Internet of Things-based method and system for monitoring electrical safety in tunnels

Abstract

This application relates to the field of data processing technology, and in particular to a method and system for monitoring electrical safety in tunnels based on the Internet of Things (IoT). It involves constructing an adjustment coefficient; using the product of the adjustment coefficient and a preset initial cutoff distance as the optimal cutoff distance; and employing the DPC algorithm based on the optimal cutoff distance to detect anomalies in electrical equipment. This application improves the accuracy of monitoring electrical equipment in tunnels.

Description

Technical Field

[0001] This application relates to the field of data processing technology, and in particular to a method and system for monitoring electrical safety in tunnels based on the Internet of Things. Background Technology

[0002] With the accelerated pace of urbanization, underground space resources are showing a trend towards in-depth development and comprehensive utilization. As a key component of the highway system, multi-level underground tunnel complexes are continuously increasing in terms of construction density and engineering scale. Limited ventilation and lighting conditions within tunnels, coupled with single evacuation routes, pose significant challenges to fire prevention and control. Furthermore, tunnels typically house ventilation, lighting, and other shared electrical equipment to provide ventilation and lighting; electrical equipment failure is a major cause of fires within tunnels. To reduce tunnel fire risks, monitoring of electrical equipment is essential. Traditional tunnel monitoring methods rely primarily on manual inspections and point-based measurements, which suffer from low monitoring frequency, discontinuous data, and delayed early warnings. With the development of the Internet of Things (IoT) and electronic information technology, IoT-based electrical equipment monitoring technologies have emerged. These technologies acquire operational data from various electrical devices, such as key parameters like current, voltage, and power, through IoT technology. Anomaly detection algorithms detect abnormal data and moments to achieve real-time monitoring and early warning of electrical equipment. DPC (Density Peak Clustering) is a density-based clustering method capable of handling data with arbitrary distributions. The DPC algorithm can detect outliers in a dataset that do not conform to the normal distribution, thereby enabling the monitoring of the operating status of electrical equipment.

[0003] In the DPC algorithm, the local density of each data point is calculated based on the neighborhood cutoff distance. This cutoff distance is determined by monitoring personnel based on their experience. When the cutoff distance is too small, it can easily lead to misjudgments of noise and cluster splitting; when the cutoff distance is too large, it can easily result in the incorrect merging of clusters of different densities and blurred boundaries of high-density areas, making it difficult to accurately identify cluster centers. During the updating of equipment data, the data distribution is affected by external environmental factors or the aging of the equipment itself. The cutoff distance determined based on experience cannot adapt to long-term monitoring of electrical equipment data, ultimately leading to low accuracy in monitoring results. Summary of the Invention

[0004] To address the issue of low accuracy in anomaly detection using traditional DPC algorithms, this application provides a method and system for monitoring tunnel electrical safety based on the Internet of Things.

[0005] Firstly, this application provides a tunnel electrical safety monitoring method based on the Internet of Things, employing the following technical solution:

[0006] An IoT-based method for monitoring electrical safety in tunnels involves constructing an adjustment coefficient; using the product of the adjustment coefficient and a preset initial cutoff distance as the optimal cutoff distance; and employing the DPC algorithm based on the optimal cutoff distance to detect anomalies in electrical equipment.

[0007] The steps for constructing the adjustment coefficient include: acquiring tunnel environmental humidity data and determining the humidity deviation coefficient at each time point; constructing a temperature matching coefficient for changes in environmental temperature and equipment temperature, and using the ratio of the normalized result of the humidity deviation coefficient to the normalized result of the temperature matching coefficient as the environmental anomaly factor; constructing voltage and current sequences based on equipment voltage and current data, and constructing a first trend index based on the similarity between the two sequences; determining a second trend index based on the changing trends between equipment temperature and current data, and using the sum of the first and second trend indices as the equipment pseudo-anomaly factor; for each time point, using the mean of the environmental anomaly factors from multiple time points prior to each time point as the standard influence coefficient, and using the mean of the equipment pseudo-anomaly factors from multiple time points prior to each time point as the standard pseudo-anomaly coefficient; and using the ratio between the standard pseudo-anomaly coefficient and the standard influence coefficient as the adjustment coefficient for each time point.

[0008] The beneficial effects are as follows: During real-time monitoring of electrical equipment, each collected data point corresponds to an adjustment coefficient. This coefficient is used to adjust the cutoff distance in the DPC algorithm, resulting in the optimal cutoff distance. By using the dynamic optimal cutoff distance, the DPC algorithm employs anomaly detection to monitor the data. Compared to the fixed cutoff distance method in traditional DPC algorithms, the method in this application can adapt to the overall deviation (non-abnormal changes) of the monitored data of the electrical equipment to a certain extent, thereby improving the accuracy of anomaly detection.

[0009] Meanwhile, the calculation of the adjustment coefficient is based on two factors. First, it considers the environmental humidity, specifically the humidity deviation coefficient. If the humidity in the tunnel environment deviates significantly from the normal range at a given moment, it indicates a potential equipment malfunction affecting the tunnel environment. Second, the calculation also considers changes in ambient and equipment temperatures. Normally, these two temperatures influence each other and change synchronously. If the change between ambient and equipment temperatures deviates from the normal range at a certain stage, it indicates an increased risk of equipment malfunction. By combining environmental factors with the analysis of the risk of equipment malfunction, and then adjusting the cutoff distance, the accuracy of the monitored data can be further improved.

[0010] Optionally, the steps of acquiring tunnel environmental humidity data and determining the humidity deviation coefficient at each moment from the normal humidity include: for any given moment, acquiring the humidity trend at that moment; acquiring the range of normal humidity data and taking the mean of the boundary values ​​of the range as the humidity center; taking the absolute difference between the humidity trend and the humidity center as the deviation value, and taking the ratio of the deviation value to the interval length of the range as the humidity deviation coefficient.

[0011] The beneficial effects are as follows: Normal humidity data in tunnels can be obtained through monitoring the tunnel environment, which varies depending on the season and the tunnel's geographical location. Under normal circumstances, humidity in the tunnel environment fluctuates to some extent, thus obtaining the normal humidity range. The mean of the boundaries of this range is used as the humidity center. The difference between the current humidity value and the humidity center determines the degree to which the current humidity trend deviates from normal humidity, thereby deriving the humidity deviation coefficient.

[0012] Optionally, the step of obtaining the humidity trend includes: for any given moment, taking the average of a first preset number of consecutive adjacent humidity data points before that moment as the humidity trend at that moment.

[0013] The beneficial effects are as follows: the humidity trend is mainly used to represent the current humidity status in the tunnel environment. The current humidity status is represented by the average of multiple data, thereby improving the robustness of the humidity trend calculation and reducing the impact of individual abnormal humidity data on the overall humidity situation.

[0014] Optionally, the steps for constructing a temperature matching coefficient based on the correlation between ambient temperature changes and equipment temperature changes include: for any given time, constructing the corresponding neighboring ambient temperature sequence and neighboring equipment temperature sequence; using wavelet coherence method to obtain the wavelet coherence coefficients between the neighboring ambient temperature sequence and the neighboring equipment temperature sequence and drawing a wavelet coherence diagram; obtaining the coherence intensity of the correlation between the changes in the neighboring ambient temperature sequence and the neighboring equipment temperature sequence based on the wavelet coherence diagram, and obtaining the isotopic intensity based on the trend of their changes; and using the product of the isotopic intensity normalization result and the coherence intensity normalization result as the temperature matching coefficient.

[0015] The beneficial effects are as follows: For any given data acquisition time, a neighboring temperature series and a neighboring equipment temperature series are constructed and analyzed to obtain the coherence intensity and isotopic intensity of the changes between the two series. The product of these two series is used as the temperature consistency coefficient, thus accurately representing the temperature consistency coefficient between changes in ambient temperature and equipment temperature.

[0016] Optionally, for any given moment, a second preset number of ambient temperature data points prior to that moment are acquired to form a nearby ambient temperature sequence; and a second preset number of device temperature data points prior to that moment are acquired to form a nearby device temperature sequence.

[0017] The beneficial effect is that, for any given moment, multiple data points adjacent to that moment are selected to form a nearby ambient temperature sequence or a nearby equipment temperature sequence, thereby effectively reflecting the changing trends of ambient temperature and equipment temperature within the corresponding time period.

[0018] Optionally, the ratio of the red and yellow regions in the wavelet coherence graph to the area of ​​the entire wavelet coherence graph can be used as the coherence intensity.

[0019] The beneficial effect is that, in the wavelet coherence diagram, the highlighted part represents the correlation strength between the changes between the two sequences. Therefore, the ratio of the red and yellow areas in the wavelet coherence diagram to the area of ​​the entire wavelet coherence diagram is used as the coherence strength.

[0020] Optionally, the steps for obtaining isotopic intensity include: obtaining the main trend of the arrow directions in the red and yellow regions of the wavelet coherence graph; and taking the ratio of the number of arrows corresponding to the main trend to the total number of arrows in the red and yellow regions as isotopic intensity.

[0021] The beneficial effect is that in wavelet coherence diagrams, the direction of the arrows indicates the phase difference between the changes of two sequences. Under normal circumstances, the changes between the two sequences are synchronized, or the change in device temperature lags slightly behind the change in ambient temperature, but the trends of the changes are consistent. Therefore, in wavelet coherence diagrams, the more arrows corresponding to the main trend, the greater the co-position intensity.

[0022] Optionally, the steps for constructing the first trend index include: using the STL decomposition algorithm to obtain the trend term sequences of the current sequence and voltage sequence and performing normalization processing to obtain the standard trend sequence corresponding to the current sequence and the standard trend sequence corresponding to the voltage sequence, and using the Pearson correlation coefficient between the standard trend sequence of the current sequence and the standard trend sequence corresponding to the voltage sequence as the first trend index.

[0023] Optionally, the step of determining the second trend index based on the changing trend between equipment temperature data and equipment current data includes: for any given moment, obtaining the difference between the equipment temperature data at that moment and the adjacent equipment temperature data before that moment, and using this difference as the first difference; obtaining the difference between the current data at that moment and the adjacent current data before that moment, and using this difference as the second difference; calculating the second trend index based on the first difference and the second difference, wherein the calculation formula for the second trend index is: In the formula, Indicates time The corresponding second trend index; Indicates the first difference; Indicates the second difference; This represents the step function.

[0024] The beneficial effects are: based on the first and second differences, it represents the change in current and the trend of current change. When the current increases, the second difference is positive; when the equipment temperature increases, the first difference is positive. If ultimately... If the value is positive, the final second trend index is 1; otherwise, it is 0. A positive second trend index indicates that the increase in equipment temperature is caused by an increase in current, which may be due to a change in the working state of the electrical equipment (such as changing a ventilation device from level two to level three). This change is normal, and in this case, the higher equipment temperature or current is not a true equipment abnormality.

[0025] Secondly, this application provides an Internet of Things-based tunnel electrical safety monitoring system, which adopts the following technical solution:

[0026] The Internet of Things (IoT) based tunnel electrical safety monitoring system includes a processor and a memory, wherein the memory stores computer program instructions, and when the computer program instructions are executed by the processor, the above-described IoT-based tunnel electrical safety monitoring method is implemented.

[0027] The beneficial effects are: the above-mentioned IoT-based tunnel electrical safety monitoring method is generated into a computer program and stored in a memory so that it can be loaded and executed by a processor, thereby making the system easy to use by creating a system based on the memory and processor.

[0028] This application has the following technical effects:

[0029] 1. Dynamically adjust the cutoff distance of the DPC clustering algorithm to adapt to the normal changes in equipment data over time, thereby improving the accuracy of detecting abnormal data in the monitored data.

[0030] 2. The adjustment coefficient is determined based on the impact of the tunnel environment on the equipment and the changes in the tunnel environment. The impact of the environment on the equipment data is taken into account, thereby further improving the accuracy of the detection of the equipment's monitoring data. Attached Figure Description

[0031] Figure 1 This is a flowchart of the tunnel electrical safety monitoring method based on the Internet of Things (IoT) proposed in this application.

[0032] Figure 2 This is a flowchart of step S1 in the Internet of Things-based tunnel electrical safety monitoring method of this application. Detailed Implementation

[0033] This application discloses an IoT-based method for monitoring electrical safety in tunnels. Based on various data collected in the tunnel environment and data from equipment operation, an adjustment coefficient is constructed. During the detection of tunnel electrical equipment data, the initial cutoff distance in the DPC algorithm is dynamically adjusted based on the adjustment coefficient corresponding to the real-time collected data, thereby completing anomaly detection. In this method, the dynamic adjustment of the cutoff distance adapts to the slow changes in equipment data over a long period, maintaining the accuracy of anomaly detection. Furthermore, the calculation of the adjustment coefficient in this application takes into account tunnel environmental factors, making the adjusted cutoff distance more accurate, thus improving the accuracy of anomaly detection.

[0034] Reference Figure 1 The method for monitoring the electrical safety of tunnels based on the Internet of Things includes steps S1-S2.

[0035] S1: Construct the adjustment coefficient.

[0036] A monitoring area typically includes multiple tunnels, and each tunnel usually contains various types and quantities of electrical equipment. Therefore, IoT technology can be used to acquire environmental data from each tunnel, as well as data from the electrical equipment within each tunnel. Furthermore, the data to be monitored for the electrical equipment can include current, power, motor temperature, etc.

[0037] Reference Figure 2 The steps for constructing the adjustment coefficient include: steps S11-S14.

[0038] S11: Obtain tunnel environmental humidity data and determine the humidity deviation coefficient at each time point from the normal humidity.

[0039] For any given moment, obtain the humidity trend at that moment; obtain the range of normal humidity data, and take the mean of the boundary values ​​of the range as the humidity center; take the absolute difference between the humidity trend and the humidity center as the deviation value, and take the ratio of the deviation value to the interval length of the range as the humidity deviation coefficient.

[0040] For any tunnel, humidity information is obtained from the location near the electrical equipment in the tunnel using a humidity sensor. The sampling frequency is once every two seconds. In other embodiments, the sampling frequency can be adjusted according to actual needs.

[0041] After humidity data is collected, it is arranged in chronological order. For any given moment, a first preset number of ambient temperature data points preceding that moment are obtained, and the average of these first preset number of ambient temperature data points is calculated. This average is used as the humidity trend at that moment. The first preset number is no more than 50 to avoid including data from too distant periods. In this embodiment, it is set to 10, but in other embodiments it can be 20 or 30, etc.

[0042] The range of normal humidity data is obtained based on the geographical location and weather of the tunnel area. It can be obtained through experiments, such as measuring the humidity in the tunnel under the condition that there are no abnormalities and the electrical equipment is working normally.

[0043] The normal humidity data range is obtained, and the mean of the two boundaries of this range is calculated to obtain the humidity center of the normal humidity data range. This humidity center reflects the humidity level of the tunnel under normal conditions. The absolute value of the difference between the humidity value and the humidity center is calculated, and this absolute value is taken as the deviation value, which can also be understood as the difference between the current humidity trend and the humidity center of the normal humidity data. The smaller the difference, the more normal the current humidity conditions are near the electrical equipment. Finally, the ratio of the deviation value to the interval length of the normal humidity data range is taken as the humidity deviation coefficient. The humidity deviation coefficient represents the degree to which the current humidity in the tunnel deviates from the normal humidity; the larger the humidity deviation coefficient, the more abnormal the current humidity conditions in the tunnel.

[0044] S12: Construct a temperature matching coefficient between ambient temperature and equipment temperature changes, and use the ratio of the normalized result of the humidity deviation coefficient to the normalized result of the temperature matching coefficient as the environmental anomaly factor.

[0045] The steps for constructing a temperature matching coefficient based on the correlation between ambient temperature changes and equipment temperature changes include: for any given time, constructing the corresponding neighboring ambient temperature sequence and neighboring equipment temperature sequence; using wavelet coherence method to obtain the wavelet coherence coefficient between the neighboring ambient temperature sequence and the neighboring equipment temperature sequence and drawing a wavelet coherence diagram; obtaining the coherence intensity of the correlation between the changes in the neighboring ambient temperature sequence and the neighboring equipment temperature sequence based on the wavelet coherence diagram, and obtaining the isotopic intensity based on the trend of their changes; and using the product of the isotopic intensity normalization result and the coherence intensity normalization result as the temperature matching coefficient.

[0046] Temperature sensors are used to collect ambient temperature data in the tunnel and equipment temperature data at specific locations on electrical equipment (such as motors, casings, etc.). The ambient temperature data and equipment temperature data are arranged in chronological order to form ambient temperature sequences and equipment temperature sequences.

[0047] To analyze the correlation between changes in ambient temperature and equipment temperature, for any given moment, a neighbor sequence is constructed. This neighbor sequence includes the neighboring ambient temperature sequence for the corresponding ambient temperature and the neighboring equipment temperature sequence for the corresponding equipment temperature. Wavelet coherence is used to process the neighbor sequence, obtaining the wavelet coherence coefficients for each moment, and a wavelet coherence diagram is plotted. This step is a standard technique in this field and will not be elaborated further.

[0048] The ratio of the red and yellow regions in the wavelet coherence graph to the area of ​​the entire wavelet coherence graph is taken as the coherence intensity.

[0049] Under normal circumstances, ambient temperature and equipment temperature are correlated; when the ambient temperature rises, the equipment temperature will also rise. Similarly, when the ambient temperature decreases, the equipment temperature will decrease. In wavelet coherence diagrams, brightly colored areas represent strong correlations. The stronger the correlation, the brighter the area; that is, the greater the proportion of red and yellow areas in the wavelet coherence diagram, the stronger the correlation between the ambient temperature sequence and the equipment temperature sequence, and the greater the coherence intensity.

[0050] For isotopic intensity, the main trend of the arrow directions in the red and yellow regions of the wavelet coherence graph is obtained; the ratio of the number of arrows corresponding to the main trend to the total number of arrows in the red and yellow regions is taken as isotopic intensity.

[0051] Locate the red and yellow regions in the wavelet coherence graph. For each arrow within these regions, count the number of arrows in any given direction, and select the direction with the most arrows as the dominant trend. For example, if a wavelet coherence graph contains 6 vertically upward arrows, 5 horizontally to the right, and 4 horizontally to the left, then the dominant trend of the arrows in the wavelet coherence graph between the ambient temperature sequence and the adjacent equipment temperature sequence is vertically upward, with 6 arrows corresponding to this dominant trend, resulting in a final isotopic intensity of 6 / 15.

[0052] In wavelet coherence plots, if the arrow points horizontally to the right, it indicates that the ambient temperature sequence and the device temperature sequence change synchronously with no significant phase difference. If the arrow points vertically upward, it indicates that the phase difference between the two sequences is +90°. Under normal circumstances, changes in ambient temperature and device temperature are synchronous, meaning there is no significant phase difference between them, or the device temperature change lags behind the ambient temperature change, but the trends are the same. Therefore, in the corresponding wavelet coherence plot, there are more arrows with the same direction, resulting in a larger in-phase intensity value.

[0053] The product of the isotopic intensity normalization result and the coherent intensity normalization result is used as the temperature fit coefficient.

[0054] The ratio of the normalized result of the humidity deviation coefficient to the normalized result of the temperature fit coefficient is used as the environmental anomaly factor.

[0055] The temperature conformity coefficient and humidity deviation coefficient are used together to calculate environmental anomaly factors. The larger the humidity deviation coefficient, the higher the risk of equipment malfunction, indicating that the current ambient humidity deviates from the normal level. At the same time, the environmental conformity coefficient represents the relationship between the equipment temperature and the ambient temperature. The smaller the environmental conformity coefficient, the higher the risk of equipment malfunction, indicating that the current equipment temperature and the ambient temperature are not changing in sync.

[0056] S13: Construct voltage and current sequences based on device voltage and current data, and construct a first trend index based on the similarity between the two sequences.

[0057] The system collects current and voltage data. For the current and voltage sequences at each acquisition time, the STL decomposition algorithm is used to obtain the trend term sequences for the current and voltage sequences. The trend term sequences for both sequences are then normalized to obtain the standard trend sequences for the current and voltage sequences, respectively. The Pearson correlation coefficient between the two standard trend sequences is used as the first trend index.

[0058] S14: Determine the second trend index based on the changing trend between equipment temperature data and equipment current data, and use the sum of the first trend index and the second trend index as the equipment pseudo-anomaly factor.

[0059] For any given moment, the difference between the device temperature data at that moment and the adjacent device temperature data before that moment is obtained, and this difference is taken as the first difference; the difference between the current data at that moment and the adjacent current data before that moment is obtained, and this difference is taken as the second difference; a second trend index is calculated based on the first difference and the second difference, wherein the formula for calculating the second trend index is: In the formula, Indicates time The corresponding second trend index; Indicates the first difference; Indicates the second difference; This represents the step function.

[0060] Under normal circumstances, voltage and current in electrical equipment are positively correlated, with current changing as voltage changes. However, when the equipment temperature rises, the resistivity in the current loop increases, leading to a decrease in current. Therefore, analyzing the overall trends of current and voltage, as well as the trends of temperature and current, can reflect the risk of equipment malfunctions. In the calculation of the equipment pseudo-anomaly factor, when the ratio of the first difference to the second difference is negative, the second trend index is 0; when the ratio is positive, the second trend index is 1, causing a sudden change in the final calculated equipment pseudo-anomaly factor. If the first difference is positive, it indicates that the equipment temperature has increased. If the second difference is also positive, it suggests that the current increase in current may be causing the temperature rise, which could be due to a normal change in the equipment's operating state. In this case, the temperature rise is not due to equipment malfunction.

[0061] If the first difference is positive and the second difference is negative, it indicates that the equipment temperature is rising while the current is decreasing. This situation may be due to an abnormal equipment condition causing the temperature to rise, which in turn increases the resistance in its current loop, ultimately leading to a decrease in current. Similarly, if the first trend index is small, it indicates that the correlation between voltage and current changes is weak, suggesting that the current changes in current and voltage are not typical of normal equipment operation, and that the equipment may be experiencing an abnormality.

[0062] S15: For each time point, the mean of the environmental anomaly factors from multiple time points prior to each time point is used as the standard influence coefficient, and the mean of the equipment pseudo-anomaly factors from multiple time points prior to each time point is used as the standard pseudo-anomaly coefficient; the ratio between the standard pseudo-anomaly coefficient and the standard influence coefficient is used as the adjustment coefficient corresponding to each time point.

[0063] In the calculation of the adjustment coefficient, the environmental anomaly factor indicates whether the current tunnel environment is at a normal level. A larger environmental anomaly factor indicates a higher risk of an abnormal environment. In cases of environmental anomalies, the cutoff factor of the subsequent DPC algorithm should be reduced to further capture the details of the monitored data, thus preventing the erroneous merging of clusters of different densities and the failure to detect abnormal data. Conversely, a smaller environmental anomaly factor indicates a lower risk of tunnel environmental anomalies, so the cutoff distance should be increased to reduce the likelihood of normal data points being detected as anomalies.

[0064] Similarly, the larger the pseudo-anomaly factor of the equipment, the higher the probability that the current state of the equipment is not abnormal. Under the current circumstances, the probability that the data to be monitored collected is abnormal is smaller. The cutoff distance should be increased to further reduce the situation where data points are mistakenly detected as abnormal data.

[0065] S2: The product of the adjustment coefficient and the preset initial cutoff distance is used as the optimal cutoff distance. Based on the optimal cutoff distance, the DPC algorithm is used to detect anomalies in electrical equipment.

[0066] For any given moment in the data acquisition process, there is a corresponding adjustment coefficient. The product of the adjustment coefficient and a preset initial cutoff distance is taken as the optimal cutoff distance. During the monitoring of electrical equipment, data to be monitored, such as equipment vibration, power, and current, are collected in real time. Simultaneously, the adjustment coefficient at the corresponding moment is calculated to obtain the optimal cutoff distance. Based on the optimal cutoff distance, the DPC algorithm is used to cluster the monitored data to identify outliers, thus achieving the monitoring of electrical equipment. The method of obtaining abnormal data using the DPC algorithm is a conventional technique in this field and will not be elaborated here.

[0067] This application also discloses an IoT-based tunnel electrical safety monitoring system, including a processor and a memory. The memory stores computer program instructions, which, when executed by the processor, implement the IoT-based tunnel electrical safety monitoring method according to this application.

[0068] The system also includes other components well known to those skilled in the art, such as communication buses and communication interfaces, the settings and functions of which are known in the art and will not be described in detail here.

[0069] The above are all preferred embodiments of this application, and are not intended to limit the scope of protection of this application. Therefore, all equivalent changes made in accordance with the structure, shape and principle of this application should be covered within the scope of protection of this application.

Claims

1. A tunnel electrical safety monitoring method based on the Internet of Things, characterized in that, Construct an adjustment coefficient; use the product of the adjustment coefficient and the preset initial cutoff distance as the optimal cutoff distance, and use the DPC algorithm to detect anomalies in electrical equipment based on the optimal cutoff distance; The steps for constructing the adjustment coefficient include: acquiring tunnel environmental humidity data and determining the humidity deviation coefficient at each moment; constructing a temperature matching coefficient for changes in environmental temperature and equipment temperature, and using the ratio of the normalized result of the humidity deviation coefficient to the normalized result of the temperature matching coefficient as an environmental anomaly factor; constructing voltage and current sequences based on equipment voltage and current data, and constructing a first trend index based on the similarity between the two sequences, including: using the STL decomposition algorithm to obtain the trend term sequences of the current and voltage sequences and performing normalization processing to obtain the standard trend sequence corresponding to the current sequence and the standard trend sequence corresponding to the voltage sequence, and using the Pearson correlation coefficient between the standard trend sequence of the current sequence and the standard trend sequence corresponding to the voltage sequence as the first trend index; determining a second trend index based on the changing trends between equipment temperature data and equipment current data, including: for any given moment, obtaining the difference between the equipment temperature data at that moment and the adjacent equipment temperature data before that moment, and using this difference as the first difference; obtaining the difference between the current data at that moment and the adjacent current data before that moment, and using this difference as the second difference; calculating the second trend index based on the first and second differences, wherein the calculation formula for the second trend index is: In the formula, Indicates time The corresponding second trend index; Indicates the first difference; Indicates the second difference; Let represent the step function, and use the sum of the first trend index and the second trend index as the equipment pseudo-anomaly factor; for each time moment, use the mean of the environmental anomaly factors of multiple time moments before each time moment as the standard influence coefficient, and use the mean of the equipment pseudo-anomaly factors of multiple time moments before each time moment as the standard pseudo-anomaly coefficient; the ratio between the standard pseudo-anomaly coefficient and the standard influence coefficient is used as the adjustment coefficient corresponding to each time moment.

2. The tunnel electrical safety monitoring method based on the Internet of Things according to claim 1, characterized in that, The steps for obtaining tunnel environmental humidity data and determining the humidity deviation coefficient at each moment from the normal humidity include: for any given moment, obtaining the humidity trend at that moment; obtaining the range of normal humidity data and taking the mean of the boundary values ​​of the range as the humidity center; taking the absolute difference between the humidity trend and the humidity center as the deviation value, and taking the ratio of the deviation value to the interval length of the range as the humidity deviation coefficient. The steps for obtaining the humidity trend include: for any given moment, taking the average of the first preset number of consecutive adjacent humidity data points before that moment as the humidity trend at that moment.

3. The tunnel electrical safety monitoring method based on the Internet of Things according to claim 1, characterized in that, The steps for constructing a temperature matching coefficient based on the correlation between ambient temperature changes and equipment temperature changes include: for any given time, constructing the corresponding nearby ambient temperature sequence and nearby equipment temperature sequence; using wavelet coherence method to obtain the wavelet coherence coefficient between the nearby ambient temperature sequence and the nearby equipment temperature sequence and drawing a wavelet coherence diagram; based on the wavelet coherence diagram, obtaining the coherence intensity of the correlation between the changes in the nearby ambient temperature sequence and the nearby equipment temperature sequence, and obtaining the isotopic intensity based on the trend of their changes; and using the product of the isotopic intensity normalization result and the coherence intensity normalization result as the temperature matching coefficient.

4. The tunnel electrical safety monitoring method based on the Internet of Things according to claim 3, characterized in that, For any given moment, a second preset number of ambient temperature data points prior to that moment are acquired to form a nearby ambient temperature sequence; a second preset number of device temperature data points prior to that moment are acquired to form a nearby device temperature sequence.

5. The tunnel electrical safety monitoring method based on the Internet of Things according to claim 3, characterized in that, The ratio of the red and yellow regions in the wavelet coherence graph to the area of ​​the entire wavelet coherence graph is taken as the coherence intensity. In wavelet coherence maps, the highlighted areas represent the correlation strength between the changes of two sequences. Therefore, the ratio of the red and yellow areas in the wavelet coherence map to the area of ​​the entire wavelet coherence map is used as the coherence strength.

6. The tunnel electrical safety monitoring method based on the Internet of Things according to claim 5, characterized in that, The steps for obtaining isotopic intensity include: obtaining the main trend of the arrow directions in the red and yellow regions of the wavelet coherence graph; and taking the ratio of the number of arrows corresponding to the main trend to the total number of arrows in the red and yellow regions as isotopic intensity. In wavelet coherence diagrams, the direction of the arrows indicates the phase difference between the changes of the two sequences; In the wavelet coherence diagram, if the arrow points horizontally to the right, it indicates that the ambient temperature sequence and the equipment temperature sequence change synchronously with no significant phase difference; if the arrow points vertically upward, it indicates that the phase difference between the two sequences is +90°.

7. A tunnel electrical safety monitoring system based on the Internet of Things, characterized in that, include: A processor and a memory, the memory storing computer program instructions that, when executed by the processor, implement the Internet of Things-based tunnel electrical safety monitoring method according to any one of claims 1-6.

Images