A detection method for electrical automation power supply system
By analyzing the current data of the generator carbon brush and calculating the trend fluctuation, peak anomaly coefficient and equilibrium coefficient, the problem that traditional detection methods cannot effectively analyze the complex problems of carbon brushes is solved, which improves detection accuracy and reduces errors.
Patent Information
- Application Number
- CN202411081754.4
- Authority / Receiving Office
- CN · China
- Patent Type
- Patents(China)
- Current Assignee / Owner
- Filing Date
- 2024-08-08
- Publication Date
- 2024-12-13
- Estimated Expiration
- 2044-08-08
AI Technical Summary
Traditional power supply system detection methods cannot effectively analyze the complex problems encountered by carbon brushes during operation, resulting in large errors in the detection of the operating status of electrical automation power supply systems.
By obtaining the current data of each carbon brush in the generator, analyzing the trend fluctuation, peak abnormality coefficient and current equalization coefficient of the current subsequence, we can judge the fault condition of the carbon brush, and reduce detection errors.
It improves the accuracy of detection of the operating status of carbon brushes, reduces the detection error of the operating status of the electrical automation power supply system, and can more comprehensively reveal the complex problems of carbon brushes in actual operation.
Smart Images
Figure CN118606874B_ABST
Abstract
Description
Technical Field
[0001] The present application relates to the technical field of electrical variable analysis, and in particular to a detection method for an electrical automation power supply system. Background Art
[0002] The power supply system is the process and equipment that generates, transmits, and distributes electrical energy from power stations to end users. The generator of the power supply system plays a vital role. The carbon brush of the generator, as a key component in the excitation system, is responsible for transmitting current to the rotor of the generator to maintain the normal operation of the generator and a stable magnetic field. However, problems such as current imbalance may occur during operation. In order to understand the working status of the carbon brush of the generator in real time, it is necessary to detect the current of the carbon brush.
[0003] The traditional detection method reflects the operating status of the power supply system by collecting the current data of the carbon brushes and directly comparing the current differences. However, during the actual operation, the carbon brushes may become stuck, worn short, or have uneven spring pressure. Directly using the current data difference to judge the status of the power supply system fails to analyze the local current fluctuations and trend changes caused by the changes in the carbon brushes, and cannot fully reveal the complex problems encountered by the carbon brushes in actual operation, resulting in large errors in the detection of the operating status of the electrical automation power supply system. Summary of the invention
[0004] In view of the above, it is necessary to provide an electrical automation power supply system detection method, which reduces the detection error of the operating status of the electrical automation power supply system compared with the traditional power supply system detection method.
[0005] The present invention discloses an electrical automation power supply system detection method using the following technical solutions:
[0006] An embodiment of the present application provides a method for detecting an electrical automation power supply system, the method comprising the following steps:
[0007] Obtain the current data of each carbon brush in the generator of the electrical automation power supply system at each collection moment;
[0008] Based on the current data of each carbon brush at all acquisition moments, each current subsequence of each carbon brush is obtained; combining the change trend and random fluctuation degree of the current data in each current subsequence, the carbon brush current trend fluctuation amount of each current subsequence is determined;
[0009] Acquire the peak value in each current subsequence, and determine the current peak abnormality coefficient of each current subsequence by combining the abnormal degree of the peak value in each current subsequence and the trend fluctuation amount of the carbon brush current;
[0010] Determine the current balancing coefficient between any two carbon brushes based on the similarity between the current data of any two carbon brushes and the corresponding current peak abnormality coefficients;
[0011] Based on the current balancing coefficient between any two carbon brushes, the fault condition of the carbon brushes in the generator is determined.
[0012] In one embodiment, the process of determining the current subsequence is as follows:
[0013] The current data of each carbon brush is formed into a current data sequence in time sequence;
[0014] For each current data sequence, a mutation point detection algorithm is used to obtain the mutation point in the current data sequence, and the current data sequence is divided into current subsequences according to the mutation point.
[0015] In one embodiment, the process of determining the carbon brush current trend fluctuation amount is as follows:
[0016] The time series decomposition algorithm is used to obtain the trend series and residual series of each current subsequence;
[0017] Determining a relative trend significance coefficient of each current subsequence of each carbon brush based on the distance between trend sequences of different current subsequences of each carbon brush;
[0018] Calculate the absolute value of the autocorrelation of the residual sequence of each current subsequence;
[0019] The carbon brush current trend fluctuation amount of each current subsequence is determined by combining the absolute value and relative trend significance coefficient corresponding to each current subsequence.
[0020] In one embodiment, the process of determining the relative trend significance coefficient is:
[0021] For each carbon brush, the average value of the distance between the trend sequence of each current subsequence of the carbon brush and the trend sequences of all other current subsequences is calculated as the relative trend significance coefficient of each current subsequence of the carbon brush.
[0022] In one embodiment, the brush current trend fluctuation amount of each current subsequence is a forward fusion result of the absolute value and relative trend significance coefficient corresponding to each current subsequence.
[0023] In one embodiment, the process of determining the current peak abnormality coefficient is as follows:
[0024] For each peak in each current subsequence, determining a prominence of the peak based on current data in a local neighborhood of the peak;
[0025] Calculate the discrete degree of prominence of all peaks in each current subsequence;
[0026] The current peak abnormal coefficient of each current subsequence is a forward fusion result of the discrete degree of each current subsequence and the carbon brush current trend fluctuation amount.
[0027] In one embodiment, the prominence of the peak value is the range of a preset number of current data before and after the peak value.
[0028] In one embodiment, the current balancing coefficient is determined as follows:
[0029] Calculate the difference between the current data means in the current data sequences of any two carbon brushes, and record it as the current mean difference;
[0030] The set of current peak abnormal coefficients of all current subsequences of each carbon brush is recorded as an abnormal coefficient set;
[0031] Calculating the similarity between the abnormal coefficient sets of any two carbon brushes;
[0032] The current balancing coefficient between the arbitrary two carbon brushes is determined by combining the similarity between the arbitrary two carbon brushes and the current mean value difference.
[0033] In one embodiment, the process of determining the current balancing coefficient is: forward fusion of the inverse proportional mapping result of the current mean value difference and the similarity, and using the forward fusion result as the current balancing coefficient between any two carbon brushes.
[0034] In one embodiment, the process of determining the fault condition of the carbon brush in the generator is as follows:
[0035] The minimum and maximum values of the current balancing coefficients between any two carbon brushes are counted. When the minimum value is less than the product of the maximum value and a preset ratio, it is determined that some of the carbon brushes in the generator are faulty. Otherwise, it is determined that the carbon brushes in the generator are operating normally.
[0036] This application has at least the following beneficial effects:
[0037] This application collects the current data of all carbon brushes in the generator of the electrical automation power supply system, deeply analyzes the trend, random fluctuation degree and abnormal peak characteristics of the current data of each carbon brush, determines the current peak abnormal coefficient of each current subsequence, and reflects the operating status of the carbon brush over a period of time;
[0038] Based on the similarity between the current data of any two carbon brushes and the similarity between the corresponding current peak abnormal coefficients, the current balancing coefficient between the any two carbon brushes is determined to judge the uniformity of the current distribution between different carbon brushes of the generator;
[0039] Furthermore, based on the uniform current distribution of all carbon brushes, carbon brush fault detection is performed, taking into account the impact of fluctuations, long-term changes and abnormal points in current data on the normal operation of the generator; by analyzing the local trends and peak characteristics between the current data of the carbon brushes, the accuracy of detecting the operating status of the carbon brushes during the operation of the electrical automation power supply system is improved, thereby reducing the detection error of the operating status of the electrical automation power supply system. BRIEF DESCRIPTION OF THE DRAWINGS
[0040] In order to more clearly illustrate the technical solutions and advantages in the embodiments of the present application or the prior art, the drawings required for use in the embodiments or the prior art descriptions are briefly introduced below. Obviously, the drawings described below are only some embodiments of the present application. For ordinary technicians in this field, other drawings can be obtained based on these drawings without creative work.
[0041] Figure 1 A flowchart of the steps of an electrical automation power supply system detection method provided in this application;
[0042] Figure 2 It is a schematic diagram of the process of determining the trend fluctuation amount of the carbon brush current;
[0043] Figure 3 It is a schematic diagram of the process of determining the current peak abnormal coefficient;
[0044] Figure 4 FIG. 1 is a schematic diagram of the process of obtaining the current balancing coefficient. DETAILED DESCRIPTION
[0045] In the description of the embodiments of the present application, words such as "exemplary", "or", "for example" and the like are used to indicate examples, illustrations or descriptions. Any embodiment or design described as "exemplary" or "for example" in the embodiments of the present application should not be interpreted as being more preferred or more advantageous than other embodiments or designs. Specifically, the use of words such as "exemplary", "or", "for example" and the like is intended to present related concepts in a concrete manner.
[0046] Unless otherwise defined, all technical and scientific terms used herein have the same meaning as those commonly understood by those skilled in the art in the present application. The terms used in the present application are only for the purpose of describing specific embodiments and are not intended to limit the present application. It should be understood that, unless otherwise specified, " / " means or.
[0047] It should also be noted that the terms "first" and "second" in this application are used to distinguish similar objects, rather than to describe a specific order or sequence. The method disclosed in the embodiments of this application or the method shown in the flow chart includes one or more steps for implementing the method. Without departing from the scope of the claims, the execution order of multiple steps can be interchanged with each other, and some steps can also be deleted.
[0048] The following is a detailed description of a specific method for detecting an electrical automation power supply system provided by the present application in conjunction with the accompanying drawings.
[0049] An embodiment of the present application provides an electrical automation power supply system detection method. Specifically, the following electrical automation power supply system detection method is provided. Figure 1 , the method comprises the following steps:
[0050] Step 1: Obtain current data of each carbon brush in a generator of an electrical automation power supply system at each collection moment.
[0051] The electrical automation power supply system needs to convert other forms of energy into electrical energy through the generator and then transmit it. The operating status of the carbon brush in the generator directly affects the current transmission efficiency and the safe and stable operation of the generator. In order to obtain the current data of the generator carbon brush, a current sensor is installed on the carbon brush to collect the current data of each carbon brush at each collection time.
[0052] In one embodiment of the present application, during the data collection process, the time interval is 0.01s, and the amount of collected data is 8000. The values of the time interval and the amount of collected data are preset manually, and the implementer can determine them according to the actual scene characteristics. This application does not impose any special restrictions.
[0053] In order to reduce the impact of instrument or environmental interference on subsequent analysis during the acquisition process, the filtering method is used to reduce the noise of the current data.
[0054] In one embodiment of the present application, a Gaussian filtering method is used to perform noise reduction processing on current data. As other implementation methods, on the basis of being able to perform noise reduction processing on current data, the implementer may use other existing technologies to perform noise reduction processing on current data, and the present application does not impose any special restrictions.
[0055] Step 2: based on the current data of each carbon brush at all acquisition moments, obtain each current subsequence of each carbon brush; and determine the carbon brush current trend fluctuation amount of each current subsequence in combination with the change trend and random fluctuation degree of the current data in each current subsequence.
[0056] Carbon brushes are a consumable component in the generator that needs to be constantly replaced and are an important part of the generator excitation system. As the generator runs longer, the carbon brushes will become stuck, worn short, or have uneven spring pressure, which will increase the contact resistance of the carbon brushes during operation, reduce the current flowing through the carbon brushes, and divert the excitation current to the carbon brushes with smaller resistance, which can easily cause their temperature to rise, and then cause the generator's power supply capacity to decrease and fail to provide normal power output. Therefore, if the carbon brush current data shows a trend of change and irregular fluctuations, it means that the carbon brushes may be excessively worn at this time.
[0057] The current data of each carbon brush is organized into a current data sequence in chronological order.
[0058] In order to analyze the changing trend characteristics of current data, this application uses a time series decomposition algorithm to decompose each current data sequence. In order to better highlight the trend analysis, for each current data sequence, a mutation point detection algorithm is used to obtain the mutation point in the current data sequence, and the current data sequence is divided into each current subsequence according to the mutation point. The input of the time series decomposition algorithm is each current subsequence, and the output is the trend sequence and residual sequence of each current subsequence.
[0059] In one embodiment of the present application, the Pettitt mutation detection method is used to obtain the mutation points in the current data sequence. As other implementation methods, on the basis of being able to obtain the mutation points in the current data sequence, the implementer may use other existing technologies to obtain the mutation points in the current data sequence, such as the Bayesian mutation point detection algorithm, the Mann-Kendall mutation point detection algorithm, etc., and the present application does not make special restrictions; the current subsequence is decomposed by using the local weighted regression (Seasonal and Trend decomposition using Loess, STL) time series decomposition method. As other implementation methods, on the basis of being able to decompose the current subsequence, the implementer may use other existing technologies to decompose the current subsequence, such as the moving average method, the state space model, etc., and the present application does not make special restrictions.
[0060] The trend component indicates the long-term growth or decline trend of the carbon brush current data over time, while the residual component usually contains the random fluctuation of the carbon brush current data. If the rising or falling characteristics of the trend component are more obvious and the autocorrelation of the residual component is stronger, it indicates that the trend change degree and random fluctuation degree of the current data in the current subsequence are greater.
[0061] The determination process of the brush current trend fluctuation amount of each current subsequence is as follows:
[0062] For each carbon brush, the average value of the distance between the trend sequence of each current subsequence of the carbon brush and the trend sequences of all other current subsequences is calculated as the relative trend significance coefficient of each current subsequence of the carbon brush;
[0063] Calculate the absolute value of the autocorrelation of the residual sequence of each current subsequence;
[0064] The brush current trend fluctuation amount of each current subsequence is a forward fusion result of the absolute value of each current subsequence and the relative trend significance coefficient.
[0065] In one embodiment of the present application, the distance may be the DTW distance. As other implementations, on the basis of the measurable distance between the two sequences, the implementer may adopt other existing technologies to obtain the distance between the two sequences, such as the Euclidean distance, the Manhattan distance, etc., and the present application does not impose any special restrictions. The autocorrelation may be an autocorrelation function. As other implementations, on the basis of the calculable autocorrelation of the residual sequence, the implementer may adopt other existing technologies to obtain the autocorrelation of the residual sequence, such as the partial autocorrelation function, and the present application does not impose any special restrictions.
[0066] It is important to understand that positive fusion refers to combining independent variables in a way that enhances the overall effect, such as multiplication or addition, and implementers can define this based on actual conditions.
[0067] In one embodiment of the present application, the brush current trend fluctuation amount of each current subsequence is the product of the absolute value of each current subsequence and the relative trend significance coefficient.
[0068] In another embodiment of the present application, the brush current trend fluctuation amount of each current subsequence is the sum of the absolute value of each current subsequence and the relative trend significance coefficient.
[0069] It should be noted that: the greater the distance between the trend sequence of each current subsequence and the trend sequences of the remaining current subsequences, the greater the relative trend change difference between the different current subsequences of the carbon brush; the greater the relative trend change difference, the greater the degree of increase or decrease of the current data in each current subsequence. Normally, the data in the residual sequence of the current subsequence are independent and randomly distributed, and the absolute value of the autocorrelation is small. If the absolute value is larger, it means that the degree of fluctuation of the current data in the current subsequence is greater. The carbon brush current trend fluctuation amount of the current subsequence reflects the stable characteristics of the current data. The larger the value of the carbon brush current trend fluctuation amount, the more unstable the current is, and the more serious the wear of the carbon brushes in the generator may be. The schematic diagram of the process of determining the carbon brush current trend fluctuation amount is shown in the figure. Figure 2 shown.
[0070] Step 3, obtaining the peak value in each current subsequence, combining the abnormal degree of the peak value in each current subsequence and the carbon brush current trend fluctuation amount, to determine the current peak abnormal coefficient of each current subsequence.
[0071] The process of determining the current peak anomaly coefficient of each current subsequence is as follows:
[0072] As the carbon brushes continue to wear, the contact between the carbon brushes and the slip rings may no longer be smooth, causing the carbon brushes to more easily jump during the operation of the generator. At the same time, the worn carbon brushes will produce carbon powder. If the carbon powder is not removed in time, a large amount of carbon powder will accumulate in the brush holder, causing the carbon brushes to get stuck, further affecting the contact between the carbon brushes and the slip rings, resulting in unstable current transmission, and then causing abnormal peaks in the carbon brush current. Since the collected current data is sinusoidal waveform data with periodicity, if the waveform of the current data remains normal, the peak value will also have a certain regularity.
[0073] The present application adopts a peak extraction method to obtain the peak value in each current subsequence.
[0074] In one embodiment of the present application, an automatic multi-scale peak search algorithm is used to obtain the peak value in each current subsequence. As other implementation methods, on the basis of being able to obtain the peak value in each current subsequence, the implementer may use other existing technologies to obtain the peak value in each current subsequence, and the present application does not impose any special restrictions.
[0075] For each peak value in each current subsequence, the range of a preset number of current data before and after the peak value is used as the prominence degree of each peak value.
[0076] In an embodiment of the present application, the value of the preset number is 10. The value of the preset number is preset manually and can be set by the implementer. This application does not impose any special restrictions.
[0077] When the generator carbon brush is working normally, the prominence will fluctuate around a specific value. If the prominence becomes abnormally smaller or larger, it means that an abnormal peak may appear in the current data. And the change in the carbon brush current trend may further increase the abnormality of the peak. Therefore, the current peak abnormality coefficient of each current subsequence is determined.
[0078] Calculate the discrete degree of prominence of all peaks in each current subsequence;
[0079] The current peak abnormal coefficient of each current subsequence is a forward fusion result of the discrete degree of each current subsequence and the carbon brush current trend fluctuation amount.
[0080] It should be understood that the degree of dispersion can be a statistic such as variance, standard deviation or coefficient of variation that can characterize the degree of uneven distribution of data.
[0081] In one embodiment of the present application, the current peak anomaly coefficient of each current subsequence is the product of the discrete degree of each current subsequence and the carbon brush current trend fluctuation amount; wherein the discrete degree is the standard deviation.
[0082] In another embodiment of the present application, the discrete degree of each current sub-sequence and the fluctuation amount of the carbon brush current trend are respectively taken as the exponents of an exponential function with a natural constant as the base, denoted as a discrete exponential function and a fluctuation index function, and the sum of the calculation results of the discrete exponential function and the fluctuation index function of each current sub-sequence is taken as the current peak anomaly coefficient of each current sub-sequence; wherein the discrete degree is the variance.
[0083] It should be noted that: the greater the degree of discreteness, the greater the difference in the peak shape of the current data in the current subsequence. The larger the value of the carbon brush current trend fluctuation, the more unstable the current is and the more likely it is to have an abnormal peak. The larger the value of the current peak abnormality coefficient, the higher the abnormality of the current data peak and the worse the contact between the generator carbon brush and the slip ring. The flow chart for determining the current peak abnormality coefficient is as follows: Figure 3 shown.
[0084] Step 4: Determine the current balancing coefficient between any two carbon brushes based on the similarity between the current data of any two carbon brushes and the corresponding current peak abnormality coefficients.
[0085] The current peak abnormality coefficient can indicate the operating status of the carbon brush. The smaller the value, the better the operating status of the carbon brush. There are multiple carbon brushes running simultaneously in the generator. Under normal circumstances, the carbon brush has good contact with the contact surface of the slip ring, the operating conditions of each carbon brush are similar, the current of each carbon brush is close, and the carbon brush current is relatively balanced. Therefore, the peak abnormality coefficient of the carbon brush is generally low, and the peak abnormality coefficients between different carbon brushes are relatively close. Based on this feature, the current balance coefficient between any two carbon brushes in the generator is calculated to determine whether the current distribution is uniform.
[0086] The process of determining the current balancing coefficient between any two carbon brushes is as follows:
[0087] Calculate the difference between the current data means in the current data sequences of any two carbon brushes, and record it as the current mean difference;
[0088] The set of current peak abnormal coefficients of all current subsequences of each carbon brush is recorded as an abnormal coefficient set;
[0089] Calculating the similarity between the abnormal coefficient sets of any two carbon brushes;
[0090] The inverse proportional mapping result of the current mean value difference between the arbitrary two carbon brushes is forward fused with the similarity, and the forward fusion result is used as the current balancing coefficient between the arbitrary two carbon brushes.
[0091] It needs to be understood that: difference refers to the difference between the two, such as the absolute value relationship of the difference, the ratio relationship, etc., which can be limited by the implementer according to the actual situation; inverse proportional mapping means that the dependent variable and the independent variable show opposite change trends, such as the division relationship, the subtraction relationship, etc., which can be limited by the implementer according to the actual situation.
[0092] In one embodiment of the present application, the difference is the absolute value of the difference between the two; the similarity is the Jaccard similarity coefficient between the two sets. As other implementation methods, on the basis of the measurable similarity between the two sets, the implementer may adopt other existing technologies to obtain the similarity between the two sets, such as the size of the intersection between the sets, and the present application does not impose any special restrictions.
[0093] In one embodiment of the present application, the ratio of the similarity between any two carbon brushes to the current mean difference is used as the current balancing coefficient between any two carbon brushes.
[0094] In another embodiment of the present application, the difference between the similarity between any two carbon brushes and the current mean difference is taken as the exponent of an exponential function with a natural constant as the base, denoted as a similarity exponential function, and the calculation result of the similarity exponential function is taken as the current balancing coefficient between any two carbon brushes.
[0095] It should be noted that the smaller the difference between the current data means in the current data sequence of any two carbon brushes, the higher the consistency of the current flowing through the two carbon brushes. The greater the similarity between the abnormal coefficient sets of any two carbon brushes, the closer the operating states of the two carbon brushes. The larger the value of the current balancing coefficient, the more uniform the distribution of the currents of the two carbon brushes. The schematic diagram of the process of obtaining the current balancing coefficient is as follows: Figure 4 shown.
[0096] Step 5: Based on the current balancing coefficient between any two carbon brushes, determine the fault condition of the carbon brushes in the generator.
[0097] The current balance coefficient can reflect the current distribution between the two carbon brushes. If a carbon brush fails and the current is too low, the value of the current balance coefficient will become smaller. On the contrary, if the carbon brush operates normally, the value of the current balance coefficient will be larger.
[0098] The minimum and maximum values of the current balancing coefficient between any two carbon brushes are counted. When the minimum value is less than the product of the maximum value and the preset ratio, it is determined that some carbon brushes in the generator are faulty, which in turn causes the power supply capacity of the power supply system to be reduced and makes it difficult to provide stable power output; when the minimum value is greater than or equal to the product of the maximum value and the preset ratio, it is determined that each carbon brush in the generator is operating normally.
[0099] In an embodiment of the present application, the value of the preset ratio is one-half. The value of the preset ratio is preset manually and can be set by the implementer. The present application does not impose any special restrictions.
[0100] In summary, this application collects the current data of all carbon brushes in the generator of the electrical automation power supply system, deeply analyzes the trend, random fluctuation degree and abnormal peak characteristics of the current data of each carbon brush, determines the current peak abnormal coefficient of each current subsequence, and reflects the operating status of the carbon brush over a period of time;
[0101] Further, based on the similarity between the current data of any two carbon brushes and the similarity between the corresponding current peak abnormal coefficients, the current balancing coefficient between the any two carbon brushes is determined to judge the uniformity of the current distribution between different carbon brushes of the generator;
[0102] Furthermore, based on the uniform current distribution of all carbon brushes, carbon brush fault detection is performed, taking into account the impact of fluctuations, long-term changes and abnormal points in current data on the normal operation of the generator; by analyzing the local trends and peak characteristics between the current data of the carbon brushes, the accuracy of detecting the operating status of the carbon brushes during the operation of the electrical automation power supply system is improved, thereby reducing the detection error of the operating status of the electrical automation power supply system.
[0103] The flowchart and block diagram in the accompanying drawings show the possible architecture, function and operation of the system, method and computer program product according to the embodiment of the present disclosure. In this regard, each box in the flowchart or block diagram can represent a module, a program segment or a part of the code, and the module, the program segment or a part of the code contains one or more executable instructions for realizing the specified logical function. In some alternative implementations, the functions marked in the box can also occur in a different order from the order marked in the accompanying drawings. For example, two consecutive boxes can actually be executed substantially in parallel, and they can sometimes be executed in the opposite order, which can depend on the functions involved. In the description corresponding to the flowchart and the block diagram in the accompanying drawings, the operations or steps corresponding to different boxes can also occur in a different order from the order disclosed in the description, and sometimes there is no specific order between different operations or steps. For example, two consecutive operations or steps can actually be executed substantially in parallel, and they can sometimes be executed in the opposite order, which can depend on the functions involved. Each block in the block diagrams and / or flowcharts, and combinations of blocks in the block diagrams and / or flowcharts, may be implemented by a dedicated hardware-based system that performs the specified functions or actions, or may be implemented by a combination of dedicated hardware and computer instructions.
[0104] It is obvious to those skilled in the art that the present application is not limited to the details of the above exemplary embodiments, and that the present application can be implemented in other specific forms without departing from the basic features of the present application. Therefore, no matter from which point of view, the above embodiments of the present application should be regarded as exemplary and non-restrictive.
Claims
1. A method for detecting an electrical automation power supply system, characterized in that: The method comprises the following steps: Obtain the current data of each carbon brush in the generator of the electrical automation power supply system at each collection moment; Based on the current data of each carbon brush at all acquisition moments, each current subsequence of each carbon brush is obtained; combining the change trend and random fluctuation degree of the current data in each current subsequence, the carbon brush current trend fluctuation amount of each current subsequence is determined; Acquire the peak value in each current subsequence, and determine the current peak abnormality coefficient of each current subsequence by combining the abnormal degree of the peak value in each current subsequence and the trend fluctuation amount of the carbon brush current; Determine the current balancing coefficient between any two carbon brushes based on the similarity between the current data of any two carbon brushes and the corresponding current peak abnormality coefficients; Based on the current balancing coefficient between any two carbon brushes, the fault condition of the carbon brushes in the generator is determined; The determination process of the carbon brush current trend fluctuation amount is: The time series decomposition algorithm is used to obtain the trend series and residual series of each current subsequence; Determining a relative trend significance coefficient of each current subsequence of each carbon brush based on the distance between trend sequences of different current subsequences of each carbon brush; Calculate the absolute value of the autocorrelation of the residual sequence of each current subsequence; Determine the brush current trend fluctuation amount of each current subsequence by combining the absolute value and relative trend significance coefficient corresponding to each current subsequence; The determination process of the relative trend significance coefficient is: For each carbon brush, the average value of the distance between the trend sequence of each current subsequence of the carbon brush and the trend sequences of all other current subsequences is calculated as the relative trend significance coefficient of each current subsequence of the carbon brush; The process of determining the current peak abnormal coefficient is as follows: For each peak in each current subsequence, determining a prominence of the peak based on current data in a local neighborhood of the peak; Calculate the discrete degree of prominence of all peaks in each current subsequence; The current peak anomaly coefficient of each current subsequence is a forward fusion result of the discrete degree of each current subsequence and the carbon brush current trend fluctuation amount; The prominence of the peak value is the range of a preset number of current data before and after the peak value; The process of determining the current balancing coefficient is as follows: Calculate the difference between the current data means in the current data sequences of any two carbon brushes, and record it as the current mean difference; The set of current peak abnormal coefficients of all current subsequences of each carbon brush is recorded as an abnormal coefficient set; Calculating the similarity between the abnormal coefficient sets of any two carbon brushes; Determine a current balancing coefficient between the arbitrary two carbon brushes by combining the similarity between the arbitrary two carbon brushes and the current mean value difference; The process of judging the fault condition of the carbon brush in the generator is as follows: The minimum and maximum values of the current balancing coefficients between any two carbon brushes are counted. When the minimum value is less than the product of the maximum value and a preset ratio, it is determined that some of the carbon brushes in the generator are faulty. Otherwise, it is determined that the carbon brushes in the generator are operating normally.
2. The electrical automation power supply system detection method according to claim 1, characterized in that: The process of determining the current subsequence is as follows: The current data of each carbon brush is formed into a current data sequence in time sequence; For each current data sequence, a mutation point detection algorithm is used to obtain the mutation point in the current data sequence, and the current data sequence is divided into current subsequences according to the mutation point.
3. The electrical automation power supply system detection method according to claim 1, characterized in that: The brush current trend fluctuation amount of each current subsequence is a forward fusion result of the absolute value and relative trend significance coefficient corresponding to each current subsequence.
4. The electrical automation power supply system detection method according to claim 1, characterized in that: The process of determining the current balancing coefficient is as follows: forward fusion is performed on the inverse proportional mapping result of the current mean value difference and the similarity, and the forward fusion result is used as the current balancing coefficient between any two carbon brushes.