Remote monitoring method and system for state of LED emergency indication board

Abstract

The invention relates to the technical field of public safety monitoring, and discloses a remote state monitoring method and system for an LED emergency indicator, and the method comprises the steps: obtaining the historical operation data and environment parameters of the indicator, and carrying out the data cleaning, and obtaining a standardized historical performance data set; calculating an aging index and predicting a fault probability to obtain an equipment risk score; fusing the equipment risk score with the real-time environment change data, and obtaining a dynamic risk level through weighted calculation; carrying out risk level classification based on the dynamic risk level and allocating a reporting frequency interval to form a preliminary frequency allocation scheme; performing communication load simulation on the scheme, optimizing resource allocation by adjusting the frequency of the low-risk equipment, and generating an optimized frequency allocation scheme; and finally, issuing a configuration instruction, confirming an execution state and generating a final reporting mechanism. According to the method, accurate and dynamic risk assessment and resource adaptive scheduling of the state of the emergency indication board can be realized.

Description

Technical Field

[0001] This invention relates to the field of public safety monitoring technology, and in particular to a method and system for remotely monitoring the status of LED emergency signs. Background Technology

[0002] Currently, in the fields of modern urban management and public safety, monitoring the operational status of emergency signs is particularly crucial. These devices are widely used in transportation, construction, and public places; any malfunction could directly impact evacuation and safety, making their stability and reliability an indispensable factor.

[0003] In one existing technology, the monitoring center typically uses wireless networking technology to communicate with remote LED signs. Existing systems usually set a fixed period or uniform standard for all devices. The signs automatically collect their operating status at this fixed interval (e.g., every 24 hours) and package the data for centralized analysis at the monitoring center. This method ignores the differences in the environments in which different devices operate and the diversity of their historical performance, making it impossible for the system to tailor the appropriate status reporting frequency based on the real-time dynamic risk level of each device (combining environmental changes and historical data).

[0004] In summary, existing technologies suffer from an imbalance between monitoring resources and efficiency due to rigid monitoring frequency settings. Summary of the Invention

[0005] This invention provides a method and system for remote monitoring of the status of LED emergency signs, in order to solve the problem of imbalance between monitoring resources and efficiency caused by rigid monitoring frequency settings.

[0006] Firstly, in order to solve the above-mentioned technical problems, the present invention provides a method for remote monitoring of the status of LED emergency signs, comprising: Obtain historical operating data and current environmental parameters of emergency signs, and perform data cleaning on the historical operating data and current environmental parameters to obtain a standardized historical performance dataset; Based on the standardized historical performance dataset, the aging index of the emergency sign is calculated, and the failure probability is predicted based on the aging index to obtain the equipment risk score. Based on the equipment risk score and the real-time environmental change data, a fusion and weighted calculation is performed to obtain the updated dynamic risk level; Based on the updated dynamic risk level, risk level classification and frequency interval allocation are performed to obtain a preliminary frequency allocation scheme. Based on the preliminary frequency allocation scheme, communication load simulation and frequency value adjustment are performed to obtain an optimized frequency allocation scheme. Based on the optimized frequency allocation scheme, configuration commands are issued and execution status is confirmed to generate the final reporting mechanism.

[0007] Secondly, the present invention provides a remote monitoring system for the status of LED emergency signs, comprising: The data acquisition and processing module is used to acquire historical operating data and current environmental parameters of emergency signs, and to perform data cleaning on the historical operating data and current environmental parameters to obtain a standardized historical performance dataset. The risk assessment module is used to calculate the aging index of the emergency sign based on the standardized historical performance dataset, and to predict the failure probability based on the aging index to obtain the equipment risk score. The dynamic calculation module is used to perform fusion and weighted calculation based on the device risk score and real-time acquired environmental change data to obtain the updated dynamic risk level; The frequency allocation module is used to classify risk levels and allocate frequency ranges based on the updated dynamic risk level to obtain a preliminary frequency allocation scheme. The load optimization module is used to perform communication load simulation and frequency value adjustment based on the preliminary frequency allocation scheme to obtain an optimized frequency allocation scheme. The instruction execution and feedback module is used to issue configuration instructions and confirm execution status according to the optimized frequency allocation scheme, and generate the final reporting mechanism.

[0008] Compared with the prior art, the present invention has the following beneficial effects: (1) This invention calculates the equipment risk score by acquiring historical operating data and current environmental parameters, and determines the dynamic risk level by integrating real-time environmental change data. This method combines the historical aging performance of the equipment with the real-time environmental impact, so that the risk assessment is no longer a static analysis based on a fixed period, but a dynamic assessment for each piece of equipment and each moment, thereby significantly improving the accuracy of the sign fault warning.

[0009] (2) This invention classifies dynamic risk levels using a decision tree algorithm and allocates shorter reporting frequency intervals for high-risk levels. This approach changes the rigid mode in the prior art where all devices report using a uniform standard. It can customize the reporting strategy according to the actual risk level of each device, so that monitoring resources can be prioritized for the high-risk devices that need the most attention, thereby achieving a reasonable allocation of monitoring resources and improving overall monitoring efficiency.

[0010] (3) This invention achieves optimized allocation by simulating communication load tests on the initial frequency scheme and reducing the frequency value of low-risk devices when resource limits are exceeded. This step adds a load balancing closed loop to the system. While ensuring that high-risk devices are monitored first, it releases resources by dynamically adjusting the frequency of low-risk devices, avoiding overload of system communication and computing resources due to high-frequency reporting. This ensures the stable operation of the system under efficient early warning and avoids resource conflicts. Attached Figure Description

[0011] Figure 1 This is a schematic flowchart of a remote monitoring method for the status of an LED emergency sign provided in the first embodiment of the present invention; Figure 2 This is a schematic diagram of the structure of a remote monitoring system for the status of an LED emergency sign provided in the second embodiment of the present invention. Detailed Implementation

[0012] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0013] Reference Figure 1 The first embodiment of the present invention provides a method for remotely monitoring the status of LED emergency signs, including the following steps: S11, acquire historical operating data and current environmental parameters of emergency signs, and perform data cleaning on the historical operating data and current environmental parameters to obtain a standardized historical performance dataset; S12, Calculate the aging index of the emergency sign based on the standardized historical performance dataset, and predict the failure probability based on the aging index to obtain the equipment risk score; S13, Based on the equipment risk score and the real-time acquired environmental change data, perform fusion and weighted calculation to obtain the updated dynamic risk level; S14. Based on the updated dynamic risk level, risk level classification and frequency range allocation are performed to obtain a preliminary frequency allocation scheme. S15, Based on the preliminary frequency allocation scheme, perform communication load simulation and frequency value adjustment to obtain an optimized frequency allocation scheme; S16, Based on the optimized frequency allocation scheme, configuration instructions are issued and execution status is confirmed to generate the final reporting mechanism.

[0014] In step S11, historical operating data and current environmental parameters of the emergency signs are acquired, and the historical operating data and current environmental parameters are cleaned to obtain a standardized historical performance dataset, including: Based on the historical operating data and current environmental parameters, outlier identification and removal are performed to obtain a clean dataset; Based on the clean dataset, format unification and unit normalization are performed to obtain an intermediate dataset; Based on the intermediate dataset, integrity checks and missing data interpolation are performed to obtain a standardized historical performance dataset.

[0015] In one implementation, the emergency sign uses its internal sensors (for monitoring historical operating data, such as brightness, operating voltage, and operating current) and external sensors (for monitoring current environmental parameters, such as ambient temperature and humidity) to capture its status at fixed time intervals (e.g., every 5 minutes). The captured data is packaged and transmitted to a central data server via a sensor network through an onboard wireless communication module (e.g., a LoRa module). The server receives and stores this time-series data, forming a raw dataset. Historical operating data refers to the time-series data of the device's own status collected and stored periodically (e.g., every 5 minutes), with an example data structure of (Device ID, Timestamp, Brightness Value, Operating Voltage, Operating Current, Display Content Switching Frequency). Current environmental parameters refer to environmental sensor data collected at the same timestamp as the historical operating data, with an example data structure of (Device ID, Timestamp, Ambient Temperature, Ambient Humidity, Ambient Light Intensity).

[0016] After obtaining the raw dataset, data cleaning is performed. First, outliers are identified and removed based on the historical operating data and current environmental parameters to obtain a clean dataset. For specific numerical features (e.g., "operating voltage"), the system calculates its mean value within a preset time window. ) and standard deviation ( The preset time window is set based on the periodicity of the data. This periodicity is determined by performing autocorrelation analysis on historical temperature data (e.g., data collected over the past year). This analysis calculates different time lags (...). (k=1 represents a lag of 5 minutes, k=2 represents a lag of 10 minutes, and so on); then, the Pearson correlation coefficient between the original temperature data sequence and its sequence lagged by k time steps is calculated; finally, the correlation coefficient values ​​corresponding to different lag amounts k are plotted. If the correlation coefficient is observed to be at k=288 (i.e., 24 hours) in the plot... If a significant peak (e.g., correlation coefficient greater than 0.5) appears at positions where 12 times / hour = 288 and its integer multiples (e.g., k = 576, k = 864), it confirms the existence of a 24-hour main period in the data. Based on this analysis, the preset time window can be set to an integer multiple of this main period (e.g., 30 times, i.e., 30 days) to ensure the accuracy of the calculation. and The statistical values ​​are stable and can represent multiple complete work cycles. If a certain characteristic value x satisfies... or If the value is not found, it is considered an outlier and is replaced with a NULL (empty) identifier in the dataset, thus creating a clean dataset.

[0017] Next, based on the clean dataset, format unification and dimension normalization are performed to obtain an intermediate dataset. The format unification operation converts all data timestamps to a standard format (e.g., Unix timestamps) and unifies the numerical data types. Dimension normalization uses the min-max normalization method, and its calculation formula is as follows: ; Among them, the physical maximum value ( ) and physical minimum ( The range is determined according to the sensor hardware manufacturer's technical specifications. (For example, if the specifications for a certain model of temperature sensor state that its measurement range is -40°C to 85°C, then...) =-40, =85). This setting method ensures the stability and objectivity of the normalization range. If... If the denominator is zero, then... Set to 0.5.

[0018] Finally, based on the intermediate dataset, integrity checks and missing data interpolation are performed to obtain a standardized historical performance dataset. Integrity checks are used to locate NULL values ​​or missing timestamps in the data sequence. The completion operation is performed on these missing points if the missing point is located between two valid consecutive data points. For example, assume time points... value Missing, but of and of If valid, then linear interpolation is used for calculation: ; in If the value is not zero, and the missing data is located at the end of the data sequence, then nearest neighbor padding is used. Specifically, if a NULL value exists at the beginning of the data sequence (e.g., from time point...), then... arrive If data is missing, backfilling is used to fill in the first valid data point ( The value of ) Assigned to If a NULL value exists at the end of the data sequence (for example, from time point...), arrive If data is missing, forward padding is used to fill in the last valid data point ( The value of ) Assigned to After this step is completed, the output standardized historical performance dataset is a time series dataset that is complete, uniformly formatted, and has all numerical features within the range of [0,1], which will be used for analysis in subsequent steps.

[0019] In step S12, based on the standardized historical performance dataset, the aging index of the emergency sign is calculated, and a failure probability prediction is performed based on the aging index to obtain an equipment risk score, including: Based on the standardized historical performance dataset, feature data related to equipment aging are extracted to obtain an aging feature set; Based on the set of aging characteristics, a weighted comprehensive calculation is performed to obtain the quantitative value of the aging index; Based on the quantified value of the aging index, combined with the preset aging threshold, the equipment risk is marked to obtain the equipment risk grouping result including high-risk equipment, medium-risk equipment and low-risk equipment; Based on the equipment risk grouping results and the quantified value of the aging index, combined with pre-stored historical fault data, fault probability prediction is performed to obtain the equipment risk score.

[0020] First, based on the standardized historical performance dataset, feature data related to device aging is extracted to obtain an aging feature set. First, predefined normalized features reflecting the physical degradation state of the device are selected from the dataset, such as cumulative operating time, display response time, and brightness decay rate. The brightness decay rate is calculated by extracting the average brightness value of the device during the initial operating phase (e.g., the first week after installation) from the standardized historical performance dataset, denoted as . ;Then, Extract its average brightness value over the most recent statistical period (e.g., the past 24 hours), and denot it as the brightness decay rate. The calculation formula is as follows: ; Next, based on the set of aging characteristics, a weighted comprehensive calculation is performed to obtain the quantitative value of the aging index. The weighted summation process is as follows: ; in, This is the quantified value of the aging index. It is the i-th feature value in the aging feature set (e.g. ), where n is the total number of features, It is the weight corresponding to the i-th feature. The weight... It is determined in advance by analyzing pre-stored historical fault data. In one implementation, The value is determined by the following formula: ; in This is attributed to the i-th aging characteristic (e.g., "LED bead degradation") The number of failures, This represents the total number of historical failures.

[0021] Then, based on the quantified value of the aging index ( ), combined with the preset aging threshold ( Equipment risk is labeled to obtain equipment risk grouping results that include high-risk and low-risk equipment. The preset aging threshold is set based on all equipment in the historical dataset. Statistical distribution analysis of the values. The threshold is chosen based on selecting a quantile (e.g., the 80th quantile) that historically... Equipment with values ​​above this quantile has a significantly higher subsequent actual failure rate (e.g., the proportion of devices failing within the next 30 days) than equipment with values ​​below this quantile, and the number of high-risk devices screened by this threshold matches the maintenance team's handling capacity. The judgment logic is as follows: If If the equipment risk grouping result is not specified, the equipment risk grouping result is marked as "high-risk group"; if If so, it is marked as "low-risk group".

[0022] Finally, based on the equipment risk grouping results and the quantified values ​​of the aging indicators, combined with pre-stored historical fault data, a fault probability prediction is performed to obtain the equipment risk score. This prediction is accomplished using a pre-trained logistic regression model. The pre-stored historical fault data originates from historical maintenance records and automatically generated fault work orders, and its data structure includes (equipment ID, fault timestamp, fault type). The training process of the model is as follows: First, historical fault data is organized to generate training labels. Specifically, a future prediction window is set ( The future forecast window is set based on the response time and spare parts procurement time of maintenance services; for example, if the average time from the discovery of high-risk equipment to the completion of repairs by maintenance personnel (including spare parts procurement and transit time) is 20 days, then... It must be greater than 20 days. A common setting method is to set it as the mean time to repair plus a safety margin (e.g., 10 days), so it is set to 30 days.

[0023] Subsequently, each historical time point t in the standardized historical performance dataset is traversed, and for the device at that time point, the pre-stored historical fault data is queried to determine whether the device is in [a certain situation]. There are fault records within the time interval; if a fault record exists, the training label y(t) at that time point t is assigned a value of 1; if no fault record exists, y(t) is assigned a value of 0. Then, the quantized value of the aging index at the corresponding time point t is extracted. The risk grouping results (G(t), high risk = 1, low risk = 0) and equipment risk grouping results (G(t), high risk = 1, low risk = 0) were used as training features. These features and labels were used to train the logistic regression model to solve for the model coefficients. ).

[0024] The process of solving the model coefficients (taking gradient descent as an example) is as follows: First, the coefficients are calculated ( First, assign random initial values; second, set a learning rate. .Should The setup involves testing the model on a validation dataset (split from the training set) before training. A set of candidate learning rates (e.g., [0.1, 0.01, 0.001]) is selected, and the model is trained on each rate for a fixed number of iterations (e.g., 100). The loss function value on the validation set is observed. The candidate value that causes the loss function to decrease steadily and at the fastest rate (e.g., 0.01) is chosen as the final learning rate. Then, iterations are performed on the training dataset. The stopping condition for iteration is determined by the maximum number of iterations (…). ) and convergence threshold ( The maximum number of iterations is jointly determined by [various factors]. This maximum number of iterations is a safety limit (e.g., typically set to 5000) to prevent excessively long training times. The convergence threshold ([…]) (For example, it is generally set to) This is used to determine whether the model has converged.

[0025] In each iteration, the predicted probability of all training samples is calculated. The total gradient of the cross-entropy loss function with respect to the true label y(t) (i.e., the loss function with respect to each coefficient) (partial derivatives); then, according to the opposite direction of the gradient, update each coefficient, with the update formula as follows: ; in This represents the total loss. After the update, calculate the difference between the loss value L(t) of the current iteration and the loss value L(t-1) of the previous iteration. Repeat this iterative process until the current iteration number is reached. or At this time ( These are the model coefficients obtained from the solution. During prediction, the current device... The values ​​of G and G are input into the trained model. The calculation formula is as follows: ; Final equipment risk score Calculated as .

[0026] In step S13, based on the equipment risk score and the real-time acquired environmental change data, a fusion and weighted calculation is performed to obtain the updated dynamic risk level, including: A comprehensive risk score is obtained by weighting and fusing the equipment risk score with real-time environmental change data. The comprehensive risk score is compared with a preset fusion threshold. If the score is higher than the fusion threshold, the weight of environmental factors is increased to obtain an adjusted weight ratio. Based on the adjusted weighting ratio, the equipment risk score and the environmental change data are weighted and summed to obtain the updated dynamic risk level.

[0027] First, the real-time acquired environmental change data is quantified to obtain a standardized environmental impact score. This process begins by quantifying the real-time acquired environmental change data (e.g., current temperature). Current humidity ) is quantified to calculate a comprehensive environmental impact score ( ). The quantification process consists of two parts. The first part is to calculate the parameter-specific influence score for each individual environmental parameter. Taking temperature as an example, its temperature influence score is (…). The mapping is determined based on a preset mapping rule. This mapping rule is formulated based on the statistical relationship between environmental deviations and failure rates in historical data. This rule relies on a preset fluctuation threshold (…). The fluctuation threshold is set based on the statistical characteristics of the standardized historical performance dataset in S11, for example, it is set to the historical standard deviation of this parameter (). twice as much as ) This is used to represent statistically significant deviations. The calculation process for this mapping rule is as follows: if the current temperature deviation... (in (The historical baseline temperature) is less than or equal to ,but =0; if Greater than ,but The value is obtained through the following linear scaling calculation and ranges between [0,1): ; in It is the extreme operating temperature (e.g., 85°C) specified in the technical specifications of the emergency sign.

[0028] The second part involves calculating the specific influence scores of all relevant parameters (such as...). , After that, they are combined into the final comprehensive environmental impact score. One approach is to take the maximum value, i.e. Next, the system calls the preset default weight ratio ( ), which is determined by the default weight of device risk. Environmental influences and default weights The default weights for device risk and environmental impact are determined by using the entire historical dataset generated in S12 (containing features). , A multivariate logistic regression model is trained using the target variable y(t). The training process aims to minimize the cross-entropy loss function, employing algorithms such as gradient descent to solve the model. The optimal coefficient in ( In this formula, It is the intermediate value (log odds) calculated by the model, which combines the linear effects of all input features. This is the intercept term of the model, representing the baseline log-odds when all input features are zero. Corresponding features The regression coefficients represent the average contribution of the equipment risk score to the logarithmic probability of failure. Corresponding features The regression coefficients represent the average contribution of the environmental impact score to the log odds of failure. The purpose of this model is to quantify... and Average contribution to the occurrence of failures.

[0029] After training, the obtained regression coefficients and The absolute value is considered as and The average contribution to the fault is calculated, and the weights are determined using the following normalization formula: ; ; Where the denominator is zero, it is set to 0.5. After determining... Then, the system determines the risk score based on the device ( ) and the environmental impact score ( ), and then perform weighted fusion to obtain a comprehensive risk score ( The calculation formula is: ; in, This represents a preliminary risk value that incorporates aging and environmental factors under default weights. It represents the probability of failure due to equipment aging. It represents the risk score of real-time environmental impact.

[0030] Then, based on the comprehensive risk score ( ), and the preset fusion threshold ( The preset fusion threshold () is compared with the data, and the final weight ratio is determined based on the comparison results. The setting is calculated through statistical analysis of all historical data within 24 hours prior to the occurrence of a failure. Values, setting the fusion threshold to these values. A specific quantile of the value (e.g., the 75th percentile) ensures that the threshold covers most failure scenarios induced by environmental mutations.

[0031] Except for those already determined In addition, the system also presets the adjusted weight ratio ( ), which is determined by the risk-adjusted weight of the equipment ( ) and environmental impact adjusted weights ( ) constitutes. To determine (Adjusted weighting ratios) First, a subset of data is selected from the entire historical dataset. This subset retains only data that meets the following criteria: Data points (i.e., historically high-risk scenarios). Then, using the same model structure as above, training is performed only on this subset of data to obtain a new set of regression coefficients. and Finally, the adjusted weights are obtained using the same normalization method: ; ; After confirming and Then, the real-time judgment logic begins to be executed. The judgment logic is: if... If the environmental impact is determined to be within a controllable range, then the appropriate method is selected. As the final weight; if If the environmental impact is deemed to pose a significant threat, then the operation of increasing the weight of environmental factors is performed. This operation refers to switching the weight ratio to the adjusted weight ratio. (For example, [0.5, 0.5]). The basis for setting and Similarly, but by analyzing only historical data... This step calculates the contribution of the regression on a subset of data at a given time (i.e., in high-risk scenarios). The output of this step is the final selected weight ratio. .

[0032] Finally, based on the final selected weight ratio ( ), for the equipment risk score ( ) and the environmental impact score ( The weighted summation is performed to obtain the updated dynamic risk level. The specific calculation formula is as follows: ; in, This is the final risk value output. and They are The corresponding equipment risk weights and environmental impact weights are specified. It is a final probability value in the range [0,1], which combines the aging state of the equipment itself and the real-time environmental impact.

[0033] In step S14, based on the updated dynamic risk level, risk level classification and frequency range allocation are performed to obtain a preliminary frequency allocation scheme, including: Based on the updated dynamic risk level, risk levels are classified to obtain a combination of classified risk levels; Based on the risk level combinations after classification, differentiated reporting frequency intervals are allocated to generate a preliminary frequency allocation scheme.

[0034] First, the system adjusts the updated dynamic risk level obtained in S13 ( This involves classifying risk levels. This classification process is performed by... This is achieved by comparing a probability value between 0 and 1 with a set of preset risk level thresholds. This threshold table contains at least two thresholds, such as a high-risk threshold (…). ) and low-risk threshold ( The threshold setting is based on historical data. The statistical distribution analysis of the values ​​and the comprehensive consideration of operational capabilities. A reproducible configuration method is to... Set as history The 90th percentile (P90) of the value distribution is used to identify the 10% of devices with the highest risk, matching their number to the maintenance team's priority handling capacity; Set as history The 50th quantile (P50, i.e., the median) of the value distribution is used to distinguish between devices with above-average and below-average risk. The comparison logic is as follows: if... The equipment was marked as "high risk"; if Marked as "medium risk"; if The risk level is marked as "low risk". The output of this step is a combination of risk levels after classification, including labels such as "high risk", "medium risk" or "low risk".

[0035] Next, the system allocates differentiated reporting frequency intervals based on the classified risk level combinations, generating a preliminary frequency allocation scheme. This allocation operation is performed based on a preset risk-frequency mapping table. This mapping table is a key-value pair structure, and its mapping rule is to assign a specific reporting frequency value (value) to each risk level (key), for example, ("high risk", ("Medium risk") ("Low risk") The frequency values ​​in the mapping table are based on the system's response time requirements and communication resource limitations for different risk levels. A common setting method is to first determine the reporting frequency for the "high-risk" level (…). Its setting is based on the requirement to meet the shortest emergency response time. ) and maximum number of retries ( For example, if safety procedures require 60 seconds ( The alarm must be acknowledged within the specified timeframe, and one retry is allowed. (meaning a total of 2 reporting opportunities), then The calculation formula is: ; In this example, Seconds. Secondly, determine the reporting frequency for "low-risk" levels ( Its setting is based on the maximum heartbeat interval required for the system to maintain the online status of the devices. This value is determined by the system operation and maintenance specifications (e.g., 3600 seconds / 1 hour), i.e. Finally, determine the reporting frequency for the "medium risk" level. Its setting is based on and One method to achieve balance between them is to use the geometric mean, which is calculated using the following formula: ; For example, if =30 seconds, =3600 seconds, then The interval is seconds, which can be rounded up to 300 seconds (5 minutes). It should be noted that the geometric mean, rather than the arithmetic mean, is chosen to strike a balance between the emergency response needs of high-risk equipment and the resource-saving needs of low-risk equipment. This value focuses more on the middle position on a sequential scale than the midpoint on a non-linear scale. The system queries this mapping table and assigns a corresponding frequency (e.g., 30 seconds) based on the current risk level of the equipment (e.g., "high risk"). This specific frequency value is the initial frequency allocation scheme, used for the subsequent load simulation in S15.

[0036] In step S15, based on the preliminary frequency allocation scheme, communication load simulation and frequency value adjustment are performed to obtain an optimized frequency allocation scheme, including: The frequency values ​​of the high-risk devices are extracted from the preliminary frequency allocation scheme, and a communication load simulation test is performed to obtain the load test judgment result. If the load test result indicates that the load is overloaded, the frequency of the low-risk device or the medium-risk device will be reduced to obtain the adjusted frequency value range. Based on the adjusted frequency range, frequency allocation schemes are integrated and verified to generate an optimized frequency allocation scheme.

[0037] The process begins with a communication load simulation test based on the preliminary frequency allocation scheme. This test summarizes the reporting frequencies allocated to all devices (including high, medium, and low-risk devices) in the scheme, and combines this with the total number of devices in the network. ) and reporting frequency ( ), calculate the total data load of the simulation ( In one implementation, the total number of messages per unit time (e.g., per hour) is first calculated: ; Then, Multiply by the estimated average packet size ( The total simulated data load is obtained as follows: ; The The estimation method involves adding the size of the application layer payload reported by the device to the size of the communication protocol overhead. For example, if the application layer payload is fixed at 18 bytes (containing device ID, status code, and voltage value), and the fixed header overhead of the communication protocol (such as NB-IoT) is 13 bytes, then... It was determined to be 31 bytes.

[0038] Subsequently, the total data load Communication capacity threshold with a preset gateway ( The comparison is as follows. It is based on the nominal data capacity of the communication link ( ) and safety margin ( The result is obtained through calculation, that is, ; The The setting is based on the data service package limit of the communication module. One method is to set it to the average hourly allowable data usage of that package (for example, if the package is 10MB / month, then...). =10,000,000 bytes / (30 days × 24 hours) ≈ 13,888 bytes / hour. The aforementioned safety margin... The setting (e.g., 0.2, or 20%) is based on a buffer to account for network signaling overhead, data retransmission, and fluctuations in package billing. If If the load test result is "load normal", then the load test result is "load normal"; if If the load test result is "load overload", the load test will determine the result as "load overload". If the load test result is "load normal", it indicates that the initial frequency allocation scheme does not need to be adjusted, and this scheme is directly determined as the optimized frequency allocation scheme. If the load test result is "load overload", the system will initiate a phased frequency reduction process. This process prioritizes ensuring the frequency of high-risk equipment, then medium-risk equipment, and finally low-risk equipment.

[0039] The first phase involves adjusting low-risk equipment, while maintaining the frequency of "high-risk" and "medium-risk" equipment in the system. The reporting frequency for "low-risk" devices (i.e., devices marked as "low-risk" in S14) remains unchanged. The adjustment is performed iteratively. This adjustment operation is based on a low-risk frequency adjustment step size ( ) and a minimum frequency limit for low risk ( The aforementioned The setting is based on using a standard unit of time. For example, if in S14... The initial value is 3600 seconds (1 hour), then It can be set to the same 3600 seconds; as a more refined adjustment strategy, the downsizing step size can also be set to a fixed percentage of the current frequency value (e.g., each iteration will...). Increase by 20% and set an absolute upper limit (e.g., 24 hours) to achieve smoother load control. The setting is based on the maximum device disconnection decision time required by the system operation specifications, such as 86,400 seconds (24 hours). In each iteration, the system calculates a new low-risk frequency: ; (like Exceed Then set it to Subsequently, the system used... Recalculate total data load And verify it. If If the verification passes, the iteration stops, and the system will contain ( The scheme described above is determined to be the optimized frequency allocation scheme. and Achieved If the first phase adjustment is insufficient, the system will start the second phase.

[0040] The second phase involves adjusting high-risk equipment. During this phase, the system maintains a high-risk equipment frequency ( ) and low-risk equipment frequencies that have reached their lower limit ( The reporting frequency for "medium-risk" devices (i.e., devices marked as "medium-risk" in S14) remains unchanged. The adjustment is iteratively reduced. This reduction operation is based on a medium-risk frequency reduction step size ( ) and a minimum frequency limit for medium risk ( The step size for adjusting the medium-risk frequency can be set to a value smaller than the step size for adjusting the low-risk frequency (e.g., 900 seconds / 15 minutes) to achieve finer load control. The lower limit for the medium-risk minimum frequency can be set to... The initial value (e.g., 3600 seconds). System calculation: ; like Exceed Then set it as Subsequently, the system uses ( Recalculate the combination of ) And verify it. If If the verification is successful, the iteration stops, and this combination scheme is determined as the optimized frequency allocation scheme. and Achieved If the first two stages of adjustment are insufficient, the system will start the third stage.

[0041] The third phase involves adjusting high-risk equipment. During this phase, the system maintains a medium-risk level that has already reached its lower limit. ) and low risk ( The equipment frequency remains unchanged, and the reporting frequency for "high-risk" equipment (i.e., equipment marked as "high-risk" in S14) is increased. The adjustment is performed iteratively. This adjustment operation is based on a high-risk frequency adjustment step size ( ) and a minimum frequency limit for high risk ( The aforementioned It can be set to a value less than The value (e.g., 10 seconds) is used for fine-tuning. The setting is based on what was determined in S14. initial value and An intermediate value between the initial values ​​(e.g., 60 seconds) must be greater than 0. The system calculates: ; Exceed Then set it as Subsequently, the system uses ( Recalculate the combination of ) And verify it. If If the verification is successful, the iteration stops, and this combination scheme is determined as the optimized frequency allocation scheme. Achieved This indicates that the frequency lower limit of all adjustable devices has been reached, but the load is still overloaded. In this case, the system will switch to the currently used ( The candidate scheme is determined as the optimized frequency allocation scheme, and an alarm for "system capacity overload" is generated at the same time to notify the operation and maintenance personnel that network expansion is required, and then the iteration stops.

[0042] In step S16, according to the optimized frequency allocation scheme, configuration instructions are issued and execution status is confirmed, and a final reporting mechanism is generated, including: Based on the optimized frequency allocation scheme, frequency configuration instructions are generated to obtain a set of instructions to be issued; Based on the set of instructions to be issued, instructions are issued to the emergency sign to obtain the instruction transmission status. Based on the instruction transmission status, the configuration execution status of the emergency sign is monitored to obtain status feedback results; Based on the status feedback result, it is compared with a preset response time threshold. If the response time threshold is not reached, the communication signal is optimized to obtain an optimized communication link. Based on the optimized communication link or the status feedback result that has reached the response time threshold, the final execution status is confirmed, and a final reporting mechanism is generated.

[0043] First, the system generates frequency configuration instructions based on the optimized frequency allocation scheme (e.g., device ID 101 is allocated a frequency of 300 seconds). This generation operation encapsulates the allocated frequency value (e.g., 300 seconds) into a downlink data packet of a specific format. A reproducible data packet format is a hexadecimal sequence [0xAA, 0x01, 0x01, 0x2C], where 0xAA is the packet header, 0x01 is the command code for "set frequency", and 0x01 and 0x2C are two bytes representing the frequency value 300. All such data packets generated for devices whose frequencies have changed constitute the set of instructions to be sent and are stored in the server's downlink message queue.

[0044] Next, based on the set of instructions to be sent, instructions are sent to the emergency sign. This sending operation is performed asynchronously: the server waits for the next active uplink data packet from the target device (ID 101). After the device sends an uplink packet, the network server (e.g., a LoRaWAN network server) pushes the instruction packets ([0xAA,0x01,0x01,0x2C]) in the queue to the device in the subsequent downlink receive window (e.g., RX1 or RX2 window). After completing this push attempt, the network server returns a network layer instruction transmission status to the application server, such as "Sent_OK" (indicating sent) or "Send_Failed" (indicating the device is unreachable).

[0045] Based on the command transmission status (assumed to be "Sent_OK"), the server starts a timer and monitors the configuration execution status of the emergency indicator. This monitoring involves waiting for the device to send back an application-layer acknowledgment (ACK) packet. This ACK packet (e.g., [0xBB, 0x01, 0x00], where 0xBB is the ACK header, 0x01 is the response to the "Set Frequency" command, and 0x00 indicates "Successful execution") indicates that the device has correctly received and applied the new frequency. If the server receives this ACK packet within the timer period, the status feedback result is marked as "Success_Ack_Received"; if the timer expires and no ACK is received, the status feedback result is marked as "Timeout_Error".

[0046] Then, based on the state feedback result, and a preset response time threshold ( The comparison is as follows. It can be set to the reporting frequency of the device before optimization in S15. ) plus a fixed network latency margin ( ),Right now, ; in This can be set based on the network's average round-trip time (RTT) (e.g., 60 seconds). If the status feedback result is "Timeout_Error" (i.e., in...), If no ACK is received within the specified time, the system determines that the response time threshold has not been reached, and then performs communication signal optimization. This optimization involves the server sending a control command to the network server, requesting it to use more robust communication parameters in the next retry of the command. These more robust communication parameters refer to configurations that sacrifice data rate for higher signal reliability and stronger penetration. For example, in LoRaWAN, this list successively adjusts the data rate from the default DR3 (spreading factor SF9) to DR2 (SF10) or DR1 (SF11); in NB-IoT, this parameter can correspond to increasing the number of repetitions of network requests. This adjusted parameter configuration constitutes the optimized communication link.

[0047] Subsequently, the system requeues the command, returns to the instruction issuance step, and uses an internal retry counter ( Count until The maximum number of retries has been reached. The maximum number of retries is set based on the network's historical average packet loss rate (PLR) and the target command delivery success rate (PCR). (e.g., 99.9%). One approach is to set the maximum number of retries. Set as the smallest integer that satisfies the condition. For example, if PLR is 20% (0.2). To achieve 99.9% (0.999), the following condition must be met. Solving for N, we get N+1=5. Therefore... It was set to 4 times.

[0048] Finally, based on the optimized communication link (i.e., the operation through retry) or the status feedback result that has reached the response time threshold (i.e., "Success_Ack_Received" is received on the first attempt or after a retry), the final execution status is confirmed. If the final status is confirmed as "Success_Ack_Received", the server updates the reporting frequency of device (ID 101) to the new value (300 seconds) in the database. If all retries (through the optimized communication link) fail, and the final status is confirmed as "Config_Failed", the server retains the original frequency of the device and generates an alarm log. The updated or retained device frequency record in the server database constitutes the final reporting mechanism.

[0049] In summary, this invention achieves intelligent collaboration between risk prediction and resource allocation by dynamically assessing the risk level of indicator signs (integrating historical aging indicators and real-time environmental changes) and combining simulated communication load testing to optimize the allocation of reporting frequencies for devices with different risk levels.

[0050] Reference Figure 2 The second embodiment of the present invention provides a remote monitoring system for the status of LED emergency signs, comprising: The data acquisition and processing module is used to acquire historical operating data and current environmental parameters of emergency signs, and to perform data cleaning on the historical operating data and current environmental parameters to obtain a standardized historical performance dataset. The risk assessment module is used to calculate the aging index of the emergency sign based on the standardized historical performance dataset, and to predict the failure probability based on the aging index to obtain the equipment risk score. The dynamic calculation module is used to perform fusion and weighted calculation based on the device risk score and real-time acquired environmental change data to obtain the updated dynamic risk level; The frequency allocation module is used to classify risk levels and allocate frequency ranges based on the updated dynamic risk level to obtain a preliminary frequency allocation scheme. The load optimization module is used to perform communication load simulation and frequency value adjustment based on the preliminary frequency allocation scheme to obtain an optimized frequency allocation scheme. The instruction execution and feedback module is used to issue configuration instructions and confirm execution status according to the optimized frequency allocation scheme, and generate the final reporting mechanism.

[0051] It should be noted that the remote monitoring system for the status of an LED emergency sign provided in this embodiment of the invention is used to execute all the process steps of the remote monitoring method for the status of an LED emergency sign in the above embodiment. The working principles and beneficial effects of the two are one-to-one, so they will not be described again.

[0052] This invention also provides an electronic device. The electronic device includes a processor, a memory, and a computer program stored in the memory and executable on the processor, such as a remote monitoring program for the status of LED emergency signs. When the processor executes the computer program, it implements the steps in the aforementioned embodiments of the remote monitoring method for the status of LED emergency signs, for example... Figure 1 The step S11 shown. Alternatively, when the processor executes the computer program, it implements the functions of each module / unit in the above system embodiments, such as the load optimization module.

[0053] For example, the computer program may be divided into one or more modules / units, which are stored in the memory and executed by the processor to complete the present invention. The one or more modules / units may be a series of computer program instruction segments capable of performing a specific function, which describe the execution process of the computer program in the electronic device.

[0054] The electronic device may be a desktop computer, laptop, handheld computer, or smart tablet, etc. The electronic device may include, but is not limited to, a processor and memory. Those skilled in the art will understand that the above components are merely examples of electronic devices and do not constitute a limitation on the electronic device. It may include more or fewer components than described above, or combine certain components, or different components. For example, the electronic device may also include input / output devices, network access devices, buses, etc.

[0055] The processor can be a Central Processing Unit (CPU), or other general-purpose processors, digital signal processors (DSPs), application-specific integrated circuits (ASICs), field-programmable gate arrays (FPGAs), or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components, etc. A general-purpose processor can be a microprocessor or any conventional processor. The processor is the control center of the electronic device, connecting all parts of the electronic device via various interfaces and lines.

[0056] The memory can be used to store the computer programs and / or modules. The processor implements various functions of the electronic device by running or executing the computer programs and / or modules stored in the memory and by calling data stored in the memory. The memory may mainly include a program storage area and a data storage area. The program storage area may store the operating system, at least one application program required for a function (such as sound playback function, image playback function, etc.), etc.; the data storage area may store data created according to the use of the mobile phone (such as audio data, phonebook, etc.). In addition, the memory may include high-speed random access memory, and may also include non-volatile memory, such as hard disk, memory, plug-in hard disk, smart media card (SMC), secure digital (SD) card, flash card, at least one disk storage device, flash memory device, or other volatile solid-state storage device.

[0057] If the modules / units integrated into the electronic device are implemented as software functional units and sold or used as independent products, they can be stored in a computer-readable storage medium. Based on this understanding, all or part of the processes in the methods of the above embodiments of the present invention can also be implemented by a computer program instructing related hardware. The computer program can be stored in a computer-readable storage medium, and when executed by a processor, it can implement the steps of the various method embodiments described above. The computer program includes computer program code, which can be in the form of source code, object code, executable files, or certain intermediate forms. The computer-readable medium can include: any entity or system capable of carrying the computer program code, recording media, USB flash drives, portable hard drives, magnetic disks, optical disks, computer memory, read-only memory (ROM), random access memory (RAM), electrical carrier signals, telecommunication signals, and software distribution media, etc. It should be noted that the content included in the computer-readable medium can be appropriately added or removed according to the requirements of legislation and patent practice in the jurisdiction. For example, in some jurisdictions, according to legislation and patent practice, computer-readable media do not include electrical carrier signals and telecommunication signals.

[0058] It should be noted that the system embodiments described above are merely illustrative. The units described as separate components may or may not be physically separate, and the components shown as units may or may not be physical units; that is, they may be located in one place or distributed across multiple network units. Some or all of the modules can be selected to achieve the purpose of this embodiment according to actual needs. Furthermore, in the accompanying drawings of the system embodiments provided by this invention, the connection relationships between modules indicate that they have communication connections, which can be specifically implemented as one or more communication buses or signal lines. Those skilled in the art can understand and implement this without any creative effort.

[0059] The specific embodiments described above further illustrate the purpose, technical solution, and beneficial effects of the present invention. It should be understood that the above descriptions are merely specific embodiments of the present invention and are not intended to limit the scope of protection of the present invention. In particular, it should be noted that any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of the present invention should be included within the scope of protection of the present invention for those skilled in the art.

Claims

1. A method for remotely monitoring the status of an LED emergency sign, characterized in that, include: Obtain historical operating data and current environmental parameters of emergency signs, and perform data cleaning on the historical operating data and current environmental parameters to obtain a standardized historical performance dataset; Based on the standardized historical performance dataset, the aging index of the emergency sign is calculated, and the failure probability is predicted based on the aging index to obtain the equipment risk score. Based on the equipment risk score and the real-time environmental change data, a fusion and weighted calculation is performed to obtain the updated dynamic risk level; Based on the updated dynamic risk level, risk level classification and frequency interval allocation are performed to obtain a preliminary frequency allocation scheme. Based on the preliminary frequency allocation scheme, communication load simulation and frequency value adjustment are performed to obtain an optimized frequency allocation scheme. Based on the optimized frequency allocation scheme, configuration commands are issued and execution status is confirmed to generate the final reporting mechanism.

2. The method for remotely monitoring the status of LED emergency signs according to claim 1, characterized in that, The process involves acquiring historical operational data and current environmental parameters of emergency signs, and then cleaning the historical operational data and current environmental parameters to obtain a standardized historical performance dataset, including: Based on the historical operating data and current environmental parameters, outlier identification and removal are performed to obtain a clean dataset; Based on the clean dataset, format unification and unit normalization are performed to obtain an intermediate dataset; Based on the intermediate dataset, integrity checks and missing data interpolation are performed to obtain a standardized historical performance dataset.

3. The method for remotely monitoring the status of LED emergency signs according to claim 1, characterized in that, The step of calculating the aging index of the emergency sign based on the standardized historical performance dataset, and predicting the failure probability based on the aging index to obtain an equipment risk score includes: Based on the standardized historical performance dataset, feature data related to equipment aging are extracted to obtain an aging feature set; Based on the set of aging characteristics, a weighted comprehensive calculation is performed to obtain the quantitative value of the aging index; Based on the quantified value of the aging index, combined with the preset aging threshold, the equipment risk is marked to obtain the equipment risk grouping result including high-risk equipment, medium-risk equipment and low-risk equipment; Based on the equipment risk grouping results and the quantified value of the aging index, combined with pre-stored historical fault data, fault probability prediction is performed to obtain the equipment risk score.

4. The method for remotely monitoring the status of LED emergency signs according to claim 1, characterized in that, The step of fusing and weighting the risk scores of the equipment and the real-time environmental change data to obtain the updated dynamic risk level includes: A comprehensive risk score is obtained by weighting and fusing the equipment risk score with real-time environmental change data. The comprehensive risk score is compared with a preset fusion threshold. If the score is higher than the fusion threshold, the weight of environmental factors is increased to obtain an adjusted weight ratio. Based on the adjusted weighting ratio, the equipment risk score and the environmental change data are weighted and summed to obtain the updated dynamic risk level.

5. The method for remotely monitoring the status of LED emergency signs according to claim 1, characterized in that, The step of classifying risk levels and allocating frequency intervals based on the updated dynamic risk level to obtain a preliminary frequency allocation scheme includes: Based on the updated dynamic risk level, risk levels are classified to obtain a combination of classified risk levels; Based on the risk level combinations after classification, differentiated reporting frequency intervals are allocated to generate a preliminary frequency allocation scheme.

6. The method for remotely monitoring the status of LED emergency signs according to claim 3, characterized in that, The step of performing communication load simulation and frequency value adjustment based on the preliminary frequency allocation scheme to obtain an optimized frequency allocation scheme includes: The frequency values ​​of the high-risk devices are extracted from the preliminary frequency allocation scheme, and a communication load simulation test is performed to obtain the load test judgment result. If the load test result indicates that the load is overloaded, the frequency of the low-risk device or the medium-risk device will be reduced to obtain the adjusted frequency value range. Based on the adjusted frequency range, frequency allocation schemes are integrated and verified to generate an optimized frequency allocation scheme.

7. The method for remotely monitoring the status of LED emergency signs according to claim 1, characterized in that, The step of issuing configuration commands and confirming execution status according to the optimized frequency allocation scheme, and generating a final reporting mechanism, includes: Based on the optimized frequency allocation scheme, frequency configuration instructions are generated to obtain a set of instructions to be issued; Based on the set of instructions to be issued, instructions are issued to the emergency sign to obtain the instruction transmission status. Based on the instruction transmission status, the configuration execution status of the emergency sign is monitored to obtain status feedback results; Based on the status feedback result, it is compared with a preset response time threshold. If the response time threshold is not reached, the communication signal is optimized to obtain an optimized communication link. Based on the optimized communication link or the status feedback result that has reached the response time threshold, the final execution status is confirmed, and a final reporting mechanism is generated.

8. A remote monitoring system for the status of LED emergency signs, characterized in that, include: The data acquisition and processing module is used to acquire historical operating data and current environmental parameters of emergency signs, and to perform data cleaning on the historical operating data and current environmental parameters to obtain a standardized historical performance dataset. The risk assessment module is used to calculate the aging index of the emergency sign based on the standardized historical performance dataset, and to predict the failure probability based on the aging index to obtain the equipment risk score. The dynamic calculation module is used to perform fusion and weighted calculation based on the device risk score and real-time acquired environmental change data to obtain the updated dynamic risk level; The frequency allocation module is used to classify risk levels and allocate frequency ranges based on the updated dynamic risk level to obtain a preliminary frequency allocation scheme. The load optimization module is used to perform communication load simulation and frequency value adjustment based on the preliminary frequency allocation scheme to obtain an optimized frequency allocation scheme. The instruction execution and feedback module is used to issue configuration instructions and confirm execution status according to the optimized frequency allocation scheme, and generate the final reporting mechanism.

Images