A smart street lamp fault diagnosis and maintenance system

Abstract

This invention relates to the field of fault diagnosis technology, specifically to a smart street light fault diagnosis and maintenance system. The system includes: a signal collaborative acquisition module, a frequency domain conversion module, a fault identification module, a coupling verification module, and a maintenance scheduling module. In this invention, multi-source data synchronous acquisition and timing alignment technology are used to eliminate the influence of external environmental interference on monitoring data, ensuring the stability and accuracy of photoelectric signals. Stable signal features are extracted through frequency domain conversion and feature filtering. Accurate fault identification is achieved based on frequency components and similarity calculations. Multi-dimensional verification is performed by combining instantaneous power comparison and communication delay detection, improving the accuracy and reliability of fault judgment, thereby avoiding misjudgments and omissions, optimizing maintenance scheduling, improving the fault response efficiency and overall stability of the street light system, effectively reducing maintenance delays and resource waste caused by inaccurate fault diagnosis, and improving the operational reliability and maintenance response speed of street lights.

Description

Technical Field

[0001] This invention relates to the field of fault diagnosis technology, and in particular to a smart street light fault diagnosis and maintenance system. Background Technology

[0002] The field of fault diagnosis technology involves the monitoring, analysis, and determination of the operating status of various types of equipment, systems, or devices. By collecting operating parameters, environmental information, and system feedback signals, potential faults in equipment are identified and located to ensure reliable system operation and timely maintenance. This technology covers core aspects such as signal acquisition, feature extraction, status recognition, and diagnostic decision-making, and is achieved through methods such as sensor detection, pattern recognition, data analysis, and diagnostic rule establishment. Therefore, it is widely used in scenarios such as industrial production, power systems, transportation facilities, and intelligent devices.

[0003] The traditional smart street light fault diagnosis and maintenance system refers to the diagnosis and maintenance system built to address problems such as light source damage, power supply abnormalities, communication failures, and environmental sensor failures that occur during the operation of smart street lights in cities. It relies on regular manual inspections or simple remote monitoring to obtain street light status information, and determines whether the street light is working properly by detecting parameters such as current, voltage, brightness, and communication signals. Then, it determines the fault type based on manual analysis or preset thresholds and arranges maintenance personnel to carry out maintenance work.

[0004] Traditional smart street light fault diagnosis and maintenance systems rely on manual inspections and simple remote monitoring. The lack of synchronization of multi-source data during signal acquisition makes monitoring results susceptible to fluctuations in ambient light and climate change, reducing diagnostic stability. Monitoring parameters such as voltage and current cannot effectively eliminate external interference, affecting fault determination accuracy. Fault determination methods based on fixed thresholds have limited ability to identify complex faults, failing to accurately distinguish between different types of faults, leading to frequent misdiagnosis or missed diagnosis. This, in turn, delays maintenance response, wastes resources, and impacts the safety and operational efficiency of the street light system. Summary of the Invention

[0005] To address the shortcomings of traditional smart street light fault diagnosis and maintenance systems, which rely on manual inspections and simple remote monitoring, lacking synchronization of multi-source data during signal acquisition, and thus susceptible to fluctuations in ambient light and climate change, leading to reduced diagnostic stability, and the inability to effectively eliminate external interference during voltage and current monitoring, thus affecting fault determination accuracy, and the limited ability of fixed-threshold-based fault determination methods to identify complex faults, resulting in frequent misdiagnosis or missed diagnosis, delayed maintenance response, wasted resources, and impact on the safety and operational efficiency of the street light system, this invention provides a smart street light fault diagnosis and maintenance system. The technical solution is as follows:

[0006] On the one hand, a smart street light fault diagnosis and maintenance system is provided, which includes:

[0007] The signal collaborative acquisition module synchronously acquires voltage time-series data, power line current and voltage phase data, and communication link transmission delay characteristics through photovoltaic sensors. It eliminates ambient light fluctuation interference and performs timestamp alignment to generate a multi-source synchronous dataset, which is then transmitted to the frequency domain conversion module.

[0008] The frequency domain conversion module calls the multi-source synchronous dataset, extracts voltage time-series data, performs Fourier transform to calculate frequency components and intensity values, adaptively sets an intensity threshold to filter effective components, calculates the average intensity value, generates photoelectric signal frequency domain features, and transmits them to the fault identification module.

[0009] The fault identification module receives the frequency domain features of the photoelectric signal, extracts the frequency component values, calculates the similarity score with the fault benchmark value using a cosine similarity algorithm, selects the first fault type in descending order, generates preliminary fault diagnosis information, and transmits it to the coupling verification module.

[0010] The coupling verification module calls the preliminary fault diagnosis information, calculates the instantaneous power comparison deviation threshold based on current and voltage to mark power abnormalities, checks if the transmission delay exceeds the delay threshold to mark communication abnormalities, verifies the matching relationship of the fault period markings, generates street light fault confirmation results, and transmits them to the maintenance scheduling module.

[0011] As a further embodiment of the present invention, the multi-source synchronization dataset includes voltage timing data, current phase data, voltage phase data, communication transmission delay characteristics, and timestamp alignment information; the photoelectric signal frequency domain characteristics include frequency components, average signal strength, strength threshold, and effective frequency components; the preliminary fault diagnosis information includes fault type, similarity score, and fault priority sequence; and the street light fault confirmation result includes power anomaly markers, communication anomaly markers, and fault time period matching information.

[0012] As a further aspect of the present invention, the signal cooperative acquisition module includes:

[0013] The photovoltaic signal acquisition submodule acquires the output voltage signal stream of the photovoltaic sensor, detects continuous time points of the sampled voltage, calculates the rate of change of light intensity based on the amplitude difference of the time points, calculates the voltage fluctuation, filters out abnormal points, and generates photovoltaic voltage fluctuation sequence values.

[0014] The electrical parameter synchronous measurement submodule synchronously acquires current and voltage signals based on the photovoltaic voltage fluctuation sequence value, analyzes the signal phase difference, compares it with the phase reference value, calculates the instantaneous power distribution, and obtains the current and voltage phase difference coefficient by comparing the time difference between the photovoltaic fluctuation sequence and the power distribution.

[0015] The timing alignment and fusion submodule extracts communication link delay features based on the current-voltage phase difference coefficient, monitors and corrects time stamp differences, filters out abnormal delay data based on the weighted average value, and generates a multi-source synchronization dataset.

[0016] As a further aspect of the present invention, the frequency domain conversion module includes:

[0017] The data timing extraction submodule acquires voltage timing data from the multi-source synchronous dataset, arranges the channel voltage sampling points in time and unifies the sampling step size, calculates the sampling interval difference and corrects the deviation, performs channel time synchronization correction, and generates voltage timing matrix values.

[0018] The frequency component calculation submodule calls the voltage time series matrix value, performs a Fourier transform on the time series, extracts the real and imaginary parts of the complex number and calculates the amplitude energy, statistically analyzes the intensity by frequency index and normalizes it, and generates a frequency intensity distribution sequence.

[0019] The threshold filtering feature generation submodule calculates the mean intensity and fluctuation deviation based on the frequency intensity distribution sequence, sets an adaptive intensity threshold, removes low-intensity frequency points and reconstructs the effective frequency set, calculates the weighted average value, and generates the frequency domain features of the photoelectric signal.

[0020] As a further aspect of the present invention, the intensity threshold is dynamically corrected based on the mean, variance, and median of the frequency intensity distribution sequence, combined with the energy concentration and the difference between the local peak and the global average.

[0021] As a further aspect of the present invention, the fault identification module includes:

[0022] The frequency domain input construction submodule calls the frequency domain features of the photoelectric signal, performs segmented sampling, filters out noise from the segmented signals and delineates the effective interval according to the time window, performs normalization adjustment on the frequency domain amplitude value, and generates the photoelectric signal frequency domain input matrix.

[0023] The frequency domain feature extraction submodule calculates the frequency component energy weighting value and extracts the main frequency component based on the frequency domain input matrix of the photoelectric signal, calculates the frequency band energy ratio and calculates the amplitude change rate, performs linear smoothing correction on the feature value, and obtains the frequency component feature vector set.

[0024] The similarity calculation and determination submodule, based on the frequency component feature vector set, calls the fault baseline data, uses the cosine similarity algorithm to compare the corresponding frequency band features, calculates the similarity score, performs weighted descending sorting, selects the fault type at the top of the sort, and generates preliminary fault diagnosis information.

[0025] As a further aspect of the present invention, the coupling verification module includes:

[0026] The signal power analysis submodule calls the preliminary fault diagnosis information to obtain the current and voltage signals at the street light end, aligns the current and voltage signals according to the timestamp, calculates the instantaneous power and maps it to the diagnostic information, and generates an instantaneous power mapping matrix.

[0027] The power anomaly determination submodule calls the instantaneous power mapping matrix to calculate the average instantaneous power over multiple time periods. It then compares the average value with a set deviation threshold. If the average value is greater than the deviation threshold, the time interval is marked as a power anomaly, and a power anomaly interval set is generated.

[0028] The communication anomaly verification submodule calls the power anomaly interval set, calculates the signal transmission delay within the interval, compares the delay threshold with the message sending and receiving time, and if the delay is greater than the set communication delay threshold, it is marked as a communication anomaly, matches the power anomaly interval, and generates a street light fault confirmation result.

[0029] As a further aspect of the present invention, the power deviation threshold is determined based on the street light rated power curve and the statistical results of the original operating power fluctuation;

[0030] The communication delay threshold is determined based on the communication protocol standard and message transmission performance statistics of the street light control network.

[0031] As a further embodiment of the present invention, the maintenance scheduling module receives the street light fault confirmation result, obtains the equipment number and location coordinates, matches the fault type with the corresponding maintenance operation type, integrates the equipment number, location coordinates, fault type, operation type and timestamp, and generates a maintenance work order;

[0032] The maintenance work order includes the equipment number, location coordinates, fault type, job type, and timestamp.

[0033] As a further aspect of the present invention, the maintenance scheduling module includes:

[0034] The device positioning and analysis submodule obtains the street light fault confirmation result, detects the device number and location coordinate format, indexes the coordinate mapping table according to the device number, calculates the coordinate difference to determine the offset range, corrects abnormal coordinate points, and generates a location coordinate correction value.

[0035] The job matching submodule, based on the location coordinate correction value, calls the fault type corresponding to the equipment number, compares the fault type with the job type mapping table, calculates the job level parameters and performs a difference comparison, selects the corresponding job type number based on the difference consistency result, and generates job type matching parameters.

[0036] The work order generation submodule, based on the job type matching parameters, calls the equipment number, location coordinates, and fault type data, concatenates the fields and adds a timestamp, calculates the combined integrity check value, and generates a maintenance work order dataset.

[0037] The beneficial effects of the technical solutions provided in the embodiments of the present invention include at least the following:

[0038] By employing multi-source data synchronous acquisition and time-series alignment technology, the impact of external environmental interference on monitoring data is eliminated, ensuring the stability and accuracy of photoelectric signals. Stable signal features are extracted through frequency domain conversion and feature filtering. Accurate fault identification is achieved based on frequency component and similarity calculations. Multi-dimensional verification is performed using instantaneous power comparison and communication delay detection, improving the accuracy and reliability of fault determination. This avoids false positives and false negatives, optimizes maintenance scheduling, ensures reasonable resource allocation, and enhances the fault response efficiency and overall stability of the streetlight system. Attached Figure Description

[0039] 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 the accompanying drawings without creative effort.

[0040] Figure 1 This is a system schematic diagram of the present invention;

[0041] Figure 2 This is a schematic diagram of the system framework of the present invention;

[0042] Figure 3 This is a flowchart of the signal collaborative acquisition module in this invention;

[0043] Figure 4 This is a flowchart of the frequency domain conversion module in this invention;

[0044] Figure 5 This is a flowchart of the fault identification module in this invention;

[0045] Figure 6 This is a flowchart of the coupling verification module in this invention;

[0046] Figure 7 This is a flowchart of the maintenance scheduling module in this invention. Detailed Implementation

[0047] The technical solution of the present invention will now be described with reference to the accompanying drawings.

[0048] In embodiments of the present invention, words such as "exemplarily," "for example," etc., are used to indicate that something is an example, illustration, or description. Any embodiment or design described as "exemplary" in the present invention should not be construed as being more preferred or advantageous than other embodiments or designs. Specifically, the use of the word "exemplary" is intended to present the concept in a concrete manner. Furthermore, in embodiments of the present invention, the meaning expressed by "and / or" can be both, or either one.

[0049] In the embodiments of this invention, the terms "image" and "picture" may sometimes be used interchangeably. It should be noted that, without emphasizing the distinction between them, they convey the same meaning. Similarly, the terms "of," "corresponding (relevant)," and "corresponding" may sometimes be used interchangeably. It should be noted that, without emphasizing the distinction between them, they convey the same meaning.

[0050] In this embodiment of the invention, sometimes the subscript such as W1 is written in a non-subscript form such as W1. When the difference is not emphasized, the meaning they express is the same.

[0051] To make the technical problems, technical solutions and advantages of the present invention clearer, a detailed description will be given below in conjunction with the accompanying drawings and specific embodiments.

[0052] This invention provides a smart street light fault diagnosis and maintenance system, such as... Figure 1-2 The diagram shown illustrates a smart street light fault diagnosis and maintenance system, which includes:

[0053] The signal collaborative acquisition module synchronously acquires voltage time-series data, power line current and voltage phase data, and communication link transmission delay characteristics through photovoltaic sensors. It eliminates ambient light fluctuation interference and performs timestamp alignment to generate a multi-source synchronous dataset, which is then transmitted to the frequency domain conversion module.

[0054] The frequency domain conversion module calls the multi-source synchronous dataset, extracts voltage time-series data, performs Fourier transform to calculate frequency components and intensity values, adaptively sets intensity thresholds to filter effective components, calculates the average intensity value, generates photoelectric signal frequency domain features, and transmits them to the fault identification module.

[0055] The fault identification module receives the frequency domain features of the photoelectric signal, extracts the frequency component values, calculates the similarity score with the fault benchmark value using the cosine similarity algorithm, selects the first fault type in descending order, generates preliminary fault diagnosis information, and transmits it to the coupling verification module.

[0056] The coupling verification module calls the preliminary fault diagnosis information, calculates the instantaneous power comparison deviation threshold based on current and voltage to mark power abnormalities, checks the transmission delay to mark communication abnormalities when it exceeds the delay threshold, verifies the matching relationship of the fault period, generates the street light fault confirmation result, and transmits it to the maintenance scheduling module.

[0057] The maintenance scheduling module receives the street light fault confirmation result, obtains the equipment number and location coordinates, matches the fault type with the corresponding maintenance job type, integrates the equipment number, location coordinates, fault type, job type and timestamp, and generates a maintenance work order;

[0058] The multi-source synchronization dataset includes voltage timing data, current phase data, voltage phase data, communication transmission delay features, and timestamp alignment information. The photoelectric signal frequency domain features include frequency components, average signal strength, strength threshold, and effective frequency components. The preliminary fault diagnosis information includes fault type, similarity score, and fault priority sequence. The street light fault confirmation results include power anomaly markers, communication anomaly markers, and fault time period matching information. The maintenance work order includes equipment number, location coordinates, fault type, work type, and timestamp.

[0059] Specifically, such as Figure 2 , 3 As shown, the signal collaborative acquisition module includes:

[0060] The photovoltaic signal acquisition submodule acquires the output voltage signal stream of the photovoltaic sensor, detects continuous time points of the sampled voltage, calculates the rate of change of light intensity based on the amplitude difference of the time points, calculates the voltage fluctuation, filters out abnormal points, and generates photovoltaic voltage fluctuation sequence values.

[0061] Set up photovoltaic sensors to The sampling frequency is connected to the analog input port of the street light controller to initiate real-time capture of the photovoltaic panel output voltage signal stream, and the sampling time window is set to [value missing]. Read continuously within this window For each voltage sampling point, the collected discrete voltage data is stored in a temporary buffer queue in chronological order, as shown in Table 1. The first [sample data point] is selected... Data from a typical sampling point to They are respectively , , , , Further analysis was conducted, and Table 1 lists the voltage sampling data of the photovoltaic sensor at some time points. The voltage data in the buffer queue was traversed, and the current time was extracted sequentially. voltage value Compared to the previous moment voltage value Perform difference operation Substituting the aforementioned data, the calculation yields... , , , Set a benchmark threshold for determining the rate of change in light intensity. The threshold Based on the maximum voltage drop rate caused by cloud cover in clear weather, the analysis of raw meteorological data shows that cloud cover leads to voltage drops at certain rates. The maximum decrease within is Therefore, it is set The calculated result and Compare them one by one, if If so, then the point is determined to be a point of sudden change in light intensity, i.e., in the above example. and Therefore, it is marked. and The time point is defined as the point of abnormal change. For the selected abnormal point and its neighborhood data, the voltage fluctuation is defined. The calculation logic is the arithmetic mean of the absolute values ​​of the differences within the window, i.e. The calculation process is applied to the entire sampling window, and the calculated fluctuation values ​​are arranged according to the time axis to generate a photovoltaic voltage fluctuation sequence value.

[0062] Table 1: Photovoltaic Sensor Sampling Voltage Data Table

[0063] Sampling point number Timestamp (s) Voltage value (V) Remark 1 10.01 18.50 Initial steady state 2 10.02 18.48 tiny fluctuations 3 10.03 17.20 Sudden drop 4 10.04 17.15 Maintaining low level 5 10.05 18.45 Mutation recovery

[0064] As shown in Table 1, this table records the photovoltaic voltage data at five consecutive time points, reflecting the dramatic changes in voltage over a short period of time.

[0065] The electrical parameter synchronous measurement submodule synchronously acquires current and voltage signals based on the photovoltaic voltage fluctuation sequence value, analyzes the signal phase difference, compares it with the phase reference value, calculates the instantaneous power distribution, and obtains the current and voltage phase difference coefficient by comparing the time difference between the photovoltaic fluctuation sequence and the power distribution.

[0066] Call the generated photovoltaic voltage fluctuation sequence value and lock the fluctuation amount. Using the time segment as the key analysis interval, high-precision current transformers and voltage transformers are activated, with the same... The frequency is used to synchronously read the current signal at the power supply terminal of the street light driver within the critical analysis range. With voltage signal Select the same time The data was used to measure the voltage. Current Using zero-crossing detection logic, the times when the voltage waveform crosses zero from the negative half-cycle to the positive half-cycle are recorded respectively. With the zero-crossing moment of the current waveform Assuming the measurement is , Calculate the time difference Combined with power grid frequency Calculate the signal phase difference Set the phase reference value The standard phase angle of the streetlight under rated full-load conditions is determined by laboratory testing. Compare and calculate the phase deviation. Next, instantaneous power distribution calculation is performed using the formula. Substitute the aforementioned numerical values ​​into the calculation. instantaneous power at time Repeat this power calculation for the sampling points within the analysis interval to construct the instantaneous power distribution curve and extract the starting moment when the power distribution curve drops. And obtain the start time of voltage abrupt change in the aforementioned photovoltaic voltage fluctuation sequence. Assuming (Corresponding to point 3 in Table 1), and the monitored power drop start time Calculate the time difference between the two. This time difference characterizes the physical transmission and processing delay from photovoltaic signal anomaly to electrical parameter response, and defines the current-voltage phase difference coefficient. This is the ratio of phase deviation to time difference, i.e. Substituting the numerical values, we obtain This coefficient reflects the dynamic response characteristics of the system under fault disturbances, and the current-voltage phase difference coefficient is obtained.

[0067] The timing alignment and fusion submodule extracts communication link delay features based on the current and voltage phase difference coefficients, monitors and corrects time stamp differences, filters out abnormal delay data based on the weighted average value, and generates a multi-source synchronization dataset.

[0068] Receive the calculated current-voltage phase difference coefficient and the corresponding time difference ,Will The latency characteristics identified as being processed by the current communication link and system are, i.e. Set the allowable range for latency correction. This range is set based on the average transmission latency statistics of the system communication protocol (such as NB-IoT or ZigBee). Statistics show that normal latency distribution is within... In between, determine the current Does it fall within this range? In Within this range, the data is considered valid delay data; if it exceeds this range, it is considered a transmission error and is discarded. For valid data, the original time stamps are extracted from the electrical parameter dataset. Execution time correction calculation For example, regarding the aforementioned power drop moment The corrected timestamp is This achieves alignment with the photovoltaic signal time axis. To further improve synchronization accuracy, time delay characteristics of five consecutive time periods are collected, assuming they are respectively... Calculate its average value and standard deviation If the deviation of a certain time delay value from the average value exceeds If so, it is defined as abnormal latency data and filtered out, for example... Those that deviate significantly from the other values ​​should be removed. For the remaining values... The final time delay correction value is calculated using a weighted average method, with data closer to the current time having a higher weight. The weight vector is set as follows: Calculate the weighted average delay Using this final correction value The timestamps of the electrical parameter data are uniformly corrected, and the corrected electrical parameter data is merged with the photovoltaic voltage fluctuation sequence according to a unified time base to generate a multi-source synchronous dataset.

[0069] Specifically, such as Figure 2 , 4 As shown, the frequency domain conversion module includes:

[0070] The data timing extraction submodule acquires voltage timing data from the multi-source synchronous dataset, arranges the channel voltage sampling points in time and unifies the sampling step size, calculates the sampling interval difference and corrects the deviation, performs channel time synchronization correction, and generates voltage timing matrix values.

[0071] Call the previously generated multi-source synchronization dataset containing photovoltaic voltage and electrical parameters, and extract the data from it. The raw voltage channel data was used to identify the data contained within the dataset. A non-uniformly distributed sampling point is rearranged according to the time stamp in ascending order to establish an initial time series. Set standard sampling step size Calculate the time interval difference between adjacent sampling points. Taking the first three items of the sequence as an example, assuming the original timestamp is... , , Then calculate the interval. , Calculate the deviation between the interval and the standard step size. , and thus , The permissible deviation threshold for time synchronization correction is set as follows: Judge the above and All values ​​exceed the threshold range, requiring interpolation correction to construct a standard time axis. ,against The voltage value at time t is obtained using a linear interpolation formula. Assuming Voltage at any moment , Voltage at any moment Substitute into numerical calculation Correction voltage at the location Similarly, for Correction is performed continuously, traversing channel data to complete full-time correction, as shown in Table 2. Table 2 lists a comparison between some original data and corrected time-series data. The corrected voltage data is then reorganized into rows and columns according to channel order (e.g., photovoltaic voltage channel, mains voltage channel), constructing a dimension of... The matrix structure generates voltage timing matrix values.

[0072] Table 2: Voltage Channel Timing Correction Comparison Table

[0073] Serial Number Original timestamp (s) Original voltage value (V) Standard timeline (s) Correction voltage value (V) 1 10.000 220.00 10.000 220.00 2 10.012 220.60 10.010 220.50 3 10.021 220.96 10.020 220.92 4 10.033 219.80 10.030 220.09 5 10.040 219.50 10.040 219.50

[0074] As shown in Table 2, this table illustrates the process of mapping the original non-uniform sampled data onto the standard 0.01s step time axis through an interpolation algorithm, correcting the jitter caused by transmission delay.

[0075] The frequency component calculation submodule calls the voltage time series matrix value, performs a Fourier transform on the time series, extracts the real and imaginary parts of the complex number and calculates the amplitude energy, counts the intensity by frequency index and normalizes it, and generates a frequency intensity distribution sequence.

[0076] The length of a set of mains voltage channels read from the above voltage timing matrix values ​​is... Time series data Perform a Discrete Fourier Transform (DFT) on it, setting the frequency resolution to . For each frequency component, an operation is performed to project the time-domain signal onto the unit circle in the complex plane. By accumulating the product of the input signal and the sine and cosine basis functions, the signal is decomposed into a complex form. Fundamental frequency (corresponding index) For example, the calculated complex number result is: Extract its actual part With the imaginary part The amplitude energy at this frequency point is calculated using the Pythagorean theorem, which involves calculating the squares of the real and imaginary parts respectively and substituting them into the numerical values ​​to obtain the result. Similarly, calculate the third harmonic. The complex number at the location is Its amplitude energy Statistically analyze the intensity values ​​under the frequency index and select the fundamental amplitude. As a normalization benchmark value Normalization is performed on the frequency intensity, that is, the amplitude at the frequency point is divided by the reference value to calculate the result. Normalized intensity at ,calculate Normalized intensity at The frequency points are calculated. Arrange them in ascending order of frequency to construct a string of length. (correspond A numerical sequence (including the range of higher harmonics) is used to generate a frequency intensity distribution sequence.

[0077] The threshold screening feature generation submodule calculates the mean intensity and fluctuation deviation based on the frequency intensity distribution sequence, sets an adaptive intensity threshold, removes low intensity frequency points and reconstructs the effective frequency set, calculates the weighted average value, and generates the frequency domain features of the photoelectric signal.

[0078] Load the above frequency intensity distribution sequence Before extracting the low-frequency band The intensity values ​​of key frequency points are used as the sample set:

[0079] ;

[0080] Calculate the intensity mean of this sample set. Calculate the variance Obtain the median value After sorting the set, take the middle value. Calculate energy concentration , defined as the ratio of the sum of the first three largest peaks to the total sum, i.e. Set the basic threshold coefficient Using the formula The formula introduces energy concentration by setting an adaptive intensity threshold. As a regulating factor, when energy is highly concentrated ( Approaching 1). As the threshold approaches 0, it moves closer to the median value to preserve weak harmonics; this is then used in numerical calculations. , put the sequence into and Compare and remove those smaller than Low intensity frequency points (such as) (etc.), reconstructing the effective frequency set The corresponding frequencies are respectively Set weight vector Calculate the weighted average The result indicates that the frequency domain composite eigenvalue after screening and weighting is... This reflects the energy distribution of the signal at the main frequency and major harmonics, generating the frequency domain characteristics of the photoelectric signal.

[0081] Specifically, such as Figure 2 , 5 As shown, the fault identification module includes:

[0082] The frequency domain input construction submodule calls the frequency domain features of the photoelectric signal, performs segmented sampling, filters out noise from the segmented signals and delineates the effective interval according to the time window, performs normalization adjustment on the frequency domain amplitude value, and generates the frequency domain input matrix of the photoelectric signal.

[0083] The dataset containing the frequency domain characteristics of photoelectric signals generated previously is called, especially the frequency intensity distribution sequence and effective frequency set calculated above. The full-band analysis range is set to... to Execute segmented sampling logic to divide the frequency band into Each frequency sub-interval has an equal width, and the bandwidth of each sub-interval is set to... , in order Iterate through each frequency sub-interval and extract the amplitude value of discrete frequency points within that interval, for example, in There exists an amplitude within the interval. The frequency components are used to set the noise filtering threshold. The threshold Based on statistical analysis of the noise floor level of photovoltaic sensors at night when there is no light and no streetlight load, the measured average noise floor value is [value missing]. ,set up For the collected amplitude values Perform a filtering decision; if Then let Otherwise, keep In order to eliminate such We eliminate weak background noise, define the effective interval based on the time window, and set the effective diagnostic time window. This refers to the period after the streetlights have entered a stable lighting state, i.e. to Only frequency domain data within the specified time period is extracted, filtering out transient fluctuations during light switching. The filtered frequency domain amplitude values ​​are then normalized using a minimum-maximum normalization formula. The advantage of this formula is that it eliminates differences in data units, mapping amplitudes from different frequency bands to a unified range to improve the consistency of subsequent weight calculations. , (Reserved) (overvoltage margin), substituted into the aforementioned fundamental amplitude value calculate Substitute the third harmonic amplitude calculate The normalized frequency band feature values ​​are reorganized according to time order and frequency band index to construct a matrix with the number of rows equal to the number of time sampling points and the number of columns equal to the number of columns. A two-dimensional matrix structure with (number of frequency bands) is used to generate the frequency domain input matrix of photoelectric signals.

[0084] The frequency domain feature extraction submodule calculates the energy weighting value of the frequency components and extracts the main frequency component based on the frequency domain input matrix of the photoelectric signal, calculates the energy ratio of the frequency band and calculates the amplitude change rate, performs linear smoothing correction on the feature values, and obtains the frequency component feature vector set.

[0085] Extracting a specific moment based on the frequency domain input matrix of the photoelectric signal. row vectors Define the frequency component energy weight vector The weighting is set based on the reciprocal of the frequency band's sensitivity to faults. (fundamental wave), (The frequency band containing the third harmonic), the rest are The energy-weighted value calculation is performed by multiplying the square of the frequency band amplitude by its corresponding weight, summing the results, and then substituting them into the numerical calculation. Extract the dominant frequency component, i.e., the recognition vector. The largest element in the set and its index are used to determine the index. (correspond () is the maximum value Calculate the energy percentage of each frequency band, and the ratio of the energy in the low-frequency band (first three intervals) to the total energy. Assuming the sum is The sum of the first three terms is ,but Calculate the rate of change of amplitude Call the previous time step main frequency amplitude Set sampling interval Execute calculation Linear smoothing correction is applied to the eigenvalues ​​using a first-order lag filter formula. ,in Set as the smoothing coefficient. To preserve the main trend of change while suppressing high-frequency jitter, the rate of change characteristic is smoothed. It is assumed that the smoothing value at the previous time step is... ,but The weighted energy obtained from the above calculation , main frequency amplitude Energy percentage and the smoothed rate of change By combining these elements, we obtain the frequency component feature vector set.

[0086] The similarity calculation and judgment submodule calls the fault baseline data based on the frequency component feature vector set, compares the corresponding frequency band features using the cosine similarity algorithm, calculates the similarity score, performs weighted descending sorting, selects the fault type at the top of the sort, and generates preliminary fault diagnosis information.

[0087] Based on the frequency component feature vector set The system calls a preset fault baseline database, which contains standard feature vectors for various typical fault modes, as shown in Table 3. The feature vector corresponding to "Drive power supply capacitor aging" in the table is selected. The comparison is performed using the cosine similarity algorithm formula:

[0088] ;

[0089] The advantage of the similarity calculation formula is that it is only sensitive to the vector direction and not to the absolute value, thus adapting to the normalization characteristics of streetlights with different power levels. Substituting the values ​​into the formula, we can calculate the numerator dot product:

[0090] ;

[0091] calculate Modulus length:

[0092] ;

[0093] calculate Modulus length:

[0094] ;

[0095] Final similarity calculation:

[0096] ;

[0097] Similarly, calculate the fault vector for "LED module open circuit". The similarity is approximately Perform a weighted descending sort, and the result is: Select the first one in the sorting. The corresponding "drive power supply capacitor aging" type indicates that the current monitoring data highly matches the characteristics of capacitor aging faults, exceeding the set diagnostic threshold. This generates preliminary fault diagnosis information.

[0098] Table 3: Reference Table for Fault Characteristic Base Values

[0099] Serial Number Fault type name Energy weighting (E) Main frequency amplitude (A) Energy percentage (R) Rate of change (Δ) 1 Normal operating status 0.8500 1.0000 0.9900 0.0500 2 Aging of drive power supply capacitors 0.7000 0.8200 0.9600 0.4000 3 LED module short circuit 1.2000 1.1500 0.9500 0.8000 4 Rectifier diode breakdown 0.5000 0.6000 0.7000 0.2000

[0100] As shown in Table 3, the table lists the standard values ​​of frequency domain characteristics under four typical conditions preset by the system, which are used as the reference vector for cosine similarity calculation. Among them, "driving power supply capacitor aging" is characterized by a moderate decay of energy and main frequency amplitude and an increase in the rate of change.

[0101] Specifically, such as Figure 2 , 6 As shown, the coupling verification module includes:

[0102] The signal power analysis submodule calls the preliminary fault diagnosis information, obtains the current and voltage signals at the street light end, aligns the current and voltage signals according to the timestamp, calculates the instantaneous power and maps it to the diagnostic information, and generates an instantaneous power mapping matrix.

[0103] The generated preliminary fault diagnosis information is retrieved, and the fault type judgment result "drive power supply capacitor aging" ranked first is extracted, along with its associated fault occurrence timestamp. Initiate a data retrieval command to obtain the timestamp before and after from the multi-source synchronized dataset. Minutes of raw current signal at the streetlight end With voltage signal The sampling frequency is set to A total of For each data point, perform timestamp alignment, iterate through the current and voltage data queues, and check the acquisition time stamp of each data set. and Then, using the voltage time marker as a reference, the current data is interpolated and corrected to ensure... After alignment is completed, for each synchronous sampling point Using the formula Calculate instantaneous power, select the first The data from each sampling point were processed to measure the voltage value at that moment. Current value Substituting into the calculation, we get The multiplication operation is performed sequentially on the sampling points to obtain a line containing A continuous curve of power values ​​is then performed, followed by a mapping operation to establish a dimension of The matrix structure, where To determine the number of sampling points, the five columns are defined as "timestamp," "voltage value," "current value," "instantaneous power value," and "diagnostic label," respectively. The "drive power supply capacitor aging" information in the aforementioned fault diagnosis information is encoded as a numerical label. Fill up to the fifth column of the matrix, for example, for... The row data, filled with the following content As shown in Table 4, Table 4 lists some key data of the generated instantaneous power mapping matrix. Through this process, physical electrical parameters and fault logic conclusions are strongly correlated at the data level to generate the instantaneous power mapping matrix.

[0104] Table 4: Data fragments of the instantaneous power mapping matrix

[0105] Index number Timestamp (s) Voltage value (V) Current value (A) Instantaneous power (W) Diagnostic label 1 22:15:00.01 219.80 0.452 99.35 2 2 22:15:00.02 219.75 0.451 99.11 2 3 22:15:00.03 219.82 0.453 99.58 2 4 22:15:00.04 219.60 0.450 98.82 2 5 22:15:00.05 219.65 0.451 99.06 2

[0106] As shown in Table 4, this table displays the electrical parameter matrix after time alignment and calculation. The diagnostic label "2" clearly points to the aging fault of the drive power supply capacitor. The instantaneous power of each row is the product of the voltage and current at the corresponding moment.

[0107] The power anomaly determination submodule calls the instantaneous power mapping matrix to calculate the average instantaneous power over multiple time periods. It then compares the average value with a set deviation threshold. If the average value is greater than the deviation threshold, the time interval is marked as a power anomaly, and a power anomaly interval set is generated.

[0108] Call the instantaneous power mapping matrix and set the time sliding window length to... (i.e., includes) (number of data points), step size is The matrix data is scanned in segments, and the first window is extracted. The data within is processed using the formula:

[0109] ;

[0110] Calculate the average instantaneous power over multiple time periods and substitute it into... Assuming the window contains The sum of the sums is Then calculate The logic for determining the power deviation threshold first retrieves the rated power curve of the street light, where the rated power of the street light is... Retrieve the past data of this street light The statistical results of the original operating power fluctuation under normal operating conditions were used to calculate the average normal power. The standard deviation is ,in accordance with The principle sets an allowable fluctuation range, i.e., a lower threshold. Upper limit threshold The average value calculated for the current window Compare with the deviation threshold and execute the judgment logic: If or If it does not, it is considered abnormal. In this example, The judgment result is true, meaning the current power is significantly lower than the lower limit of the normal operating range. This is consistent with the characteristics of power factor decline and active power loss caused by capacitor aging. This time interval is marked. The power anomaly interval is defined and the anomaly type is recorded as "low power anomaly". The sliding window continues to the next interval, and the above calculation and comparison process is repeated. The continuous time periods that are judged to be abnormal are merged to generate a power anomaly interval set.

[0111] The communication anomaly verification submodule calls the power anomaly interval set, calculates the signal transmission delay within the interval, compares the delay threshold with the message sending and receiving time, and if the delay is greater than the set communication delay threshold, it is marked as a communication anomaly, matches the power anomaly interval, and generates a street light fault confirmation result.

[0112] Call a specific interval from the power anomaly interval set Access the communication log database of the street light controller, extract the status report message records sent within the specified time period, and retrieve a total of [number missing] records. From the given messages, select one key message and record its device-side sending timestamp. and the timestamp of the cloud server receiving the message. Using the formula Calculate the signal transmission delay and substitute it into the numerical calculation. A reference standard for setting the communication latency threshold should be established based on the NB-IoT communication protocol standard. Under normal coverage levels, the end-to-end latency should be less than [a certain value]. Based on the statistical results of message transmission performance of the street light control network in this area, the average latency over the past week was [data missing]. The maximum delay is Set communication delay threshold For the maximum delay This eliminates the possibility of regular network jitter, i.e. The calculated result and Perform a comparison and execute the judgment logic: If If the condition is met, it is marked as a communication error; otherwise, it is marked as a normal communication. In this example,... The system determined that the communication status was normal. This result indicates that the power data monitored in the previous steps was not affected by communication packet loss or high latency, and the real-time performance and integrity of the data are reliable. Based on this, the system matched the power anomaly interval. Since the communication verification passed (normal), the system confirmed that the power anomaly determination was valid, eliminating the probability of "data alignment error caused by communication delay" or "mean calculation error caused by packet loss". Finally, it was confirmed that the anomaly was caused by a physical component failure of the street light. Combined with the aforementioned diagnostic tags, a comprehensive report was generated containing "fault type: driver power supply capacitor aging", "fault time: 22:15:00", "power status: low power (99.2W)" and "communication status: normal". This result indicates that the street light did indeed experience hardware-level performance degradation, rather than a false alarm caused by the network environment, and a street light fault confirmation result was generated.

[0113] Specifically, such as Figure 2 , 7 As shown, the maintenance scheduling module includes:

[0114] The device positioning and analysis submodule obtains the street light fault confirmation result, detects the device number and location coordinate format, indexes the coordinate mapping table according to the device number, calculates the coordinate difference to determine the offset range, corrects abnormal coordinate points, and generates location coordinate correction values.

[0115] Obtain the generated street light fault confirmation result and extract the key identification information, namely the device's unique serial number. Simultaneously, it extracts the real-time Global Positioning System (GPS) coordinate data reported by the device during the previous communication handshake. First, a format compliance check is performed, which uses regular expression logic to check whether the longitude value is located within the specified range. Within the interval, are the latitude values ​​located at... Within the interval, and whether the precision of the decimal places meets at least After confirming the format is correct, determine the location requirements based on the device number. The system uses a pre-built Geographic Information System (GIS) coordinate mapping table to retrieve the high-precision reference coordinates entered during the installation and deployment phase of the equipment. The coordinate difference calculation is performed, using the Euclidean distance approximation algorithm (for short-distance planar projection) to calculate the spatial distance between the reported coordinates and the reference coordinates, and the longitude difference is calculated separately. Latitude difference Set the conversion factor for converting latitude and longitude to meter-level distance, where the longitude conversion factor is... (Corresponding to around 39 degrees North latitude), latitude conversion factor Substitute the numerical values ​​to calculate the actual physical offset:

[0116] ;

[0117] The calculation process is as follows Determine the offset range and set the first-level threshold for position drift determination. and secondary threshold This threshold setting is based on the statistical value of the average positioning accuracy of civilian GPS modules in urban canyon environments. If it is determined to be a normal fluctuation; If it is determined to be a slight drift; The positioning was determined to be invalid in this example. The fluctuations are within the acceptable range, but to ensure the accuracy of maintenance work orders, abnormal coordinate points need to be corrected. The system executes forced repositioning logic, discarding reported coordinates with fluctuations. Directly using the reference coordinates As the final work point, if the offset in the example is Then, the same repositioning operation is performed to eliminate drift interference, and the final coordinate data is obtained. Format as a standard WGS-84 string to generate position coordinate correction values.

[0118] The job matching submodule, based on the location coordinate correction value, calls the fault type corresponding to the equipment number, compares the fault type with the job type mapping table, calculates the job level parameters and performs a difference comparison, selects the corresponding job type number based on the difference consistency result, and generates job type matching parameters.

[0119] Based on the location coordinate correction value, the geographical area is determined to be "Dongcheng District, Section A". The device number is then retrieved. The corresponding fault type determination result is "Drive power supply capacitor aging" (fault code). The system compares and retrieves fault codes and standardized work procedures based on a fault type-job type mapping table, as shown in Table 5. Table 5 lists the correspondence between the system's preset fault codes and standardized work procedures. The retrieved corresponding job type number is... (Power module replacement operation) Calculate the operation level parameter, which is determined by the fault urgency weight. Technical complexity weight and environmental risk weights The weighted summation is obtained, and the weights are assigned as follows: For the "capacitor aging" fault, an urgency score is set based on the original records in the operation and maintenance database. (Because the streetlights can still be turned on, they are not completely extinguished), technical complexity rating. (Requires live operation or dismantling at height), Environmental risk assessment (Regular road section), perform calculations Substitute into numerical calculation Perform a difference comparison and calculate the results. The work was compared against a preset work classification standard, which was set as: Level 1 work. (Simplified Inspection), Level 2 Operation (Routine maintenance), Level 3 operation (Emergency repair), judgment If the job falls within the Level 2 work zone, based on the consistency of discrepancies, the system checks the currently waiting maintenance team qualification database and confirms the team number as follows. The work team possesses a Level II electrician qualification (qualification points). If the resources match, select the corresponding job type number. And add a level suffix Generate job type matching parameters.

[0120] Table 5: Mapping Table of Fault Types and Operation Types

[0121] Serial Number Fault Codes Fault Description Job type number Basic Description of the Assignment 1 01 LED module open circuit M_OPT_01 Complete replacement of the light source module 2 02 Aging of drive power supply capacitors M_ELEC_02 Power driver replacement and debugging 3 03 Communication module offline M_COMM_03 NB-IoT module reset or replacement 4 04 abnormal tilt of the light pole M_MECH_04 Mechanical structure reinforcement and correction

[0122] As shown in Table 5, this table establishes a detailed logical mapping from fault symptoms to specific maintenance actions, ensuring the automation and standardization of work order generation. The job type number directly determines the retrieval of the subsequent bill of materials.

[0123] The work order generation submodule matches parameters based on the job type, calls up equipment number, location coordinates and fault type data, concatenates fields and adds a timestamp, calculates the combined integrity check value, and generates a maintenance work order dataset.

[0124] Match parameters based on job type Call device number Position coordinate correction value The fault type data "Drive power supply capacitor aging" is used to perform field concatenation operations to construct a standardized JSON format data packet: {"DevID":"10205088","Loc":"116.397500,39.908700","Fault":"02","Type":"LV2"}, with a timestamp appended to obtain the current system time. Convert it to Unix timestamp format The data is added to the end of the data packet, and the integrity check value is calculated. To prevent the work order data from being tampered with or lost during transmission to the dispatch system, a weighted character summation and modulo algorithm is used to extract the last four digits of the device number. Extract fault codes Extract the last four digits of the timestamp. According to the formula The advantage of this algorithm lies in its ability to amplify the impact of minute data changes on the verification value through non-linear weighting, which is then substituted into the numerical calculation. Perform division operation Therefore, the remainder is , check value Write the header field of the work order. This result indicates that the integrity fingerprint of the work order data has been generated. Any change in any key number will cause the check value to mismatch. Finally, encapsulate the fields to generate the maintenance work order dataset.

[0125] The above description is merely a specific embodiment of the present invention, but the scope of protection of the present invention is not limited thereto. Any variations or substitutions that can be easily conceived by those skilled in the art within the technical scope disclosed in the present invention should be included within the scope of protection of the present invention. Therefore, the scope of protection of the present invention should be determined by the scope of the claims.

Claims

1. A smart street light fault diagnosis and maintenance system, characterized in that, The system includes: The signal collaborative acquisition module synchronously acquires voltage timing data, power line current and voltage phase data, and communication link transmission delay characteristics through photovoltaic sensors. It eliminates ambient light fluctuation interference and performs timestamp alignment to generate a multi-source synchronous dataset, which is then transmitted to the frequency domain conversion module. The signal collaborative acquisition module includes: The photovoltaic signal acquisition submodule acquires the output voltage signal stream of the photovoltaic sensor, detects continuous time points of the sampled voltage, calculates the rate of change of light intensity based on the amplitude difference of the time points, calculates the voltage fluctuation, filters out abnormal points, and generates photovoltaic voltage fluctuation sequence values. The electrical parameter synchronous measurement submodule synchronously acquires current and voltage signals based on the photovoltaic voltage fluctuation sequence value, analyzes the signal phase difference, compares it with the phase reference value, calculates the instantaneous power distribution, and obtains the current and voltage phase difference coefficient by comparing the time difference between the photovoltaic fluctuation sequence and the power distribution. The timing alignment and fusion submodule extracts communication link delay features based on the current-voltage phase difference coefficient, monitors and corrects time stamp differences, filters out abnormal delay data based on the weighted average value, and generates a multi-source synchronization dataset. The frequency domain conversion module calls the multi-source synchronous dataset, extracts voltage time-series data, performs Fourier transform to calculate frequency components and intensity values, adaptively sets an intensity threshold to filter effective components, calculates the average intensity value, generates photoelectric signal frequency domain features, and transmits them to the fault identification module. The fault identification module receives the frequency domain features of the photoelectric signal, extracts the frequency component values, calculates the similarity score with the fault benchmark value using a cosine similarity algorithm, selects the first fault type in descending order, generates preliminary fault diagnosis information, and transmits it to the coupling verification module. The coupling verification module calls the preliminary fault diagnosis information, calculates the instantaneous power comparison deviation threshold based on current and voltage to mark power abnormalities, checks if the transmission delay exceeds the delay threshold to mark communication abnormalities, verifies the matching relationship of the fault period, generates street light fault confirmation results, and transmits them to the maintenance scheduling module. The coupling verification module includes: The signal power analysis submodule calls the preliminary fault diagnosis information to obtain the current and voltage signals at the street light end, aligns the current and voltage signals according to the timestamp, calculates the instantaneous power and maps it to the diagnostic information, and generates an instantaneous power mapping matrix. The power anomaly determination submodule calls the instantaneous power mapping matrix to calculate the average instantaneous power over multiple time periods. It then compares the average value with a set deviation threshold. If the average value is greater than the deviation threshold, the time interval is marked as a power anomaly, and a power anomaly interval set is generated. The communication anomaly verification submodule calls the power anomaly interval set, calculates the signal transmission delay within the interval, compares the delay threshold with the message sending and receiving time, and if the delay is greater than the set communication delay threshold, it is marked as a communication anomaly, matches the power anomaly interval, and generates a street light fault confirmation result.

2. The intelligent street light fault diagnosis and maintenance system according to claim 1, characterized in that, The multi-source synchronization dataset includes voltage timing data, current phase data, voltage phase data, communication transmission delay features, and timestamp alignment information. The photoelectric signal frequency domain features include frequency components, average signal strength, strength threshold, and effective frequency components. The preliminary fault diagnosis information includes fault type, similarity score, and fault priority sequence. The street light fault confirmation result includes power anomaly markers, communication anomaly markers, and fault time period matching information.

3. The intelligent street light fault diagnosis and maintenance system according to claim 1, characterized in that, The frequency domain conversion module includes: The data timing extraction submodule acquires voltage timing data from the multi-source synchronous dataset, arranges the channel voltage sampling points in time and unifies the sampling step size, calculates the sampling interval difference and corrects the deviation, performs channel time synchronization correction, and generates voltage timing matrix values. The frequency component calculation submodule calls the voltage time series matrix value, performs a Fourier transform on the time series, extracts the real and imaginary parts of the complex number and calculates the amplitude energy, statistically analyzes the intensity by frequency index and normalizes it, and generates a frequency intensity distribution sequence. The threshold filtering feature generation submodule calculates the mean intensity and fluctuation deviation based on the frequency intensity distribution sequence, sets an adaptive intensity threshold, removes low-intensity frequency points and reconstructs the effective frequency set, calculates the weighted average value, and generates the frequency domain features of the photoelectric signal.

4. The intelligent street light fault diagnosis and maintenance system according to claim 3, characterized in that, The intensity threshold is dynamically corrected based on the mean, variance, and median of the frequency intensity distribution sequence, combined with energy concentration and the difference between local peak values ​​and global average values.

5. The intelligent street light fault diagnosis and maintenance system according to claim 1, characterized in that, The fault identification module includes: The frequency domain input construction submodule calls the frequency domain features of the photoelectric signal, performs segmented sampling, filters out noise from the segmented signals and delineates the effective interval according to the time window, performs normalization adjustment on the frequency domain amplitude value, and generates the photoelectric signal frequency domain input matrix. The frequency domain feature extraction submodule calculates the frequency component energy weighting value and extracts the main frequency component based on the frequency domain input matrix of the photoelectric signal, calculates the frequency band energy ratio and calculates the amplitude change rate, performs linear smoothing correction on the feature value, and obtains the frequency component feature vector set. The similarity calculation and determination submodule, based on the frequency component feature vector set, calls the fault baseline data, uses the cosine similarity algorithm to compare the corresponding frequency band features, calculates the similarity score, performs weighted descending sorting, selects the fault type at the top of the sort, and generates preliminary fault diagnosis information.

6. The intelligent street light fault diagnosis and maintenance system according to claim 1, characterized in that, The set deviation threshold is a power deviation threshold, which is determined based on the street light rated power curve and the statistical results of the original operating power fluctuation. The communication delay threshold is determined based on the communication protocol standard and message transmission performance statistics of the street light control network.

7. The intelligent street light fault diagnosis and maintenance system according to claim 1, characterized in that, The maintenance scheduling module receives the street light fault confirmation result, obtains the equipment number and location coordinates, matches the fault type with the corresponding maintenance operation type, and integrates the equipment number, location coordinates, fault type, operation type and timestamp to generate a maintenance work order. The maintenance work order includes the equipment number, location coordinates, fault type, job type, and timestamp.

8. The intelligent street light fault diagnosis and maintenance system according to claim 7, characterized in that, The maintenance scheduling module includes: The device positioning and analysis submodule obtains the street light fault confirmation result, detects the device number and location coordinate format, indexes the coordinate mapping table according to the device number, calculates the coordinate difference to determine the offset range, corrects abnormal coordinate points, and generates a location coordinate correction value. The job matching submodule, based on the location coordinate correction value, calls the fault type corresponding to the equipment number, compares the fault type with the job type mapping table, calculates the job level parameters and performs a difference comparison, selects the corresponding job type number based on the difference consistency result, and generates job type matching parameters. The work order generation submodule, based on the job type matching parameters, calls the equipment number, location coordinates, and fault type data, concatenates the fields and adds a timestamp, calculates the combined integrity check value, and generates a maintenance work order dataset.

Images