A fault monitoring method, system and storage medium based on LED classroom lamp

Abstract

The present application provides a fault monitoring method, system and storage medium based on LED classroom lamps, which relate to the field of fault monitoring technology. The method comprises the following steps: obtaining an operating data matrix of an LED classroom lamp; extracting an operating current sequence from the operating data matrix, converting the operating current sequence into an operating current unbiased domain, and determining a current abnormality deviation characteristic of the LED classroom lamp under non-steady-state operation through the operating current unbiased domain; obtaining operating light intensity data of the LED classroom lamp under non-steady-state operation, performing non-steady-state feature extraction on the operating light intensity data, and obtaining a light intensity fault feature vector of the LED classroom lamp under non-steady-state operation; determining fault type information of the LED classroom lamp using the current abnormality deviation characteristic and the light intensity fault feature vector, and sending a fault report of the LED classroom lamp according to the fault type information. The present application can effectively extract the operating current characteristic and light intensity characteristic of the LED classroom lamp under non-steady-state operation, so as to improve the fault monitoring accuracy of the LED classroom lamp.

Description

Technical Field

[0001] The present application relates to the technical field of fault monitoring, and more specifically, to a fault monitoring method, system and storage medium based on LED classroom lamps. Background Art

[0002] Fault monitoring technology based on LED classroom lamps is an intelligent technology that collects lamp operation data (such as current, voltage, light intensity, etc.) in real time, combines data analysis and intelligent algorithms to achieve accurate detection and prediction of lamp operation status. The application of this technology can not only improve the operating reliability of lamps, but also reduce maintenance costs and improve the lighting quality of classroom environments.

[0003] However, under non-steady-state working conditions, and under the influence of power grid fluctuations, ambient light changes or lamp aging, the light intensity signal may show obvious nonlinear and non-stationary characteristics. Traditional monitoring and analysis methods are difficult to accurately describe its dynamic change process, and it is difficult to accurately determine the fault type of LED classroom lamps based on a single light intensity data. Therefore, how to effectively extract the operating current characteristics and light intensity characteristics of LED classroom lamps under non-steady-state operation to improve the fault monitoring accuracy of LED classroom lamps is a difficult problem faced by the industry. Summary of the invention

[0004] The present application provides a fault monitoring method, system and storage medium based on LED classroom lamps, which can effectively extract the operating current characteristics and light intensity characteristics of LED classroom lamps under non-steady-state operation to improve the fault monitoring accuracy of LED classroom lamps.

[0005] In a first aspect, the present application provides a fault monitoring method based on an LED classroom lamp, the monitoring method comprising the following steps:

[0006] Start the fault monitoring of LED classroom lights, monitor the operating status of LED classroom lights under non-steady-state operation, and then obtain the operating data matrix of LED classroom lights;

[0007] Extracting the operating current sequence of the LED classroom lamp under non-steady-state operation from the operating data matrix, converting the operating current sequence into an operating current unbiased domain, and determining the current abnormal variation deviation characteristics of the LED classroom lamp under non-steady-state operation through the operating current unbiased domain;

[0008] Acquire operating light intensity data of the LED classroom lamp under non-steady-state operation, perform non-steady-state feature extraction on the operating light intensity data, and obtain a light intensity fault feature vector of the LED classroom lamp under non-steady-state operation;

[0009] The fault type information of the LED classroom lamp is determined using the current abnormal variation deviation feature and the light intensity fault feature vector, and a fault report of the LED classroom lamp is generated based on the fault type information.

[0010] In this embodiment, the operating state of the LED classroom lamp in the non-steady-state operation is monitored by an embedded microcontroller.

[0011] In this embodiment, extracting the operating current sequence of the LED classroom lamp under non-steady-state operation in the operating data matrix specifically includes:

[0012] For each operation data vector in the operation data matrix, extract the operation current of the LED classroom lamp from the operation data vector, and then obtain the operation currents of the plurality of LED classroom lamps;

[0013] The operating current sequence of the LED classroom lamp under non-steady-state operation is constructed based on all the operating currents.

[0014] In this embodiment, converting the operating current sequence into the operating current unbiased domain specifically includes:

[0015] Determining the unbiased current according to the operating current sequence;

[0016] Converting each operating current in the operating current sequence into a corresponding unbiased operating current according to the unbiased current amount;

[0017] The unbiased operating current range is determined by all unbiased operating currents.

[0018] In this embodiment, determining the current abnormal deviation characteristics of the LED classroom lamp under non-steady-state operation through the operating current unbiased domain specifically includes:

[0019] Determining the current anomaly in the bias-free region of the operating current;

[0020] Determining the deviation degree of abnormality corresponding to each unbiased operating current in the unbiased domain of the operating current according to the current abnormality degree;

[0021] The current deviation characteristics of LED classroom lights under non-steady-state operation are determined by the deviation of the currents corresponding to each unbiased operating current.

[0022] In this embodiment, obtaining the operating light intensity data of the LED classroom lamp under non-steady-state operation specifically includes:

[0023] The operating light intensity value of the LED classroom lamp is extracted from each operating data vector of the operating data matrix of the LED classroom lamp, and then the data set composed of all the operating light intensity values ​​is combined as the operating light intensity data of the LED classroom lamp under non-steady-state operation.

[0024] In this embodiment, the non-steady-state feature extraction is performed on the operating light intensity data to obtain the light intensity fault feature vector of the LED classroom lamp under non-steady-state operation, which specifically includes:

[0025] Converting the operating light intensity data into the frequency domain to obtain operating light intensity frequency domain data;

[0026] Performing wavelet packet decomposition on the operating light intensity frequency domain data to obtain the wavelet packet power corresponding to each decomposition layer;

[0027] Determine the wavelet packet intensity entropy corresponding to each decomposition layer through the corresponding wavelet packet power;

[0028] The light intensity fault feature vector of LED classroom lamp under non-steady-state operation is constructed based on all wavelet packet light intensity entropies.

[0029] In this embodiment, the fault type information of the LED classroom lamp is determined by using the current abnormal variation deviation feature and the light intensity fault feature vector to input the current abnormal variation deviation feature and the light intensity fault feature vector into the confidence rule base expert system for fault diagnosis, thereby obtaining the fault type information of the LED classroom lamp.

[0030] In a second aspect, the present application provides a fault monitoring system based on an LED classroom lamp, which is used to perform a fault monitoring method based on an LED classroom lamp, and the monitoring system includes:

[0031] An operation monitoring module is used to start fault monitoring of LED classroom lights, monitor the operation status of LED classroom lights under non-steady-state operation, and then obtain an operation data matrix of LED classroom lights;

[0032] A current feature extraction module is used to extract the operating current sequence of the LED classroom lamp under non-steady-state operation from the operating data matrix, convert the operating current sequence into an operating current unbiased domain, and determine the current abnormal variation deviation feature of the LED classroom lamp under non-steady-state operation through the operating current unbiased domain;

[0033] A light intensity feature extraction module is used to obtain the operating light intensity data of the LED classroom lamp under non-steady-state operation, perform non-steady-state feature extraction on the operating light intensity data, and obtain a light intensity fault feature vector of the LED classroom lamp under non-steady-state operation;

[0034] A fault diagnosis module is used to determine the fault type information of the LED classroom lamp using the current abnormal variation deviation feature and the light intensity fault feature vector, and generate a fault report of the LED classroom lamp according to the fault type information.

[0035] In a third aspect, the present application provides a computer-readable storage medium, in which instructions or codes are stored. When the instructions or codes are run on a computer, the computer implements the above-mentioned fault monitoring method based on LED classroom lights when executed.

[0036] The technical solution provided by the embodiments disclosed in this application has the following beneficial effects:

[0037] By starting the fault monitoring of the LED classroom lamp, the operating state of the LED classroom lamp under non-steady-state operation is monitored, and then the operating data matrix of the LED classroom lamp is obtained; the operating current sequence of the LED classroom lamp under non-steady-state operation is extracted from the operating data matrix, the operating current sequence is converted into an operating current unbiased domain, and the current abnormal deviation characteristics of the LED classroom lamp under non-steady-state operation are determined through the operating current unbiased domain; the operating light intensity data of the LED classroom lamp under non-steady-state operation is obtained, and the non-steady-state characteristics of the operating light intensity data are extracted to obtain the light intensity fault feature vector of the LED classroom lamp under non-steady-state operation; the fault type information of the LED classroom lamp is determined by using the current abnormal deviation characteristics and the light intensity fault feature vector, and a fault report of the LED classroom lamp is generated according to the fault type information.

[0038] It can be seen that in the present application, firstly, by extracting the operating current sequence and converting it into the operating current unbiased domain, the deviation caused by the external environment or the initial conditions can be removed, so as to accurately capture the abnormal fluctuations in the operation of the LED classroom lamp, and the current abnormal variation deviation characteristics can reflect the relative deviation degree of the operating current, so that the current abnormal variation deviation characteristics can sensitively reflect the initial signs of LED classroom lamp failure; then, by extracting non-steady-state characteristics through the operating light intensity data, the abnormal fluctuations, sudden changes and long-term trend changes in the light intensity signal can be accurately captured, and by extracting the light intensity fault feature vector, the light intensity change characteristics of the LED lamp can be mapped to a high-dimensional feature space, providing comprehensive data support for machine learning models or expert systems, thereby improving the accuracy of fault monitoring; finally, the current abnormal variation deviation characteristics and the light intensity fault feature vector are used in combination for fault diagnosis, and a fault report is generated, so as to accurately detect and analyze the non-steady-state faults of LED classroom lamps.

[0039] In summary, the technical solution adopted in this application can effectively extract the operating current characteristics and light intensity characteristics of LED classroom lights under non-steady-state operation, so as to improve the fault monitoring accuracy of LED classroom lights. BRIEF DESCRIPTION OF THE DRAWINGS

[0040] Figure 1 is a flow chart of a fault monitoring method based on LED classroom lamps provided in the present application;

[0041] Figure 2 is an exemplary flow chart for determining the unbiased domain of operating current provided by the present application;

[0042] Figure 3 is an exemplary flow chart for determining the light intensity fault characteristic vector of an LED classroom lamp under non-steady-state operation according to the present application;

[0043] Figure 4 It is a module structure diagram of the fault monitoring system based on LED classroom lights provided in this application. DETAILED DESCRIPTION

[0044] The following will be combined with the drawings in the embodiments of the present application to clearly and completely describe the technical solutions in the embodiments of the present application. Obviously, the described embodiments are only part of the embodiments of the present application, not all of the embodiments. Based on the embodiments in the present application, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of this application.

[0045] The embodiment of the present application provides a fault monitoring method, system and storage medium based on LED classroom lamps, the core of which is to start fault monitoring of LED classroom lamps, monitor the operating state of LED classroom lamps under non-steady-state operation, and then obtain the operating data matrix of LED classroom lamps; extract the operating current sequence of LED classroom lamps under non-steady-state operation from the operating data matrix, convert the operating current sequence into an unbiased domain of operating current, and determine the current abnormal deviation characteristics of LED classroom lamps under non-steady-state operation through the unbiased domain of operating current; obtain the operating light intensity data of LED classroom lamps under non-steady-state operation, extract non-steady-state characteristics of the operating light intensity data, and obtain the light intensity fault feature vector of LED classroom lamps under non-steady-state operation; use the current abnormal deviation characteristics and the light intensity fault feature vector to determine the fault type information of LED classroom lamps, and generate a fault report of LED classroom lamps based on the fault type information. The above scheme can effectively extract the operating current characteristics and light intensity characteristics of LED classroom lamps under non-steady-state operation to improve the fault monitoring accuracy of LED classroom lamps.

[0046] Embodiment 1: In order to better understand the above technical solution, the above technical solution will be described in detail below in conjunction with the accompanying drawings and specific implementation methods. Figure 1 As shown in FIG. 1 , this figure is an exemplary flow chart of a fault monitoring method based on an LED classroom lamp according to this embodiment of the present application, and the monitoring method includes the following steps:

[0047] In step S1, fault monitoring of the LED classroom lamp is started to monitor the operating state of the LED classroom lamp under non-steady-state operation, thereby obtaining an operating data matrix of the LED classroom lamp.

[0048] In specific implementation, the operating status of the LED classroom lamp under non-steady-state operation can be monitored by an embedded microcontroller; the embedded microcontroller is set to collect current, voltage, light intensity, temperature and other data of the LED classroom lamp from each sensor at a certain frequency (for example, sampling 10 times per second), and an operating data vector is constructed through the data collected once. For example, the operating data vector includes the operating current, operating light intensity value, operating voltage and operating temperature of the LED classroom lamp. Multiple operating data vectors can be obtained in the above manner, and all the operating data vectors are constructed into a matrix in chronological order, and the obtained matrix can be used as the operating data matrix of the LED classroom lamp.

[0049] In step S2, an operating current sequence of the LED classroom lamp under non-steady-state operation is extracted from the operating data matrix, the operating current sequence is converted into an operating current unbiased domain, and the current abnormal deviation characteristics of the LED classroom lamp under non-steady-state operation are determined through the operating current unbiased domain.

[0050] In this embodiment, the operating current sequence of the LED classroom lamp under non-steady-state operation can be extracted from the operating data matrix in the following manner, namely:

[0051] For each operation data vector in the operation data matrix, extract the operation current of the LED classroom lamp from the operation data vector, and then obtain the operation currents of the plurality of LED classroom lamps;

[0052] The operating current sequence of the LED classroom lamp under non-steady-state operation is constructed based on all the operating currents.

[0053] In the specific implementation, first, for each operating data vector in the operating data matrix, the operating current of the LED classroom lamp can be extracted from the operating data vector by traversing. The operating currents of multiple LED classroom lamps can be extracted in the above manner, and each operating current has a time label; then, the operating current sequence of the LED classroom lamp under non-steady-state operation can be constructed based on all the operating currents, that is, all the operating currents are arranged in chronological order to obtain a time series, which can be used as the operating current sequence of the LED classroom lamp under non-steady-state operation.

[0054] Preferably, in this embodiment, reference Figure 2 As shown, this figure is an exemplary flow chart for determining the unbiased domain of the operating current in an embodiment of the present application. In this embodiment, converting the operating current sequence into the unbiased domain of the operating current can be specifically implemented by the following steps:

[0055] First, in step S21, the current unbiased amount is determined according to the operating current sequence;

[0056] Then, in step S22, each operating current in the operating current sequence is converted into a corresponding unbiased operating current according to the current unbiased amount;

[0057] Finally, in step S23, the unbiased operating current range is determined by using all the unbiased operating currents.

[0058] In specific implementation, first, the current unbiased value can be determined according to the operating current sequence, wherein the current unbiased value is used to indicate the overall value level of the operating current sequence without deviation. The mean of all operating currents in the operating current sequence can be used as the current unbiased value, or the mode of the operating currents in the operating current sequence can be used as the current unbiased value. Then, each operating current in the operating current sequence can be converted into a corresponding unbiased operating current according to the current unbiased value, wherein the unbiased operating current is the operating current after deviation correction. For each operating current in the operating current sequence, the operating current can be deviation-corrected according to the current unbiased value, thereby obtaining the corresponding unbiased operating current. In actual implementation, the unbiased operating current can be determined according to the following formula:

[0059] ;

[0060] in, Indicates the operating current The corresponding unbiased operating current, represents the i-th operating current in the operating current sequence, represents the jth operating current in the operating current sequence, m represents the total number of operating currents in the operating current sequence, It represents the unbiased amount of current. The unbiased operating current corresponding to each operating current in the operating current sequence can be obtained by the above method. Finally, the operating current unbiased domain can be determined by all the unbiased operating currents, wherein the operating current unbiased domain is the data domain composed of the operating current after bias correction, that is, the data composed of all the unbiased operating currents can be combined as the operating current unbiased domain.

[0061] In this embodiment, the current variation deviation characteristics of the LED classroom lamp under non-steady-state operation can be determined by the operating current unbiased domain in the following manner, namely:

[0062] Determining the current anomaly in the bias-free region of the operating current;

[0063] Determining the deviation degree of abnormality corresponding to each unbiased operating current in the unbiased domain of the operating current according to the current abnormality degree;

[0064] The current deviation characteristics of LED classroom lights under non-steady-state operation are determined by the deviation of the currents corresponding to each unbiased operating current.

[0065] In specific implementation, first, the current variability of the unbiased domain of the operating current can be determined, wherein the current variability is an indicator used to represent the relative variability of the unbiased operating current in the unbiased domain of the operating current, and the standard deviation of all unbiased operating currents in the unbiased domain of the operating current can be calculated, so that the ratio of the standard deviation to the minimum unbiased operating current in the unbiased domain of the operating current can be used as the current variability of the unbiased domain of the operating current; then, the variability deviation corresponding to each unbiased operating current in the unbiased domain of the operating current can be determined according to the current variability, wherein the variability deviation is used to represent the degree of deviation of the unbiased operating current from the domain center, and the mean of all unbiased operating currents in the unbiased domain of the operating current can be used as the domain center of the unbiased domain of the operating current. For each unbiased operating current in the unbiased domain of the operating current, the absolute value of the difference between the unbiased operating current and the domain center can be calculated, and then the product of the result and the current variability degree can be used as the variability deviation degree corresponding to the unbiased operating current. The variability deviation degree corresponding to each unbiased operating current in the unbiased domain of the operating current can be obtained in the above manner; finally, the current variability deviation characteristic of the LED classroom lamp under non-steady-state operation can be determined through the variability deviation degree corresponding to each unbiased operating current, wherein the current variability deviation characteristic is a characteristic used to represent the relative variability degree of the operating current of the LED classroom lamp under non-steady-state operation, and the standard deviation of the variability deviation degrees corresponding to all unbiased operating currents can be used as the current variability deviation characteristic of the LED classroom lamp under non-steady-state operation.

[0066] It should be noted that by extracting the operating current sequence and converting it into the operating current unbiased domain, the deviation caused by the external environment or initial conditions can be removed, so as to accurately capture the abnormal fluctuations in the operation of the LED classroom lamp, and the current abnormal deviation feature can reflect the relative deviation degree of the operating current, so that the current abnormal deviation feature can sensitively reflect the early signs of LED classroom lamp failure. For example, when the internal circuit of the LED lamp is abnormal, its current may fluctuate slightly or suddenly, and these changes can be sensitively captured by the current abnormal deviation feature.

[0067] In step S3, the operating light intensity data of the LED classroom lamp under non-steady-state operation is obtained, and non-steady-state feature extraction is performed on the operating light intensity data to obtain a light intensity fault feature vector of the LED classroom lamp under non-steady-state operation.

[0068] In this embodiment, the operating light intensity data of the LED classroom lamp under non-steady-state operation can be obtained in the following manner, namely:

[0069] The operating light intensity value of the LED classroom lamp is extracted from each operating data vector of the operating data matrix of the LED classroom lamp, and then the data set composed of all the operating light intensity values ​​is combined as the operating light intensity data of the LED classroom lamp under non-steady-state operation; in specific implementation, the operating light intensity value of the LED classroom lamp can be extracted from each operating data vector of the operating data matrix of the LED classroom lamp by traversing.

[0070] Preferably, in this embodiment, reference Figure 3 As shown in FIG. 1 , this figure is an exemplary flow chart of determining the light intensity fault feature vector of the LED classroom lamp under non-steady-state operation in an embodiment of the present application. In this embodiment, the non-steady-state feature extraction is performed on the operating light intensity data to obtain the light intensity fault feature vector of the LED classroom lamp under non-steady-state operation, which can be specifically implemented by the following steps:

[0071] First, in step S31, the operating light intensity data is converted into the frequency domain to obtain operating light intensity frequency domain data;

[0072] Then, in step S32, the operating light intensity frequency domain data is subjected to wavelet packet decomposition to obtain the wavelet packet power corresponding to each decomposition layer;

[0073] Secondly, in step S33, the wavelet packet intensity entropy corresponding to each decomposition layer is determined by the corresponding wavelet packet power;

[0074] Finally, in step S34, the light intensity fault feature vector of the LED classroom lamp under non-steady-state operation is constructed based on all wavelet packet light intensity entropies.

[0075] In the specific implementation, first, the operating light intensity data can be converted to the frequency domain through Fourier transform to obtain the operating light intensity frequency domain data. The frequency characteristics can be analyzed and potential fault modes can be identified through the operating light intensity frequency domain data; then, the operating light intensity frequency domain data can be decomposed by wavelet packet decomposition, wherein wavelet packet decomposition (WPD) is a method of multi-scale signal analysis, which can effectively capture the local characteristics of the signal. The operating light intensity frequency domain data is decomposed by wavelet packets to obtain detailed information of each frequency band, that is, the result of wavelet packet decomposition will obtain components within different frequency ranges, and each layer represents a different degree of decomposition of the signal. For example, the first layer represents low-frequency information, the second layer represents medium-frequency information, the third layer represents high-frequency information, etc. For each decomposition layer, the wavelet packet power can be calculated, which represents the energy of the corresponding frequency band, and the wavelet packet power is the sum of the squares of the signals of each decomposition layer.

[0076] In addition, in specific implementation, the wavelet packet intensity entropy corresponding to each decomposition layer can be determined by the corresponding wavelet packet power, wherein the wavelet packet intensity entropy is an indicator used to measure the complexity of the corresponding decomposition layer. In the fault state, the wavelet packet intensity entropy usually increases. By calculating the wavelet packet intensity entropy of each decomposition layer, the fault characteristics of the light intensity can be further analyzed. In actual implementation, the wavelet packet intensity entropy can be determined by the following formula:

[0077]

[0078] Among them, E represents the wavelet packet intensity entropy corresponding to the decomposition layer, represents the wavelet packet power corresponding to the decomposition layer, n represents the total number of discrete points in the frequency band in the decomposition layer, It represents the power corresponding to the discrete point of the i-th frequency band in the decomposition layer. The power corresponding to the discrete point of the frequency band can be calculated in the process of wavelet packet decomposition. The wavelet packet intensity entropy corresponding to each decomposition layer can be obtained in the above way. Finally, the light intensity fault feature vector of the LED classroom lamp under non-steady-state operation can be constructed according to all the wavelet packet intensity entropies, wherein the light intensity fault feature vector is a feature vector used to represent the light intensity change of the LED classroom lamp under non-steady-state operation, that is, the feature vector composed of all the wavelet packet intensity entropies can be used as the light intensity fault feature vector of the LED classroom lamp under non-steady-state operation.

[0079] It should be noted that by running light intensity data to extract non-steady-state features, it is possible to accurately capture abnormal fluctuations, sudden changes and long-term trend changes in the light intensity signal. By extracting the light intensity fault feature vector, the light intensity change characteristics of the LED lamp can be mapped to a high-dimensional feature space, providing comprehensive data support for machine learning models or expert systems. It can not only diagnose fault types in real time, but also improve the sensitivity, robustness and accuracy of fault monitoring.

[0080] In step S4, the fault type information of the LED classroom lamp is determined using the current abnormal variation deviation feature and the light intensity fault feature vector, and a fault report of the LED classroom lamp is generated according to the fault type information.

[0081] In this embodiment, the fault type information of the LED classroom lamp is determined by using the current abnormal variation deviation feature and the light intensity fault feature vector to input the current abnormal variation deviation feature and the light intensity fault feature vector into the confidence rule base expert system for fault diagnosis, thereby obtaining the fault type information of the LED classroom lamp.

[0082] In the specific implementation, first, the current abnormal deviation characteristics and light intensity fault feature vectors are normalized; then, the normalized current abnormal deviation characteristics and light intensity fault feature vectors are input into the confidence rule base expert system as input variables. The confidence rule base expert system establishes association rules between input variables and fault types based on historical experience and expert knowledge, and calculates the activation weight of each association rule according to the matching degree between the input variables and the rule conditions; finally, the confidence rule base expert system synthesizes the confidence distribution of the final fault type through Bayesian reasoning or weighted average method based on the output confidence of all association rules, and selects the fault type with the highest confidence as the final diagnosis result, that is, the fault type information of the LED classroom lamp is obtained.

[0083] In this embodiment, a fault report of the LED classroom lamp is generated according to the fault type information. In a specific implementation, a fault report including key diagnostic results is generated according to the fault type information. The fault report of the LED classroom lamp may include the following information:

[0084] Lamp number: unique identification of the faulty LED classroom lamp.

[0085] Lamp location: classroom number, specific location of the lamp.

[0086] Fault occurrence time: records the timestamp when the fault monitoring system detects the problem.

[0087] Fault type: Specific fault type, such as light attenuation, current abnormality, heat dissipation problem, etc.

[0088] Fault severity: Assess the severity of the fault (such as minor, severe, critical, etc.) based on the confidence level of the diagnostic model output.

[0089] Recommended actions: Provides repair suggestions for specific faults, such as replacing LED lamp beads, checking the drive circuit, cleaning the heat sink, etc.

[0090] It should be noted that the comprehensive use of current variation deviation characteristics and light intensity fault feature vectors for fault diagnosis and generation of fault reports can not only accurately detect and analyze the non-steady-state faults of LED classroom lights, but also optimize the entire maintenance process and improve the reliability and economy of the system. At the same time, structured fault reports provide important data support in operation, management and optimization design.

[0091] It can be seen that in the present application, firstly, by extracting the operating current sequence and converting it into the operating current unbiased domain, the deviation caused by the external environment or the initial conditions can be removed, so as to accurately capture the abnormal fluctuations in the operation of the LED classroom lamp, and the current abnormal variation deviation characteristics can reflect the relative deviation degree of the operating current, so that the current abnormal variation deviation characteristics can sensitively reflect the initial signs of LED classroom lamp failure; then, by extracting non-steady-state characteristics through the operating light intensity data, the abnormal fluctuations, sudden changes and long-term trend changes in the light intensity signal can be accurately captured, and by extracting the light intensity fault feature vector, the light intensity change characteristics of the LED lamp can be mapped to a high-dimensional feature space, providing comprehensive data support for machine learning models or expert systems, thereby improving the accuracy of fault monitoring; finally, the current abnormal variation deviation characteristics and the light intensity fault feature vector are used in combination for fault diagnosis, and a fault report is generated, so as to accurately detect and analyze the non-steady-state faults of LED classroom lamps.

[0092] In summary, the technical solution adopted in this application can effectively extract the operating current characteristics and light intensity characteristics of LED classroom lights under non-steady-state operation, so as to improve the fault monitoring accuracy of LED classroom lights.

[0093] Embodiment 2: This application provides a fault monitoring system based on LED classroom lamps, referring to Figure 4 As shown, this figure is a schematic diagram of a monitoring system according to this embodiment of the present application, and the monitoring system includes:

[0094] The operation monitoring module 100 is used to start the fault monitoring of the LED classroom lamp, monitor the operation state of the LED classroom lamp under the non-steady state operation, and then obtain the operation data matrix of the LED classroom lamp;

[0095] The current feature extraction module 200 is used to extract the operating current sequence of the LED classroom lamp under non-steady-state operation from the operating data matrix, convert the operating current sequence into an operating current unbiased domain, and determine the current abnormal variation deviation feature of the LED classroom lamp under non-steady-state operation through the operating current unbiased domain;

[0096] The light intensity feature extraction module 300 is used to obtain the operating light intensity data of the LED classroom lamp under non-steady-state operation, perform non-steady-state feature extraction on the operating light intensity data, and obtain the light intensity fault feature vector of the LED classroom lamp under non-steady-state operation;

[0097] The fault diagnosis module 400 is used to determine the fault type information of the LED classroom lamp using the current abnormal variation deviation feature and the light intensity fault feature vector, and generate a fault report of the LED classroom lamp according to the fault type information.

[0098] In the third embodiment, the present application also provides a computer-readable storage medium, in which instructions or codes are stored. When the instructions or codes are executed on a computer, the computer implements the above-mentioned fault monitoring method based on LED classroom lights.

[0099] The present application is described with reference to the flowcharts and / or block diagrams of the methods, devices (systems) and computer program products according to the embodiments of the present application. It should be understood that each process and / or box in the flowchart and / or block diagram, as well as the combination of the processes and / or boxes in the flowchart and / or block diagram, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, a special-purpose computer, an embedded processor or other programmable data processing device to generate a machine, so that the instructions executed by the processor of the computer or other programmable data processing device generate instructions for implementing the processes in the flowchart and / or block diagram. Figure 1 A process or multiple processes and / or boxes Figure 1 A device that provides the functions specified in a block or multiple blocks.

Claims

1. A fault monitoring method based on LED classroom lamps, characterized in that: The monitoring method comprises the following steps: Start the fault monitoring of LED classroom lights, monitor the operating status of LED classroom lights under non-steady-state operation, and then obtain the operating data matrix of LED classroom lights; Extracting the operating current sequence of the LED classroom lamp under non-steady-state operation from the operating data matrix, converting the operating current sequence into an operating current unbiased domain, and determining the current abnormal variation deviation characteristics of the LED classroom lamp under non-steady-state operation through the operating current unbiased domain; Acquire operating light intensity data of the LED classroom lamp under non-steady-state operation, perform non-steady-state feature extraction on the operating light intensity data, and obtain a light intensity fault feature vector of the LED classroom lamp under non-steady-state operation; The fault type information of the LED classroom lamp is determined by using the current abnormal variation deviation feature and the light intensity fault feature vector, and a fault report of the LED classroom lamp is generated according to the fault type information; specifically, the method includes: first, normalizing the current abnormal variation deviation feature and the light intensity fault feature vector; then, inputting the normalized current abnormal variation deviation feature and the light intensity fault feature vector as input variables into the confidence rule base expert system, and the confidence rule base expert system establishes association rules between the input variables and the fault types according to historical experience and expert knowledge, and calculates the activation weight of each association rule according to the matching degree between the input variables and the rule conditions; finally, the confidence rule base expert system synthesizes the confidence distribution of the final fault type according to the output confidence of all association rules through Bayesian reasoning or weighted average method, and selects the fault type with the highest confidence as the final diagnosis result, that is, the fault type information of the LED classroom lamp is obtained; In the above steps, determining the current abnormal deviation characteristics of the LED classroom lamp under non-steady-state operation through the operating current unbiased domain specifically includes: Determining the current anomaly in the bias-free region of the operating current; Determining the deviation degree of abnormality corresponding to each unbiased operating current in the unbiased domain of the operating current according to the current abnormality degree; The current deviation characteristics of LED classroom lights under non-steady-state operation are determined by the deviation of the currents corresponding to each unbiased operating current.

2. A fault monitoring method based on LED classroom lamps as claimed in claim 1, characterized in that: The operating status of LED classroom lights under non-steady-state operation is monitored by an embedded microcontroller.

3. A fault monitoring method based on LED classroom lamps as claimed in claim 1, characterized in that: Extracting the operating current sequence of the LED classroom lamp under non-steady-state operation in the operating data matrix specifically includes: For each operation data vector in the operation data matrix, extract the operation current of the LED classroom lamp from the operation data vector, and then obtain the operation currents of the plurality of LED classroom lamps; The operating current sequence of the LED classroom lamp under non-steady-state operation is constructed based on all the operating currents.

4. A fault monitoring method based on LED classroom lamps as claimed in claim 1, characterized in that: Converting the operating current sequence into the operating current unbiased domain specifically includes: Determining the unbiased current according to the operating current sequence; Converting each operating current in the operating current sequence into a corresponding unbiased operating current according to the unbiased current amount; The unbiased operating current range is determined by all unbiased operating currents.

5. A fault monitoring method based on LED classroom lamps as claimed in claim 1, characterized in that: Obtaining the operating light intensity data of LED classroom lights under non-steady-state operation specifically includes: The operating light intensity value of the LED classroom lamp is extracted from each operating data vector of the operating data matrix of the LED classroom lamp, and then the data set composed of all the operating light intensity values ​​is combined as the operating light intensity data of the LED classroom lamp under non-steady-state operation.

6. A fault monitoring method based on LED classroom lamps as claimed in claim 1, characterized in that: The non-steady-state feature extraction is performed on the operating light intensity data to obtain the light intensity fault feature vector of the LED classroom lamp under non-steady-state operation, which specifically includes: Converting the operating light intensity data into the frequency domain to obtain operating light intensity frequency domain data; Performing wavelet packet decomposition on the operating light intensity frequency domain data to obtain the wavelet packet power corresponding to each decomposition layer; Determine the wavelet packet intensity entropy corresponding to each decomposition layer through the corresponding wavelet packet power; The light intensity fault feature vector of LED classroom lamp under non-steady-state operation is constructed based on all wavelet packet light intensity entropies.

7. A fault monitoring system based on LED classroom lamps, used to execute a fault monitoring method based on LED classroom lamps as claimed in any one of claims 1 to 6, characterized in that: The monitoring system comprises: An operation monitoring module is used to start fault monitoring of LED classroom lights, monitor the operation status of LED classroom lights under non-steady-state operation, and then obtain an operation data matrix of LED classroom lights; A current feature extraction module is used to extract the operating current sequence of the LED classroom lamp under non-steady-state operation from the operating data matrix, convert the operating current sequence into an operating current unbiased domain, and determine the current abnormal variation deviation feature of the LED classroom lamp under non-steady-state operation through the operating current unbiased domain; A light intensity feature extraction module is used to obtain the operating light intensity data of the LED classroom lamp under non-steady-state operation, perform non-steady-state feature extraction on the operating light intensity data, and obtain a light intensity fault feature vector of the LED classroom lamp under non-steady-state operation; A fault diagnosis module is used to determine the fault type information of the LED classroom lamp using the current abnormal variation deviation feature and the light intensity fault feature vector, and generate a fault report of the LED classroom lamp according to the fault type information.

8. A computer-readable storage medium, characterized in that: The computer-readable storage medium stores instructions or codes, and when the instructions or codes are executed on a computer, the computer implements the fault monitoring method based on LED classroom lamps as described in any one of claims 1 to 6.

Images