data.

[0097] Preferably, in some implementations of the embodiments of the present invention, the specific method of dividing all state data points into a number of feature groups includes:

[0098] Preset a clustering parameter , wherein this embodiment is based on The present embodiment is not specifically limited. It depends on the specific implementation situation.

[0099] The K value is used as the clustering parameter The K-means clustering algorithm clusters all state data points in the three-dimensional coordinate system to obtain several clusters, and divides all state data points in each cluster into a feature group to obtain several feature groups.

[0100] Among them, the K-means clustering algorithm is an existing technology and will not be described in detail in this embodiment.

[0101] It should be noted that by obtaining the operating status of each feature group, a feature matrix in the information gain ID3 decision tree algorithm is constructed; the optimized decision tree is constructed through the feature matrix, so that the optimized decision tree can more accurately judge the operating status of the circuit breaker through the operating status data.

[0102] Preferably, in some implementations of the embodiments of the present invention, the calculation method for obtaining the operating status of each feature group is:

[0103] Preset a threshold parameter , wherein this embodiment is based on The present embodiment is not specifically limited. It depends on the specific implementation situation.

[0104] For any feature group, the mean value of the probability of all state data points in the feature group belonging to normal data is taken as the probability of the feature group belonging to the disconnected circuit; if the probability of the feature group belonging to the disconnected circuit is greater than or equal to the threshold parameter , the running state of the feature group is recorded as the disconnected state; if the probability of the feature group belonging to the disconnected circuit is less than the threshold parameter , and record the operating status of the feature group as other status.

[0105] It should be noted that, generally, when constructing a decision tree, the operating state is simply divided by thresholds, and then the probability value is obtained. However, the gains brought by the disconnected operating state may have different effects. The greater the possibility, the higher the degree of confusion should be, and the greater the information entropy should be. Therefore, they cannot be generally regarded as consistent. Therefore, the possibility of the disconnected operating state is combined with the probability, and the information entropy construction is improved according to its possibility when it belongs to the operating state, so that the greater the possibility, the greater the information entropy, making the construction of the decision tree more accurate.

[0106] Specifically, all feature groups and the operating status of each feature group are input into the information gain ID3 decision tree algorithm to construct a feature matrix, and an optimized decision tree is constructed through the feature matrix.

[0107] Among them, the information gain ID3 decision tree algorithm is a prior art and will not be described in detail in this embodiment.

[0108] At this point, the optimized decision tree is obtained through the above method.

[0109] Step S004: classify the operating status of the circuit breaker through the optimized decision tree.

[0110] Specifically, all state data points are classified through the optimized decision tree, and the operating state of the SF6 high-voltage circuit breaker is divided into a state when the SF6 circuit is disconnected and other states.

[0111] See also Figure 2 , which shows a characteristic relationship flow chart of circuit breaker operation status monitoring.

[0112] Through the above steps, a circuit breaker operation status monitoring is completed.

[0113] The present invention also proposes a circuit breaker operating status monitoring system, comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein when the processor executes the computer program, the steps of a circuit breaker operating status monitoring method in steps S001 to S004 are implemented.

[0114] 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 circuit breaker operating status monitoring method, characterized in that: The method comprises the following steps: Acquire a circuit breaker operation status training set; the circuit breaker operation status training set includes a plurality of status data points; the status data points refer to temperature data, pressure data and vibration data of the circuit breaker at each sampling time; According to the temperature data fluctuation of the state data points in the surrounding area of ​​any two state data points, the possibility of forming a disconnected state interval of any two state data points is obtained; and the disconnected state interval is obtained by screening according to the possibility of forming a disconnected state interval; The specific method of obtaining the possibility of forming a disconnected state interval of any two state data points according to the temperature data fluctuation of the state data points in the surrounding area of ​​any two state data points includes: For any two state data points in the circuit breaker operation state training set, one state data point is recorded as the first state data point; the other state data point is recorded as the second state data point; Obtain the temperature fluctuation within the left neighborhood of the first state data point and the temperature fluctuation within the right neighborhood of the second state data point; Obtaining a temperature fluctuation degree between a first state data point and a second state data point; The sum of the temperature fluctuation within the left neighborhood of the first state data point and the temperature fluctuation within the right neighborhood of the second state data point is recorded as the first sum value; the inverse proportional value of the first sum value and the product of the temperature fluctuation degree between the first state data point and the second state data point are taken as the possibility of the first state data point and the second state data point constituting a disconnected state interval; The specific method of obtaining the temperature fluctuation within the left neighborhood of the first state data point and the temperature fluctuation within the right neighborhood of the second state data point includes: Preset a range parameter , the left side of the first state data point state data points constitute a data point sequence as the left neighborhood range of the first state data point; for any state data point in the left neighborhood range of the first state data point, the absolute value of the difference between the temperature data of the state data point and the first state data point is recorded as the first difference absolute value of the state data point; the cumulative sum of the first difference absolute values ​​of all state data points in the left neighborhood range of the first state data point is taken as the temperature volatility in the left neighborhood range of the first state data point; The right side of the second state data point state data points constitute a data point sequence as the right neighborhood range of the second state data point; for any state data point in the left neighborhood range of the second state data point, the absolute value of the difference between the temperature data of the state data point and the second state data point is recorded as the second difference absolute value of the state data point; the cumulative sum of the second difference absolute values ​​of all state data points in the right neighborhood range of the second state data point is taken as the temperature volatility in the right neighborhood range of the second state data point; The specific method of obtaining the temperature fluctuation degree between the first state data point and the second state data point includes: Fitting all state data points between the first state data point and the second state data point by the least square method to obtain a fitting curve; taking the integral result of the fitting curve as the degree of temperature fluctuation between the first state data point and the second state data point; According to the temperature data difference and pressure data difference between each state data point and the surrounding state data points, and whether it is in the disconnected state interval, the possibility of each state data point being normal data is obtained; The specific method of obtaining the possibility of each state data point belonging to normal data according to the temperature data difference and pressure data difference between each state data point and the surrounding state data points, and whether it is in the disconnected state interval, includes: Preset a range parameter , for any state data point, the left side of the state data point state data points and the right state data points, and the data point sequence formed by the data points is used as the surrounding area sequence of the state data points; According to whether the state data point is in a disconnected state interval, obtaining a disconnection characteristic value of the state data point; Obtaining the temperature and pressure data volatility of the state data point according to the temperature data difference and the pressure data difference between the state data point and the state data points in the surrounding area sequence; Obtaining the temperature and pressure data correlation of the state data point; The product of the disconnection characteristic value, the temperature and pressure data fluctuation and the temperature and pressure data correlation is used as the possibility that the state data point belongs to normal data; The specific method of obtaining the disconnection characteristic value of the state data point includes: For any state data point, if the state data point is in the disconnected state interval, 1 is used as the disconnected characteristic value of the state data point; if the state data point is not in the disconnected state interval, 0 is used as the disconnected characteristic value of the state data point; The specific method of obtaining the temperature and pressure data volatility of the state data point includes: Record any state data point as a reference state data point, and for any state data point in a surrounding region sequence of the reference state data point, record the absolute value of the difference in temperature data between the state data point and the reference state data point as a first temperature difference of the state data point; record the cumulative sum of the first temperature differences of all state data points in the surrounding region sequence of the reference state data point as a first cumulative sum; The absolute value of the difference between the pressure data of the state data point and the reference state data point is recorded as the first pressure difference of the state data point; the cumulative sum of the first pressure differences of all state data points in the surrounding area sequence of the reference state data point is recorded as the second cumulative sum; The product of the first accumulated sum and the second accumulated sum is used as the temperature and pressure data volatility of the reference state data point; Cluster all state data points to obtain several feature groups; obtain the operating status of each feature group based on the possibility of belonging to normal data; The specific method of obtaining the operating status of each feature group according to the possibility of belonging to normal data is as follows: Preset a threshold parameter For any feature group, the mean value of the probability of all state data points in the feature group belonging to normal data is taken as the probability of the feature group belonging to the disconnected circuit; if the probability of the feature group belonging to the disconnected circuit is greater than or equal to the threshold parameter , the running state of the feature group is recorded as the disconnected state; if the probability of the feature group belonging to the disconnected circuit is less than the threshold parameter , recording the operating state of the feature group as other states; Construct an optimized decision tree based on the operating status of the feature group; The operating status of the circuit breaker is classified through the optimized decision tree.

2. A circuit breaker operating status monitoring method according to claim 1, characterized in that: The specific method of obtaining the temperature and pressure data correlation of the state data point includes: For any state data point, the data set consisting of the temperature data of all state data points in the surrounding area sequence of the state data point is recorded as the temperature data set; the data set consisting of the pressure data of all state data points in the surrounding area sequence of the state data point is recorded as the pressure data set; the Pearson correlation coefficient between the temperature data set and the pressure data set is taken as the correlation between the temperature and pressure data of the state data point.

3. A circuit breaker operating status monitoring system, comprising a memory, a processor, and a computer program stored in the memory and running on the processor, characterized in that: When the processor executes the computer program, the steps of the circuit breaker operating status monitoring method as claimed in claim 1 or 2 are implemented.

Images