Navigation light circuit fault detection method, device and electronic equipment

Abstract

This invention relates to the field of navigational lighting monitoring technology, and more particularly to a method, device, and electronic equipment for detecting faults in navigational lighting circuits. The method includes: collecting navigational lighting data from an airport; extracting and aligning the frequency bands of the navigational lighting data to generate harmonic vectors; constructing a harmonic feature extraction model based on the harmonic vectors to obtain multidimensional health features; performing health quantification and fault tracing based on feature clustering and the multidimensional health features; and providing dynamic early warning based on the multidimensional health features. This invention analyzes the harmonic family as a whole dynamic feature, and the five extracted feature categories all correspond to specific physical fault mechanisms, possessing strong interpretability. It eliminates the need to label fault samples and locates the source of anomalies through feature deviation combination patterns, achieving an upgrade from "whether there is a fault" to "where and what kind of fault."

Description

Technical Field

[0001] This invention relates to the field of navigation light monitoring technology, and in particular to a method, device and electronic equipment for detecting faults in navigation light circuits. Background Technology

[0002] Navigational lighting systems are critical facilities for ensuring safe takeoffs and landings of aircraft at night or in low-visibility conditions, and their reliability directly affects flight safety and airport operational efficiency. Currently, the navigational lighting systems widely used in airports have complex structures, including constant current dimmers, underground cable networks, isolation transformers, and numerous lamp arrays. Over long-term operation, they are susceptible to factors such as electrical aging, environmental corrosion, and loose connections, leading to various hidden faults in the lighting circuits. If these faults are not detected and located in a timely manner, they may cause abnormal light brightness, flickering, or even system failure, seriously affecting aviation safety. Therefore, developing a health status detection technology for navigational lighting circuits that can monitor, intelligently diagnose, and provide early warnings in real time has become an important issue that urgently needs to be addressed in the field of airport operations and maintenance.

[0003] Existing fault detection technologies for navigation lighting circuits primarily rely on periodic manual inspections, current and voltage alarms based on fixed thresholds, or simple waveform monitoring, which have significant limitations and defects. First, manual inspections are inefficient, lack comprehensive coverage, and struggle to detect latent or intermittent faults, resulting in severely delayed responses. Second, traditional threshold alarm methods typically monitor only macroscopic parameters such as fundamental current and voltage, making them insensitive to early-stage faults, leading to high false negative rates and an inability to differentiate fault types. Third, some methods employing harmonic analysis often focus only on single indicators such as total harmonic distortion (THD), failing to fully utilize the rich fault information inherent in harmonic intensity, phase, and their dynamic changes, resulting in insufficient diagnostic accuracy and difficulty in tracing the source. Furthermore, existing methods generally lack adaptive compensation capabilities to environmental factors, are prone to false alarms due to seasonal variations, and lack quantitative assessment and trend prediction functions for circuit health status, making it difficult to support intelligent preventative maintenance decisions. Summary of the Invention

[0004] The purpose of this invention is to provide a method, apparatus, and electronic device for detecting faults in navigation lighting circuits, so as to solve at least one of the problems existing in the prior art.

[0005] To achieve the above objectives, the present invention adopts the following technical solution:

[0006] A method for detecting faults in navigation light circuits includes:

[0007] Collect data on airport navigation lights;

[0008] Frequency band extraction and alignment are performed on the navigation light data to generate harmonic vectors;

[0009] A harmonic feature extraction model is constructed based on harmonic vectors to obtain multidimensional health features;

[0010] Health measurement and fault tracing are performed based on feature clustering and combined with multidimensional health features;

[0011] Dynamic early warning based on multidimensional health characteristics.

[0012] Preferably, a Fast Fourier Transform is performed on the total loop current waveform in the collected navigation lighting data to extract the harmonic intensity and harmonic phase of each harmonic. The harmonic intensity and harmonic phase are then arranged according to their harmonic order to form a harmonic vector, which is set as A, where A = [I h ,phi h In the formula, I h Phy represents harmonic intensity. h The harmonic phase is represented by h, and the harmonic order number is represented by h∈N. + And h≤H, where H represents the number of harmonics.

[0013] Preferably, the harmonic intensity spectral entropy is analyzed based on the harmonic intensity in the harmonic vector;

[0014] The harmonic vectors are stored as historical samples, and historical samples from the past 30 days that were in a healthy state are extracted. The average values ​​of the harmonic intensities numbered 3, 5, 7, and 9 are calculated as the baseline for odd-order harmonic intensities. The current relative drift d is then analyzed based on the odd-order harmonic intensity baseline and the harmonic intensities in the currently analyzed harmonic vectors. h d h =|I h -Ib h | / Ib h h = 3, 5, 7, 9, where Ib h This represents the baseline for odd harmonic intensity, where the average value of the current relative drift is used as the odd harmonic intensity.

[0015] Preferably, the phase difference sequence of the harmonic phase is extracted from the harmonic vector, wherein the phase difference sequence of the harmonic phase is from Δph2 to Δph 50 The phase difference sequence contains 49 values, each value being the difference between the current harmonic phase and the previous harmonic phase. The variance of the phase difference sequence of harmonic phases is used as the phase coherence parameter.

[0016] Preferably, the harmonic intensities greater than or equal to 0.001×I1 in the harmonic vector are labeled as qualified, and the number of harmonic vectors labeled as qualified in 2≤h≤H is counted as the qualified intensity quantity. The qualified intensity quantity of historical samples 5 minutes ago is extracted to calculate the harmonic scale growth rate. The harmonic scale growth rate = (current qualified intensity quantity - qualified intensity quantity of historical samples 5 minutes ago) / 5.

[0017] The energy centroid is analyzed based on the harmonic intensity in the harmonic vector. The average value of the energy centroid of historical samples that were in a healthy state in the past 30 days is used as the reference centroid, and the absolute value of the difference between the current energy centroid and the reference centroid is used as the centroid drift.

[0018] Preferably, a health cluster is constructed based on historical samples. Historical samples that were in a healthy state within the past 30 days and their corresponding multidimensional health features are extracted, and a health feature vector F is constructed, F=[S,Do,C,G,ΔHc], where Do represents the odd harmonic intensity, C represents the phase coherence parameter, G represents the harmonic scale growth rate, and ΔHc represents the centroid drift. The K-means++ algorithm is used to perform cluster analysis on the health feature vector of the historical samples. The number of clusters is determined by the elbow rule. The sum of squares within clusters with K values ​​between [2,10] is calculated, and the K value corresponding to the inflection point is selected to obtain K health clusters and their covariance matrices.

[0019] Preferably, for the health feature vector of the currently analyzed navigation light data, the health cluster center point with the closest Euclidean distance is found, and the Mahalanobis distance from the health feature vector of the currently analyzed navigation light data to the health cluster center point is calculated as the feature distance. The feature distance is mapped to a health score, and the expression of the health score is: HS=100×exp(-α×Dm), where HS represents the health score, α represents the attenuation coefficient, and Dm represents the feature distance.

[0020] Tracing rules are constructed based on health feature vectors and health scores to enable fault tracing of navigation lights.

[0021] Preferably, for each healthy cluster, a dynamic range of its feature threshold is set. For all historical samples belonging to the healthy cluster, the mean and standard deviation of each feature are calculated, and its dynamic range is set as: [mean of feature - 3 × standard deviation of feature, mean of feature + 3 × standard deviation of feature]. A regression model is established between ambient temperature, ambient humidity and the mean of each feature. When analyzing whether the current feature deviates, the mean of the current feature is first compensated and corrected using ambient temperature and ambient humidity, and then compared. An early warning is issued when the current feature does not belong to its corresponding dynamic range.

[0022] On the other hand, the present invention also provides a navigation light circuit fault detection device, comprising:

[0023] The data acquisition unit is used to collect data on airport navigation lights.

[0024] The harmonic extraction unit is used to extract and align the frequency bands of the navigation light data to generate harmonic vectors;

[0025] The feature analysis unit is used to construct a harmonic feature extraction model based on harmonic vectors to obtain multidimensional health features;

[0026] Clustering analysis unit is used for health measurement and fault tracing based on feature clustering and combined with multidimensional health features;

[0027] The feature-based early warning unit is used for dynamic early warning based on multi-dimensional health characteristics.

[0028] On the other hand, the present invention also provides an electronic device, the electronic device comprising:

[0029] One or more processors;

[0030] Storage device for storing one or more programs;

[0031] When the one or more programs are executed by the one or more processors, the one or more processors implement the navigation light circuit fault detection method as described above.

[0032] The beneficial effects of this invention are as follows: By performing deep harmonic extraction and feature modeling on the loop current waveform, a refined quantification of the health status of the lighting loop and early fault warning are achieved. This method overcomes the problems of lag and high false alarm rate of traditional manual inspection and threshold alarm methods. It can automatically identify various fault modes such as loop aging, poor contact, and interference intrusion from electrical signals, and provide fault type judgment and location clues. This significantly improves the intelligent operation and maintenance level and fault response speed of the airport lighting system, provides reliable data support for preventive maintenance, and effectively ensures the continuous and stable operation of the navigation lighting system and flight safety. Attached Figure Description

[0033] To more clearly illustrate the technical solutions in the embodiments of the present invention, the accompanying drawings used in the description of the embodiments will be briefly introduced below. Obviously, the accompanying drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0034] Figure 1 This is a flowchart of the navigation light circuit fault detection method in this embodiment.

[0035] Figure 2 This is a flowchart of the harmonic feature extraction method in this embodiment.

[0036] Figure 3 This is a flowchart of the navigation light circuit fault detection device in this embodiment.

[0037] Figure 4 This is a schematic diagram of the electronic device in this embodiment. Detailed Implementation

[0038] The following detailed description, in conjunction with the accompanying drawings and specific embodiments, provides a further detailed account of the navigation light circuit fault detection method, apparatus, and electronic equipment disclosed in this invention. It should be noted that the technical features or combinations of technical features described in the following embodiments should not be considered isolated; they can be combined to achieve better technical effects. In the accompanying drawings of the following embodiments, the same reference numerals in each drawing represent the same features or components, which can be applied to different embodiments. Therefore, once an item is defined in one drawing, it does not need to be further discussed in subsequent drawings.

[0039] It should be noted that the structures, proportions, sizes, etc., illustrated in the accompanying drawings are merely for illustrative purposes and to aid those skilled in the art in understanding and reading the invention. They are not intended to limit the conditions under which the invention can be implemented. Any modifications to the structure, changes in proportions, or adjustments to size, provided they do not affect the effectiveness or purpose of the invention, should fall within the scope of the technical content disclosed in the invention. The scope of the preferred embodiments of the present invention includes other implementations, wherein functions may be performed not in the order stated or discussed, including substantially simultaneously or in reverse order, depending on the functions involved. This should be understood by those skilled in the art to which the embodiments of the present invention pertain.

[0040] Techniques, methods, and apparatus known to those skilled in the art may not be discussed in detail, but where appropriate, such techniques, methods, and apparatus should be considered part of the specification. In all examples shown and discussed herein, any specific values ​​should be interpreted as merely exemplary and not as limitations. Therefore, other examples of exemplary embodiments may have different values.

[0041] In the description of the embodiments of this application, " / " means "or", and "and / or" is used to describe the relationship between related objects, indicating that there can be three relationships. For example, "A and / or B" means: A and B exist alone, B exists alone, and A and B exist simultaneously. In the description of the embodiments of this application, "multiple" refers to two or more embodiments.

[0042] Please see Figure 1 As shown, this is the method for detecting faults in the navigation light circuit in this embodiment, including:

[0043] Step S1: Collect airport navigation lighting data. The navigation lighting data refers to the data from the airport navigation lighting system, which includes a power supply, circuit cables, a lamp array, and a monitoring terminal. The power supply is a constant current dimmer. The circuit cables are buried cables connected in series with an isolation transformer. The lamp array includes multiple navigation lights. The monitoring terminal includes an embedded sensor module deployed at key points in the circuit, such as the dimmer output, the middle section of the circuit, and the end of the circuit. The embedded sensor module includes a high-precision current transformer, a voltage acquisition channel, a temperature and humidity sensor, and a microprocessor and communication module for processing and transmitting the data collected by the sensor. The navigation lighting data consists of the total circuit current waveform, single lamp voltage waveform, ambient temperature, ambient humidity, timestamp, and circuit identifier, all synchronously collected under steady-state lighting brightness. The sampling frequency of the total circuit current waveform should be greater than or equal to 10kHz to ensure that at least 50 harmonics are captured in the data.

[0044] Specifically, in step S1 of this embodiment, the total circuit current, single lamp voltage, and environmental parameters are synchronously collected using high-precision sensors to form a multi-dimensional, high-sampling-rate steady-state lighting data foundation. This step ensures that the data relied upon for subsequent analysis has sufficient timeliness, completeness, and accuracy, and can comprehensively reflect the electrical behavior and environmental impact of the lighting circuit under real working conditions. This provides a high-quality data source for subsequent harmonic feature extraction and health modeling, laying the cornerstone for reliable fault detection.

[0045] Please continue reading. Figure 1 As shown, the navigation light circuit fault detection method further includes:

[0046] Step S2 involves extracting and aligning the frequency bands of the navigation light data to generate harmonic vectors.

[0047] Specifically, in step S2 of this embodiment, a Fast Fourier Transform is performed on the total loop current waveform in the collected navigation light data to extract the harmonic intensity and harmonic phase of each harmonic. The harmonic intensity and harmonic phase are then arranged according to the harmonic order to form a harmonic vector, which is set as A, where A = [I h ,phi h In the formula, I h Phy represents harmonic intensity. h The harmonic phase is represented by h, and the harmonic order number is represented by h∈N. + And h≤H, where H represents the number of harmonics. The harmonic intensity is the effective value of the h-th harmonic, the harmonic phase is the phase difference relative to the fundamental wave, and the harmonic order number is defined as the number that distinguishes the number of harmonics collected.

[0048] Specifically, in step S2 of this embodiment, a fast Fourier transform is performed on the acquired current waveform to accurately extract the intensity and phase information of each harmonic and construct a structured harmonic vector. This step converts the time-domain signal into frequency-domain features, effectively highlighting the harmonic distortion characteristics caused by fault factors such as nonlinear loads and impedance changes in the circuit. It provides a core data carrier for subsequent in-depth feature mining and is a key technical step in realizing the conversion from the original waveform to fault-sensitive features.

[0049] Please continue reading. Figure 1 As shown, the navigation light circuit fault detection method further includes:

[0050] Step S3: Construct a harmonic feature extraction model based on the harmonic vector to obtain multidimensional health features, which include harmonic intensity spectral entropy, odd harmonic drift, phase coherence parameter, harmonic scale growth rate, and centroid drift.

[0051] Please see Figure 2 As shown, this is a method for extracting harmonic features, including:

[0052] Step S31: Analyze the harmonic intensity spectrum entropy based on the harmonic intensity in the harmonic vector.

[0053] Specifically, in step S31 of this embodiment, the harmonic intensity spectral entropy is analyzed based on the harmonic intensity in the harmonic vector. The expression for the harmonic intensity spectral entropy is: In the formula, S represents the harmonic intensity spectrum entropy.

[0054] Please continue reading. Figure 2 As shown, the method for extracting harmonic features further includes:

[0055] Step S32: Analyze the odd harmonic drift degree based on the harmonic intensity in the harmonic vector.

[0056] Specifically, in step S32 of this embodiment, the harmonic vector is stored as a historical sample, and historical samples that were in a healthy state in the past 30 days are extracted. The average value of the harmonic intensity of harmonics numbered 3, 5, 7, and 9 is calculated as the baseline of odd-order harmonic intensity. The current relative drift d is analyzed based on the baseline of odd-order harmonic intensity and the harmonic intensity in the currently analyzed harmonic vector. h d h =|I h -Ib h | / Ib h h = 3, 5, 7, 9, where Ib h This represents the baseline for odd harmonic intensity, where the average value of the current relative drift is used as the odd harmonic intensity.

[0057] Specifically, in step S32 of this embodiment, the health status is defined as collecting all navigation light data that have been running stably for at least one month after the navigation light circuit has been installed and debugged, and the default value is "health status".

[0058] Please continue reading. Figure 2 As shown, the method for extracting harmonic features further includes:

[0059] Step S33: Analyze the phase coherence parameters based on the harmonic phase in the harmonic vector.

[0060] Specifically, in step S33 of this embodiment, the phase difference sequence of the harmonic phase is extracted from the harmonic vector. The phase difference sequence of the harmonic phase is from Δph2 to Δph 50 The phase difference sequence contains 49 values, each value being the difference between the current harmonic phase and the previous harmonic phase. The variance of the phase difference sequence of harmonic phases is used as the phase coherence parameter.

[0061] Please continue reading. Figure 2 As shown, the method for extracting harmonic features further includes:

[0062] Step S34: Determine the harmonic scale growth rate based on the harmonic intensity in the harmonic vector.

[0063] Specifically, in step S34 of this embodiment, the harmonic intensities greater than or equal to 0.001×I1 in the harmonic vector are labeled as qualified, the number of harmonic vectors labeled as qualified in 2≤h≤H is counted as the qualified intensity quantity, the qualified intensity quantity of historical samples 5 minutes ago is extracted, and the harmonic scale growth rate is calculated. The harmonic scale growth rate = (current qualified intensity quantity - qualified intensity quantity of historical samples 5 minutes ago) / 5.

[0064] Please continue reading. Figure 2 As shown, the method for extracting harmonic features further includes:

[0065] Step S35: Analyze the centroid drift based on the harmonic intensity in the harmonic vector.

[0066] Specifically, in step S35 of this embodiment, the energy centroid is analyzed based on the harmonic intensity in the harmonic vector. The expression for the energy centroid is: In the formula, Hc represents the energy centroid, which is the average energy centroid of historical samples that were in a healthy state over the past 30 days as the baseline centroid, and the absolute value of the difference between the current energy centroid and the baseline centroid as the centroid drift.

[0067] Specifically, in step S3 of this embodiment, a feature extraction model is constructed based on harmonic vectors, resulting in a multi-dimensional health feature set including spectral entropy, odd harmonic drift, and phase coherence. These features characterize the circuit state from multiple perspectives, such as energy distribution uniformity, specific harmonic stability, phase relationship consistency, and dynamic changes in harmonic scale, forming a comprehensive, three-dimensional, and sensitive digital description of the circuit health, which greatly enhances the ability to distinguish and identify fault modes.

[0068] Please continue reading. Figure 1 As shown, the navigation light circuit fault detection method further includes:

[0069] Step S4: Based on feature clustering and combined with multidimensional health features, perform health quantification and fault tracing.

[0070] Specifically, in step S4 of this embodiment, a health cluster is constructed based on historical samples. Historical samples that were in a healthy state within the past 30 days and their corresponding multidimensional health features are extracted, and a health feature vector F is constructed, F=[S,Do,C,G,ΔHc], where Do represents the odd harmonic intensity, C represents the phase coherence parameter, G represents the harmonic scale growth rate, and ΔHc represents the centroid drift. The K-means++ algorithm is used to perform cluster analysis on the health feature vector of the historical samples. The number of clusters is determined using the elbow rule. The sum of squares within clusters with K values ​​between [2,10] is calculated, and the K value corresponding to the inflection point is selected to obtain K healthy clusters and their covariance matrices. The health feature vector constructed based on the historical samples is the mean of the corresponding data, and it is used as the feature baseline.

[0071] Specifically, in step S4 of this embodiment, for the health feature vector of the currently analyzed navigation light data, the health cluster center point with the closest Euclidean distance is found, and the Mahalanobis distance from the health feature vector of the currently analyzed navigation light data to the health cluster center point is calculated as the feature distance. The feature distance is mapped to a health score, and the expression for the health score is: HS = 100 × exp(-α × Dm), where HS represents the health score, α represents the attenuation coefficient, and Dm represents the feature distance. The value of the attenuation coefficient is determined by inversely calculating that when Dm = 2, HS ≈ 60.

[0072] Specifically, in step S4 of this embodiment, a tracing rule is constructed based on the health feature vector and health score to achieve fault tracing of the navigation lights. The tracing rule is as follows:

[0073] When Do is greater than twice its corresponding characteristic baseline and S increases synchronously, the fault type is determined to be power supply / dimmer abnormality. The auxiliary judgment method is to check whether the current waveform at the beginning of the circuit is distorted. The location clue is to compare the Do of other circuits. If the Do of more than 60% of the circuits increases, it points to the common dimming cabinet.

[0074] When C is greater than 3 times its corresponding characteristic baseline and G is positive, the fault type is determined to be intermittent arcing due to loose cable joint / poor contact. The auxiliary judgment method is to check the historical data to see if the C value increases intermittently and in a pulse-like manner. The location clue is to compare the C values ​​at different locations, and the C value upstream of the fault point changes more drastically.

[0075] When ΔHc>2 and the number of qualified intensity increases, the fault type is determined to be high-frequency interference intrusion / lamp electronic component failure. The auxiliary judgment method is to check whether there is a new device connected to the same power grid. The location clue is: when it is a single circuit problem, use a portable high-frequency probe to inspect along the circuit to find the high-frequency noise source.

[0076] When S is less than 0.5 times its corresponding characteristic baseline, the fault type is determined to be a nonlinear load fault. The auxiliary judgment method is to analyze the harmonic spectrum, locate the harmonic number of abnormal energy concentration, and provide the following location clues: the harmonic number of abnormal energy concentration may be related to the circuit topology of a specific lamp, which can be investigated in a targeted manner.

[0077] When all data in the health feature vector are within [0.9, 1) times its corresponding feature baseline and the health score shows a downward trend, the fault type is determined to be comprehensive circuit aging. The auxiliary judgment method is trend analysis, and the location clues are: observe the slow drift trend of each feature over time and formulate a preventive maintenance plan.

[0078] Specifically, in step S4 of this embodiment, various normal patterns are constructed using historical health data through a clustering algorithm, and Mahalanobis distance is used to calculate the deviation between the current state and the nearest healthy cluster, which is then mapped to an intuitive health score. This step achieves a quantitative assessment of health status, transforming loop status from qualitative judgment to precise scoring. Combined with preset tracing rules, it is possible to associate characteristic anomaly combinations with specific fault types and provide auxiliary judgment methods and location clues, thereby achieving a leap from "abnormal status" to "fault diagnosis and location".

[0079] Please continue reading. Figure 1 As shown, the navigation light circuit fault detection method further includes:

[0080] Step S5: Dynamic early warning based on multidimensional health characteristics.

[0081] Specifically, in step S5 of this embodiment, for each healthy cluster, a dynamic range of its feature threshold is set. For all historical samples belonging to the healthy cluster, the mean and standard deviation of each feature are calculated, and its dynamic range is set as: [mean of feature - 3 × standard deviation of feature, mean of feature + 3 × standard deviation of feature]. A regression model of environmental temperature, environmental humidity and the mean of each feature is established. When analyzing whether the current feature deviates, the mean of the current feature is first compensated and corrected using environmental temperature and environmental humidity, and then compared to eliminate the influence of seasonal environment. An early warning is issued when the current feature does not belong to its corresponding dynamic range.

[0082] Specifically, in step S5 of this embodiment, a dynamic early warning threshold based on historical statistics is set for the characteristics of each healthy cluster, and an environmental temperature and humidity regression model is introduced for compensation and correction. This step fully considers the natural fluctuations during normal equipment operation and the impact of seasonal environmental factors, effectively avoiding false alarms caused by environmental changes. Through dynamic threshold comparison, the system can issue an early warning when the characteristics deviate in the early stages but before a fault occurs, truly realizing predictive maintenance and shifting the operation and maintenance mode from post-maintenance to pre-intervention.

[0083] Please see Figure 3 As shown, this is the navigation light circuit fault detection device of this embodiment, including:

[0084] The data acquisition unit is used to collect data on airport navigation lights.

[0085] The harmonic extraction unit is used to extract and align the frequency bands of the navigation light data to generate harmonic vectors;

[0086] The feature analysis unit is used to construct a harmonic feature extraction model based on harmonic vectors to obtain multidimensional health features;

[0087] Clustering analysis unit is used for health measurement and fault tracing based on feature clustering and combined with multidimensional health features;

[0088] The feature-based early warning unit is used for dynamic early warning based on multi-dimensional health characteristics.

[0089] Please see Figure 4 As shown, it is a structural schematic diagram of an electronic device in this embodiment. The electronic device 60 in this embodiment may include, but is not limited to, mobile terminals such as mobile phones, laptops, digital broadcast receivers, PDAs (personal digital assistants), PADs (tablet computers), PMPs (portable multimedia players), vehicle terminals (such as vehicle navigation terminals), wearable electronic devices, etc., as well as fixed terminals such as digital TVs, desktop computers, smart home devices, etc. Figure 4The electronic device shown is merely an example and should not be construed as limiting the functionality and scope of use of the embodiments of the present invention.

[0090] like Figure 4 As shown, the electronic device 60 may include a processing unit 61, which can perform various appropriate actions and processes to implement the methods of the embodiments described herein, based on a program stored in ROM 62 or a program loaded from storage device 68 into RAM 63. RAM 63 also stores various programs and data required for the operation of the electronic device 60. The processing unit 61, ROM 62, and RAM 63 are interconnected via bus 64. I / O interface 65 is also connected to bus 64. Typically, the following devices can be connected to I / O interface 65: input devices 66 including, for example, touchscreens, touchpads, keyboards, mice, cameras, microphones, accelerometers, gyroscopes, etc.; output devices 67 including, for example, liquid crystal displays (LCDs), speakers, vibrators, etc.; storage devices 68 including, for example, magnetic tapes, hard disks, etc.; and communication devices 69. Communication device 69 allows the electronic device 60 to communicate wirelessly or wiredly with other devices to exchange data. Although Figure 4 An electronic device 60 with various devices is shown; however, it should be understood that it is not required to implement or possess all of the devices shown. More or fewer devices may be implemented or possessed alternatively.

[0091] Specifically, according to embodiments of the present invention, the processes described above with reference to the flowcharts can be implemented as computer software programs. For example, embodiments of the present invention include a computer program product comprising a computer program carried on a non-transitory computer-readable medium, the computer program containing program code for performing the methods shown in the flowcharts, thereby implementing the methods as described above. In such embodiments, the computer program can be downloaded and installed from a network via a communication device 69, or installed from a storage device 68, or installed from a ROM 62. When the computer program is executed by the processing device 61, it performs the functions defined in the methods of the embodiments of the present invention.

[0092] Specifically, the computer-readable medium described in this embodiment may be included in the aforementioned electronic device; or it may exist independently and not assembled into the electronic device.

[0093] Specifically, in this embodiment, the computer-readable medium carries one or more programs. When the electronic device executes the one or more programs, the electronic device causes the following: to collect airport navigation light data; to extract and align the navigation light data in frequency bands to generate harmonic vectors; to construct a harmonic feature extraction model based on the harmonic vectors to obtain multidimensional health features; to perform health quantification and fault tracing based on feature clustering and combined with multidimensional health features; and to provide dynamic early warning based on multidimensional health features.

[0094] Obviously, the above embodiments of the present invention are merely examples for clearly illustrating the present invention, and are not intended to limit the implementation of the present invention. For those skilled in the art, other variations or modifications can be made based on the above description. It is impossible to exhaustively list all the implementation methods here. All obvious variations or modifications derived from the technical solutions of the present invention are still within the protection scope of the present invention.

Claims

1. A method for detecting faults in navigation light circuits, characterized in that, include: Collect data on airport navigation lights; Frequency band extraction and alignment are performed on the navigation light data to generate harmonic vectors; A harmonic feature extraction model is constructed based on harmonic vectors to obtain multidimensional health features; Health measurement and fault tracing are performed based on feature clustering and combined with multidimensional health features; Dynamic early warning based on multidimensional health characteristics.

2. The method for detecting faults in navigation light circuits according to claim 1, characterized in that, A Fast Fourier Transform (FFT) is performed on the total loop current waveform in the collected navigation lighting data to extract the harmonic intensity and phase of each harmonic. The harmonic intensity and phase are then arranged according to their harmonic order to form a harmonic vector, denoted as A, where A = [I...]. h ,phi h In the formula, I h Phy represents harmonic intensity. h The harmonic phase is represented by h, and the harmonic order number is represented by h∈N. + And h≤H, where H represents the number of harmonics.

3. The method for detecting faults in navigation light circuits according to claim 2, characterized in that, Analysis of harmonic intensity spectrum entropy based on harmonic intensity in harmonic vectors; The harmonic vectors are stored as historical samples, and historical samples from the past 30 days that were in a healthy state are extracted. The average values ​​of the harmonic intensities numbered 3, 5, 7, and 9 are calculated as the baseline for odd-order harmonic intensities. The current relative drift d is then analyzed based on the odd-order harmonic intensity baseline and the harmonic intensities in the currently analyzed harmonic vectors. h d h =|I h -Ib h | / Ib h h = 3, 5, 7, 9, where Ib h This represents the baseline for odd harmonic intensity, where the average value of the current relative drift is used as the odd harmonic intensity.

4. The method for detecting faults in navigation light circuits according to claim 3, characterized in that, Extract the phase difference sequence of the harmonic phase from the harmonic vector. The phase difference sequence of the harmonic phase is from Δph2 to Δph 50 The phase difference sequence contains 49 values, each value being the difference between the current harmonic phase and the previous harmonic phase. The variance of the phase difference sequence of harmonic phases is used as the phase coherence parameter.

5. The method for detecting faults in navigation light circuits according to claim 4, characterized in that, Mark the harmonic intensities greater than or equal to 0.001×I1 in the harmonic vectors as qualified. Count the number of harmonic vectors marked with qualified labels in 2≤h≤H as the qualified intensity quantity. Extract the qualified intensity quantity of historical samples from 5 minutes ago to calculate the harmonic scale growth rate. The harmonic scale growth rate = (current qualified intensity quantity - qualified intensity quantity of historical samples from 5 minutes ago) / 5. The energy centroid is analyzed based on the harmonic intensity in the harmonic vector. The average value of the energy centroid of historical samples that were in a healthy state in the past 30 days is used as the reference centroid, and the absolute value of the difference between the current energy centroid and the reference centroid is used as the centroid drift.

6. The method for detecting faults in navigation light circuits according to claim 5, characterized in that, Based on historical samples, health clusters are constructed. Historical samples that were in a healthy state within the past 30 days and their corresponding multidimensional health features are extracted, and a health feature vector F is constructed, F=[S,Do,C,G,ΔHc], where Do represents the odd harmonic intensity, C represents the phase coherence parameter, G represents the harmonic scale growth rate, and ΔHc represents the centroid drift. The K-means++ algorithm is used to perform cluster analysis on the health feature vectors of historical samples. The number of clusters is determined by the elbow rule. The sum of squares within clusters with K values ​​between [2,10] is calculated, and the K value corresponding to the inflection point is selected to obtain K health clusters and their covariance matrices.

7. The method for detecting faults in navigation light circuits according to claim 6, characterized in that, For the health feature vector of the currently analyzed navigation light data, find the health cluster center point with the closest Euclidean distance, and calculate the Mahalanobis distance from the health feature vector of the currently analyzed navigation light data to the health cluster center point as the feature distance. Map the feature distance to the health score. The expression of the health score is: HS=100×exp(-α×Dm), where HS represents the health score, α represents the attenuation coefficient, and Dm represents the feature distance. Tracing rules are constructed based on health feature vectors and health scores to enable fault tracing of navigation lights.

8. The method for detecting faults in navigation light circuits according to claim 7, characterized in that, For each healthy cluster, a dynamic range for its feature thresholds is set. For all historical samples belonging to the healthy cluster, the mean and standard deviation of each feature are calculated, and its dynamic range is set as: [mean of feature - 3 × standard deviation of feature, mean of feature + 3 × standard deviation of feature]. A regression model is established between ambient temperature, ambient humidity and the mean of each feature. When analyzing whether the current feature deviates, the mean of the current feature is first compensated and corrected using ambient temperature and ambient humidity, and then compared. An early warning is issued when the current feature does not belong to its corresponding dynamic range.

9. A navigation light circuit fault detection device, applied to the navigation light circuit fault detection method as described in any one of claims 1-8, characterized in that, include: The data acquisition unit is used to collect data on airport navigation lights. The harmonic extraction unit is used to extract and align the frequency bands of the navigation light data to generate harmonic vectors; The feature analysis unit is used to construct a harmonic feature extraction model based on harmonic vectors to obtain multidimensional health features; Clustering analysis unit is used for health measurement and fault tracing based on feature clustering and combined with multidimensional health features; The feature-based early warning unit is used for dynamic early warning based on multi-dimensional health characteristics.

10. An electronic device, characterized in that, The electronic device includes: One or more processors; Storage device for storing one or more programs; When the one or more programs are executed by the one or more processors, the one or more processors implement the navigation light circuit fault detection method as described in any one of claims 1-8.

Images