A thermal radiation risk warning method, device and medium in live working scenarios

Abstract

The present application relates to the field of intelligent sensor temperature measurement technology, and specifically to a method, device, and medium for early warning of thermal radiation risk in live working scenarios. The method comprises: determining the temperature deviation of each intelligent sensor based on the average distribution and discreteness of all temperature data of each intelligent sensor in each time period, as well as the distribution of deviation data; determining the temperature variation similarity based on the correlation of all temperature correction data between any two intelligent sensors; determining the temperature offset value based on the average distribution of all temperature variation similarities in each detection area; and determining the total abnormality score of each intelligent sensor based on the abnormality score of the electromagnetic interference vector to determine the risk assessment coefficient and provide early warning of thermal radiation risk in live working scenarios. The present application aims to eliminate the influence of electromagnetic interference on the accuracy of thermal radiation risk early warning and improve the accuracy of thermal radiation risk early warning in live working scenarios.

Description

Technical Field

[0001] The present application relates to the field of intelligent sensor temperature measurement technology, and specifically to a method, device and medium for early warning of thermal radiation risks in live working scenarios. Background Art

[0002] Live working refers to the maintenance and testing of operating high-voltage electrical equipment, which avoids maintenance outages and ensures normal power supply. High-voltage electrical equipment and transmission lines are installed at high altitudes, and the live working shielding suits worn by power workers have poor breathability, increasing the risk of heat stroke for power workers working live. Due to the unique nature of live working environments, cooling measures cannot be taken immediately upon the onset of mild symptoms. Furthermore, evacuation is difficult and time-consuming, further increasing the risk of heat stroke. Therefore, issuing heat stroke risk warnings to power workers in live working environments and ensuring their timely evacuation are crucial measures to protect their lives.

[0003] Currently, the power industry uses smart sensors to monitor the temperature inside the shielding suits worn by power workers during live-line work in real time, providing early warnings of thermal radiation risks during these operations. The strong electromagnetic environment generated by high-voltage electrical equipment and transmission lines surrounding power workers during live-line work creates significant electromagnetic interference for the smart sensors, affecting their internal circuitry and increasing errors in their measured temperature data. This, in turn, reduces the accuracy of thermal radiation risk warnings for power workers during live-line work. Summary of the Invention

[0004] In a first aspect, an embodiment of the present application provides a method for early warning of thermal radiation risk in a live working scenario, the method comprising the following steps:

[0005] In the live working shield suits of power workers, the temperature data of each smart sensor in different detection areas is obtained at the current moment and all collection moments before the preset time, and the preset time is divided into multiple time periods;

[0006] Based on the average distribution and dispersion of all temperature data of each smart sensor in each time period, all deviation data of each smart sensor in each time period are determined; based on the difference between each deviation data of each smart sensor in each time period and the average distribution of all temperature data, as well as the distribution of all deviation data of each smart sensor within a preset time period, the temperature deviation of each smart sensor in any detection area in each time period is determined;

[0007] For each smart sensor, all temperature data except deviation data within a preset time period is recorded as temperature correction data. Based on the correlation of all temperature correction data between any two smart sensors, the temperature change similarity between any two smart sensors within the same time period is determined;

[0008] Determine the temperature offset value of each smart sensor in any detection area within the same time period based on the average distribution of all the temperature change similarities in each detection area;

[0009] The temperature deviation and temperature offset value of each smart sensor in each time period are combined into an electromagnetic interference vector, and the total abnormality score of each smart sensor is determined based on the abnormality score of the electromagnetic interference vector;

[0010] In each detection area, based on the difference between the total abnormality score of each smart sensor and the extreme distribution of the total abnormality scores of all smart sensors, as well as the average distribution of temperature data of all smart sensors at the current moment and the change trend of all temperature correction data of each smart sensor in each time period, the risk assessment coefficient of each detection area at the current moment is determined, and an early warning of the thermal radiation risk in the live working scenario at the current moment is issued.

[0011] Preferably, the method for determining all deviation data of each smart sensor in each time period is:

[0012] The mean and standard deviation of all temperature data of each smart sensor in each time period are calculated respectively, and the temperature data that differs from the mean by more than three times the standard deviation is used as the deviation data of each smart sensor in each time period.

[0013] Preferably, the method for determining the temperature deviation of each smart sensor in any detection area within each time period is:

[0014] In any detection area, calculate the sum of all deviation data of all smart sensors within a preset time period before the current moment, and record it as the first deviation sum value of any detection area;

[0015] Calculate the cumulative sum of the differences between all deviation data of each smart sensor in each time period and the mean value of all temperature data, and record it as the second deviation sum value of each smart sensor in each time period;

[0016] The temperature deviation of each smart sensor in any detection area in each time period is the result of the fusion of the first deviation sum value and the second deviation sum value.

[0017] Preferably, the method for determining the temperature change similarity between any two smart sensors within the same time period is:

[0018] In each detection area, a cross-correlation sequence of all temperature correction data between any two smart sensors in the same period is obtained, and all elements in the cross-correlation sequence are numbered;

[0019] Calculate the difference between the maximum value in the cross-correlation sequence between any two smart sensors in the same period and the corresponding sequence number of the element value located in the middle of the cross-correlation sequence, and record it as the correlation difference between any two smart sensors in the same period;

[0020] Calculate the cumulative sum of the ratios of all element values ​​in the cross-correlation sequence between any two smart sensors in the same period to the element value located in the middle of the cross-correlation sequence, and record it as the correlation sum value between any two smart sensors in the same period;

[0021] The temperature change similarity between any two smart sensors in the same period is the ratio of the correlation sum to the correlation difference between any two smart sensors in the same period.

[0022] Preferably, the expression for the temperature offset value of each smart sensor in any detection area within the same time period is: Where, represents the temperature offset value of smart sensor m in the rth detection area within the time period s; represents the mean value of the temperature change similarity of all smart sensors in the rth detection area within the time period s; represents the mean of the temperature change similarity of all smart sensors in the rth detection area and the ith detection area within the time period s; represents the mean value of the temperature variation similarity between the smart sensor m in the rth detection area and the smart sensor j in the ith detection area within the time period s; R represents the number of all detection areas; M i Represents the number of all smart sensors in the detection area i; ε represents a preset constant greater than 0.

[0023] Preferably, the method for determining the total abnormality score of each smart sensor is:

[0024] The electromagnetic interference vectors of all smart sensors in all detection areas at all time periods are used as input to the isolation forest algorithm, where the number of isolated trees is set to a preset first value, the number of randomly selected samples is set to a preset second value, and the anomaly score of each electromagnetic interference vector under each isolated tree is output;

[0025] The total abnormal score F of the smart sensor m in the rth detection area r,m The expression is: Where, It represents the time interval between the end time of the period of the wth electromagnetic interference vector of the smart sensor m in the detection area r in the randomly selected sample and the current time; W represents the mean of the abnormal scores of the w-th electromagnetic interference vector of the smart sensor m in the detection area r in the randomly selected samples under all isolated trees; r,m It represents the number of all electromagnetic interference vectors selected as random samples of the smart sensor m in the detection area r in all time periods; ln() represents the logarithmic function with a natural constant as the base; δ represents a constant preset to be greater than 0.

[0026] Preferably, the method for determining the risk assessment coefficient of each detection area at the current moment is:

[0027] Obtain the backward difference sequence of all temperature correction data of each smart sensor in each time period, and calculate the cumulative sum of all elements in the backward difference sequence, which is recorded as the temperature difference sum value of each smart sensor in each time period;

[0028] Calculate the average of all temperature differences and values ​​of all smart sensors in any detection area within a preset time period, and record it as the average temperature difference of any detection area at the current moment;

[0029] The risk assessment coefficient E of the detection area r at the current moment r The expression is: Where, represents the maximum value of the total abnormality scores of all smart sensors in the detection area r at the current moment; F r,m represents the total abnormal score of the mth smart sensor in the detection area r at the current moment; Represents the average temperature data of all smart sensors in the detection area r at the current moment; represents the mean temperature difference of the detection area r at the current moment; exp() represents an exponential function with a natural constant as the base.

[0030] Preferably, the process of providing early warning for the risk of thermal radiation in the live working scenario at the current moment is:

[0031] According to the method for obtaining the risk assessment coefficient of each detection area at the current moment, the risk assessment coefficients of all detection areas at multiple moments during the normal working state of the power staff before the current moment are calculated, and the maximum value of the risk assessment coefficients of all detection areas at multiple moments is used as the risk assessment threshold;

[0032] If the risk assessment coefficient of any detection area at the current moment is greater than the risk assessment threshold, there is a thermal radiation risk in the live working scenario at the current moment; otherwise, there is no thermal radiation risk in the live working scenario at the current moment.

[0033] In a second aspect, an embodiment of the present application provides a thermal radiation risk warning device in a live working scenario, the device comprising:

[0034] The intelligent sensor data acquisition module is used to obtain temperature data of each intelligent sensor in different detection areas at the current moment and all acquisition moments before the preset time period in the live working shield suit of the power workers, and divide the preset time period into multiple time periods;

[0035] The smart sensor data analysis module is used to determine all deviation data of each smart sensor in each time period based on the average distribution and dispersion of all temperature data of each smart sensor in each time period; based on the difference between each deviation data of each smart sensor in each time period and the average distribution of all temperature data, as well as the distribution of all deviation data of each smart sensor within a preset time period, determine the temperature deviation of each smart sensor in any detection area in each time period;

[0036] For each smart sensor, all temperature data except deviation data within a preset time period is recorded as temperature correction data. Based on the correlation of all temperature correction data between any two smart sensors, the temperature change similarity between any two smart sensors within the same time period is determined;

[0037] Determine the temperature offset value of each smart sensor in any detection area within the same time period based on the average distribution of all the temperature change similarities in each detection area;

[0038] The temperature deviation and temperature offset value of each smart sensor in each time period are combined into an electromagnetic interference vector, and the total abnormality score of each smart sensor is determined based on the abnormality score of the electromagnetic interference vector;

[0039] The intelligent sensor early warning module is used to determine the risk assessment coefficient of each detection area at the current moment based on the difference between the total abnormal score of each intelligent sensor and the extreme distribution of the total abnormal scores of all intelligent sensors, the average distribution of temperature data of all intelligent sensors at the current moment, and the changing trend of all temperature correction data of each intelligent sensor in each time period, and to issue an early warning of the thermal radiation risk in the live working scenario at the current moment.

[0040] In a third aspect, an embodiment of the present application also provides a thermal radiation risk warning medium in a live working scenario, comprising a memory, a processor, and a computer program stored in the memory and running on the processor. When the processor executes the computer program, the steps of any one of the above-mentioned thermal radiation risk warning methods in a live working scenario are implemented.

[0041] As can be seen from the above embodiments, the thermal radiation risk warning method in live working scenarios provided by the embodiments of the present application has at least the following beneficial effects:

[0042] This application constructs a temperature deviation by analyzing the distribution and discreteness of temperature data in each time period. By calculating the temperature deviation, the deviation data caused by electromagnetic interference can be identified and eliminated, thereby improving the data quality. The deviation data in the temperature data is further eliminated to obtain temperature correction data. By analyzing the correlation between the temperature correction data of different sensors, the temperature change similarity is constructed, which reflects the similarity of the temperature data change trends between the smart sensors, helps to screen out more accurate temperature data, and improves the accuracy of the heat radiation risk warning. Furthermore, by analyzing the average distribution of the temperature change similarity in each detection area, a temperature offset value is constructed, which helps to reduce the impact of electromagnetic interference on the temperature data. According to the influence of the temperature data, the changing trend of the temperature data can be evaluated more accurately, thereby improving the accuracy of the thermal radiation risk warning; further, by forming an electromagnetic interference vector based on the temperature deviation and temperature offset value, and based on the abnormal score of the electromagnetic interference vector, an abnormal total score is constructed, which reflects the degree to which the smart sensor is affected by electromagnetic interference, helps to identify abnormal data points, improve data quality, more accurately evaluate the thermal radiation risk, and improve the accuracy of the warning. Finally, the risk assessment coefficient is constructed by combining the abnormal total score, the distribution of temperature data at the current moment, and the changing trend of temperature data at all collection moments before the current moment, thereby improving the accuracy of the thermal radiation risk warning in the live working scenario. This application eliminates the problem of inaccurate thermal radiation risk warning caused by electromagnetic interference by analyzing the degree of influence of electromagnetic interference on the accuracy of thermal radiation risk warning, and improves the accuracy of thermal radiation risk warning in the live working scenario. BRIEF DESCRIPTION OF THE DRAWINGS

[0043] In order to more clearly illustrate the technical solutions and advantages of the embodiments of the present application or the prior art, the following is a brief introduction to the drawings required for use in the embodiments or the description of the prior art. Obviously, the drawings described below are only some embodiments of the present application. For ordinary technicians in this field, other drawings can be obtained based on these drawings without any creative work.

[0044] Figure 1 A flowchart of a method for early warning of thermal radiation risk in a live working scenario provided by one embodiment of the present application;

[0045] Figure 2 A schematic diagram of the smart sensor installation area provided in one embodiment of the present application;

[0046] Figure 3 A schematic diagram of a cross-correlation sequence acquisition process provided in one embodiment of the present application;

[0047] Figure 4 A block diagram of a thermal radiation risk warning device for live working scenarios provided in one embodiment of the present application. DETAILED DESCRIPTION

[0048] In order to further illustrate the technical means and effects adopted by this application to achieve the predetermined invention purpose, the following, in conjunction with the accompanying drawings and preferred embodiments, describes in detail the specific implementation method, structure, features and effects of a thermal radiation risk warning method, device and medium for live working scenarios proposed in this application. In the following description, different "one embodiment" or "another embodiment" does not necessarily refer to the same embodiment. In addition, specific features, structures or characteristics in one or more embodiments may be combined in any suitable form.

[0049] Unless defined otherwise, all technical and scientific terms used herein have the same meaning as commonly understood by one of ordinary skill in the art to which this application belongs.

[0050] The following describes in detail a specific scheme of a thermal radiation risk warning method, equipment and medium in a live working scenario provided by this application with reference to the accompanying drawings.

[0051] See also Figure 1 , which shows a flowchart of a method for early warning of thermal radiation risk in a live working scenario provided by one embodiment of the present application, the method comprising the following steps:

[0052] Step S1: In the live working shielding suit of the power worker, the temperature data of each smart sensor in different detection areas at the current moment and all collection moments before the preset time are obtained, and the preset time is divided into multiple time periods.

[0053] Taking into account the high temperature environment in live working scenarios, smart sensors are placed on live working shielding suits to provide real-time warnings of the heat radiation risks of live working workers to ensure their life safety.

[0054] In order to detect the temperature changes at different locations on the live working shielding suit worn by the live working personnel, a total of R detection areas are divided on the live working shielding suit. In this embodiment, R is set to 7, and the smart sensor is DS18B20 smart sensor. The smart sensor installation area diagram is shown in the figure below. Figure 2 As shown, Figure 1 The left image is the front view of the live working shielding suit, and the right image is the rear view of the live working shielding suit. The 7 shaded areas represent different detection areas. Figure 1Serial number 1 represents the right arm area, serial number 2 represents the left arm area, serial number 3 represents the abdomen area, serial number 4 represents the right leg area, serial number 5 represents the left leg area, serial number 6 represents the head area, and serial number 7 represents the back area. M smart sensors are evenly installed in each detection area of ​​the live working shielding suit. In this embodiment, the value of M is 8.

[0055] In the live working shield suit of the power workers, the temperature data of each smart sensor in different detection areas at the current moment and all the collection moments before the preset time are obtained, and the preset time is evenly divided into K time periods, each of which contains the temperature data at S collection moments, where the data sampling frequency is set to f.

[0056] It should be noted that the value of the preset duration, the value of the sampling frequency f, and the values ​​of K and S are all set manually. In this embodiment, the value of the preset duration is 600s, the value of the sampling frequency f is 1Hz, the value of K is 20, and the value of S is 30. The implementer can also set them according to the specific situation. This embodiment does not impose any special restrictions.

[0057] Step S2: Based on the average distribution and discreteness of all temperature data of each smart sensor in each time period, all deviation data of each smart sensor in each time period are determined; based on the difference between each deviation data of each smart sensor in each time period and the average distribution of all temperature data, as well as the distribution of all deviation data of each smart sensor within a preset time period, the temperature deviation of each smart sensor in any detection area in each time period is determined.

[0058] When inspecting high-voltage electrical equipment and transmission lines, power workers' body movements can cause variations in the distance between smart sensors and the equipment in different detection areas. The closer the smart sensor is to the high-voltage equipment, the greater the impact of electromagnetic interference and electrostatic discharge. This transient high electrical interference within the smart sensor's internal circuitry can cause deviations in the detected temperature data. The greater the electrical interference a smart sensor experiences, the more deviated temperature data it reads and the greater the degree of deviation.

[0059] Since the time interval for temperature data reading is short and the changes in temperature data in the same detection area are continuous, the temperature measurements of different smart sensors in the same detection area in the same period can be approximately regarded as multiple continuous temperature data measurements of the same location by the same model of smart sensor. Therefore, all temperature data of different smart sensors in the same detection area in the same period approximately obey Gaussian distribution.

[0060] Therefore, based on the characteristic that the temperature data of different smart sensors in the same detection area in the same period are approximately Gaussian in normal circumstances, the temperature data with large deviations are screened out to improve the accuracy of the heat radiation risk warning. Specifically:

[0061] The mean and standard deviation of all temperature data of each smart sensor in each time period are calculated respectively, and the temperature data that differs from the mean by more than three times the standard deviation is used as the deviation data of each smart sensor in each time period, indicating that the temperature data is seriously affected by electromagnetic interference and has a large deviation.

[0062] It should be noted that the reason why three times the standard deviation is chosen as the screening criterion is because the method used here is the Laida criterion, which complies with the 3σ principle. In this embodiment, when the temperature data differs from the calculated standard deviation by more than three times the standard deviation, the temperature data is regarded as deviation data. The Laida criterion is a well-known technology in statistical methods, and the specific principles and processes of the Laida criterion will not be elaborated on.

[0063] Furthermore, in order to analyze the interference caused by electromagnetic interference on temperature data, thereby reducing the accuracy and timeliness of electromagnetic interference-induced thermal radiation risk warnings, the temperature deviation of each smart sensor in any detection area in each time period is determined by analyzing the difference between the individual deviation data of each smart sensor in each time period and the average distribution of all temperature data, as well as the distribution of all deviation data of each smart sensor within a preset time period. Specifically:

[0064] In any detection area, calculate the sum of all deviation data of all smart sensors within a preset time period before the current moment, and record it as the first deviation sum value of any detection area;

[0065] Further, the cumulative sum of the differences between all deviation data of each smart sensor in each time period and the mean value of all temperature data is calculated, and recorded as the second deviation sum value of each smart sensor in each time period;

[0066] It should be noted that there are many methods for measuring the difference between data. In this embodiment, the square of the difference between all deviation data of each smart sensor in each time period and the mean of all temperature data is calculated as a measure of the difference between the deviation data and the mean of the temperature data. The implementer can also reasonably adopt other methods of measuring the difference between data, such as taking the absolute value or ratio of the difference, based on the specific situation. This embodiment does not impose any special restrictions on the selection of the method for measuring the difference between data.

[0067] Further, based on the first deviation sum value and the second deviation sum value, the temperature deviation of each smart sensor in any detection area in each time period is determined, and the temperature deviation of each smart sensor in any detection area in each time period is the result of the fusion of the first deviation sum value and the second deviation sum value.

[0068] It should be understood that fusion refers to combining two or more indicators through addition or multiplication to obtain a comprehensive indicator, thereby more comprehensively and accurately evaluating a phenomenon or problem. This fusion method is not limited to simple arithmetic operations and can also include more complex statistical models and analysis methods. Implementers can choose according to their specific circumstances and this embodiment does not impose any special restrictions.

[0069] Preferably, in this embodiment, the temperature deviation of each smart sensor in any detection area within each time period is the product of the first deviation sum value and the second deviation sum value; in actual application, as other implementation methods, the temperature deviation of each smart sensor in any detection area within each time period is an exponential function value with a natural constant as the base and the sum of the first deviation sum value and the second deviation sum value as the independent variable.

[0070] Furthermore, based on the temperature deviation of each smart sensor in any detection area within each time period, it can be understood that, on the one hand, the more deviation data of all smart sensors in the detection area in each time period, the more serious electromagnetic interference the smart sensors in the detection area as a whole have been subjected to for a long time, the greater the temperature data deviation, and the greater the obtained temperature deviation; on the other hand, the greater the difference between the deviation data and the mean of the temperature data in each time period, the greater the temperature data deviation caused by the electromagnetic interference suffered by the smart sensor, and the greater the obtained temperature deviation; conversely, if the fewer the deviation data of all smart sensors in the detection area in each time period, and the smaller the difference between the deviation data and the mean of the temperature value, the smaller the obtained temperature deviation, indicating that the smart sensor is less affected by the electromagnetic interference.

[0071] Step S3: For each smart sensor, all temperature data except deviation data within a preset time period is recorded as temperature correction data. Based on the correlation of all temperature correction data between any two smart sensors, the temperature change similarity between any two smart sensors within the same time period is determined.

[0072] Due to differences between smart sensors and their varying resistance to electromagnetic interference, the degree of temperature data drift caused by electromagnetic interference affecting their internal electronic components varies. Furthermore, the movement of power workers during operation causes different detection areas to be illuminated at different times, leading to lags in temperature data between different smart sensors.

[0073] Considering that most smart sensors maintain drift errors within a normal range, the more temperature data acquired, the higher the corresponding data redundancy. To expand the comparison range of temperature data and filter out more accurate temperature data based on the similarity of temperature data change trends, the correlation of all temperature correction data between any two smart sensors is analyzed to determine the temperature change similarity between any two smart sensors in the same period. Specifically,

[0074] First, considering that the deviation data is quite different from the actual temperature data, in order to eliminate the interference of the deviation data, all deviation data within the preset time length are eliminated; further, the linear interpolation method is used to replace the removed deviation data, and all temperature data in each time period after processing are recorded as temperature correction data.

[0075] Among them, the linear interpolation method is a well-known technology in the interpolation method, and its specific principle process of estimating the value of an unknown data point between known data points will not be repeated here.

[0076] Furthermore, in each detection area, a cross-correlation sequence of all temperature correction data between any two smart sensors in the same time period is obtained, and all elements in the cross-correlation sequence are numbered;

[0077] Calculate the difference between the maximum value in the cross-correlation sequence between any two smart sensors in the same period and the corresponding sequence number of the element value located in the middle of the cross-correlation sequence, and record it as the correlation difference between any two smart sensors in the same period;

[0078] Calculate the cumulative sum of the ratios of all element values ​​in the cross-correlation sequence between any two smart sensors in the same period to the element value located in the middle of the cross-correlation sequence, and record it as the correlation sum value between any two smart sensors in the same period;

[0079] Furthermore, the ratio of the correlation sum to the correlation difference between any two smart sensors in the same period is taken as the temperature change similarity between any two smart sensors in the same period.

[0080] It should be noted that the process of obtaining the cross-correlation sequence between two known data sequences is a well-known technique. To facilitate understanding of this embodiment, a simple diagram is provided. The cross-correlation sequence obtaining process is shown in FIG. Figure 3 As shown, Figure 3 In the equation, sequence A = {1, 2, 3}, sequence B = {4, 5, 6}, and the final cross-correlation sequence of sequence A and sequence B is {12, 23, 32, 17, 6}.

[0081] Furthermore, it should be understood that according to the cross-correlation sequence solution process, the number of all elements in the cross-correlation sequence is odd, and there is no even number. Therefore, there is only one element in the middle of the cross-correlation sequence, and there is no zero element.

[0082] Furthermore, based on the temperature variation similarity between any two smart sensors within the same time period, it can be understood that the cross-correlation sequence between any two smart sensors within the same time period reflects the degree of correlation between the temperature data changes between the two smart sensors. On the one hand, the smaller the difference in the corresponding sequence number between the maximum value in the cross-correlation sequence and the value of the element located in the middle of the cross-correlation sequence, the shorter the time interval between the maximum value in the cross-correlation sequence and the value of the element located in the middle of the cross-correlation sequence, indicating a smaller hysteresis in the temperature fluctuation data between the corresponding smart sensors and a greater temperature variation similarity. On the other hand, the larger the ratio of each element in the cross-correlation sequence to the element located in the middle of the cross-correlation sequence, the more gradual the temperature data variation between the corresponding smart sensors, the closer the temperature variation trends of the two smart sensors, and the greater the temperature variation similarity. Conversely, the smaller the ratio of each element in the cross-correlation sequence to the element located in the middle of the sequence, the more severe the temperature data fluctuation between the two smart sensors, the greater the difference in the temperature data variation trends, and the lower the temperature variation similarity between the two smart sensors.

[0083] Step S4: determining the temperature offset value of each smart sensor in any detection area within the same time period based on the average distribution of all the temperature change similarities in each detection area.

[0084] In order to eliminate electromagnetic interference from monitoring and judging thermal radiation risks, thereby protecting the personal safety of power workers, the average distribution of all temperature change similarities in each detection area is analyzed to determine the temperature offset value, thereby improving the accuracy and reliability of temperature data, thereby improving the accuracy of thermal radiation risk warnings. Specifically:

[0085] Temperature offset value of smart sensor m in detection area r during time period s The expression is: Where, represents the mean value of the temperature change similarity of all smart sensors in the rth detection area within the time period s; represents the mean of the temperature change similarity of all smart sensors in the rth detection area and the ith detection area within the time period s; represents the mean value of the temperature variation similarity between the smart sensor m in the rth detection area and the smart sensor j in the ith detection area within the time period s; R represents the number of all detection areas; M irepresents the number of all smart sensors in the detection area i; ε represents a preset constant greater than 0, which is used to prevent the denominator from being 0. The value of ε is set manually. In this embodiment, the value of ε is 0.01. Under the premise of ensuring that the denominator is not 0 and does not excessively affect the calculation results, the implementer can also set it according to the specific situation. This embodiment does not impose any special restrictions.

[0086] According to the temperature offset value of each smart sensor in any detection area during the same period, it can be understood that the larger the mean value of the temperature change similarity of all smart sensors in the detection area during the same period, the greater the influence of the location of the power workers. The mean value of the temperature change similarity of all smart sensors in different detection areas during the same period will decrease. The greater the decrease, the greater the difference in the degree of temperature change in the detection area, which means that the effect of this area on screening more accurate temperature data is smaller. The larger the value, the smaller the calculated temperature offset value. In addition, the smaller the average value of the temperature change similarity between the smart sensor in the detection area and the smart sensor in another detection area during the same period, the closer the temperature data change trends of the two smart sensors are. When the change trend is closer to the change trend of the temperature data read by a larger number of smart sensors, it reflects that the average error caused by electromagnetic interference during the temperature data reading period of the smart sensor is smaller, and the calculated temperature offset value is smaller.

[0087] On the contrary, if The smaller it is, the smaller the difference in temperature data between smart sensors in different detection areas is, and the larger the average value of the temperature change similarity between the smart sensor in the detection area and the smart sensor in another detection area during the same period is, indicating that the difference in the change trend of the temperature data read by the two smart sensors is greater, reflecting that the average error caused by electromagnetic interference during the time period of temperature data reading of the smart sensor is greater, and the calculated temperature offset value is greater.

[0088] Step S5: The temperature deviation and temperature offset value of each smart sensor in each time period are combined into an electromagnetic interference vector, and based on the abnormality score of the electromagnetic interference vector, the total abnormality score of each smart sensor is determined.

[0089] To further identify abnormal data points in temperature data and improve data quality, thereby increasing the accuracy of thermal radiation risk warnings in live working scenarios, the temperature deviation and temperature offset values ​​of each smart sensor in each time period are analyzed to form an electromagnetic interference vector. Based on the abnormality score of the electromagnetic interference vector, the total abnormality score of each smart sensor is determined to filter out abnormal data points and improve the accuracy and reliability of warnings. Specifically:

[0090] The temperature deviation and temperature offset values ​​of each smart sensor in each time period are combined to form the electromagnetic interference vector of each smart sensor in each time period, which represents the instantaneous error caused by strong electromagnetic interference in a short period of time and the drift error that persists for a longer period of time.

[0091] The electromagnetic interference vectors of all smart sensors in all detection areas at all time periods are used as input to the isolation forest algorithm, where the number of isolated trees is set to a preset first value, the number of randomly selected samples is set to a preset second value, and the anomaly score of each electromagnetic interference vector under each isolated tree is output;

[0092] It should be noted that the values ​​of the preset first value and the preset second value are both manually set. In this embodiment, the value of the preset first value is 256, and the value of the preset second value is 100. The implementer can also set them according to the specific situation. This embodiment does not impose any special restrictions.

[0093] Among them, the isolation forest algorithm is a well-known technology in the field of anomaly detection algorithms, and its specific principle process is not repeated here.

[0094] Furthermore, based on the abnormality score of the electromagnetic interference vector, the total abnormality score of each smart sensor is determined, specifically:

[0095] The total abnormal score F of the smart sensor m in the rth detection area r,m The expression is: Where, It represents the time interval between the end time of the period of the wth electromagnetic interference vector of the smart sensor m in the detection area r in the randomly selected sample and the current time; W represents the mean of the abnormal scores of the w-th electromagnetic interference vector of the smart sensor m in the detection area r in the randomly selected samples under all isolated trees; r,m represents the number of all electromagnetic interference vectors selected as random samples by the smart sensor m in the detection area r in all time periods; ln() represents a logarithmic function with a natural constant as the base; δ represents a constant preset greater than 0, which is used to prevent the calculation result from being 0. In this embodiment, the value of δ is 2. Under the premise of ensuring that the calculation result is not 0 and not excessively affecting the calculation result, the implementer can also set it according to the specific situation. This embodiment does not impose any special restrictions.

[0096] Based on the total anomaly score of each smart sensor, it can be understood that the anomaly score of the electromagnetic interference vector output by the isolation forest algorithm in each isolated tree reflects the degree of anomaly of the corresponding smart sensor in the subsample. The higher the anomaly score, the greater the degree of influence of the electromagnetic interference of the high-voltage transmission line on the smart sensor in the corresponding time period. The larger the average anomaly score of the electromagnetic interference vector of the smart sensors in the detection area in the randomly selected sample under all isolated trees. At the same time, the shorter the time interval between the end time of the electromagnetic interference vector and the current time, the closer the electromagnetic interference vector is to the current time, and the greater the impact on the heat radiation risk warning.

[0097] On the contrary, the lower the anomaly score, the less the smart sensor is affected by the electromagnetic interference of the high-voltage transmission line in the corresponding period, and the smaller the average anomaly score of the electromagnetic interference vector of the smart sensor in the detection area in the randomly selected sample under all isolated trees; at the same time, the longer the time interval between the end time of the period where the electromagnetic interference vector is located and the current time, the farther the period where the electromagnetic interference vector is located is from the current time, and the smaller the impact on the thermal radiation risk warning.

[0098] Step S6: In each detection area, based on the difference between the total abnormality score of each smart sensor and the extreme distribution of the total abnormality scores of all smart sensors, as well as the average distribution of temperature data of all smart sensors at the current moment and the difference in temperature data between different smart sensors, the risk assessment coefficient of each detection area at the current moment is determined, and an early warning of the thermal radiation risk in the live working scenario at the current moment is issued.

[0099] Considering the temperature differences between different parts of the human body, when providing heatstroke risk warnings, different detection areas correspond to different parts of the power worker's body, and the temperature thresholds for heatstroke risk are different. Therefore, a backward difference sequence of all temperature correction data from each smart sensor in each time period is obtained, and the cumulative sum of all elements in the backward difference sequence is calculated, which is recorded as the temperature difference sum of each smart sensor in each time period.

[0100] Calculate the cumulative sum of the temperature difference values ​​of all time periods of each smart sensor in any detection area within the preset time period, and record it as the cumulative sum of the temperatures of each smart sensor in any detection area at the current moment, and record the average of the cumulative sum of the temperatures of all smart sensors in any detection area at the current moment as the average temperature difference of any detection area at the current moment.

[0101] The process of obtaining the backward difference sequence is a well-known technique, and the specific steps of obtaining the sequence are not described in detail here.

[0102] Furthermore, based on the difference between the total abnormality score of each smart sensor and the extreme distribution of the total abnormality score of all smart sensors, as well as the average distribution of the temperature data of all smart sensors at the current moment and the difference in temperature data between different smart sensors, the risk assessment coefficient of each detection area at the current moment is determined, specifically:

[0103] The risk assessment coefficient E of the detection area r at the current moment r The expression is: Where, represents the maximum value of the total abnormality scores of all smart sensors in the detection area r at the current moment; F r,m represents the total abnormal score of the mth smart sensor in the detection area r at the current moment; Represents the average temperature data of all smart sensors in the detection area r at the current moment; represents the mean temperature difference of the detection area r at the current moment; exp() represents an exponential function with a natural constant as the base.

[0104] Based on the risk assessment coefficients of each detection area at the current moment, it can be understood that the total abnormality score of the smart sensor reflects the degree to which it is affected by the electromagnetic interference of the high-voltage transmission equipment. The larger the ratio of the maximum total abnormality score of all smart sensors to the abnormality score of each smart sensor, the lower the accuracy of the temperature data, and the greater the weight when calculating the risk assessment coefficient. In addition, the larger the average value of the temperature data of all smart sensors in the detection area at the current moment, and the larger the temperature difference mean of the detection area at the current moment, the larger the calculated risk assessment coefficient.

[0105] On the contrary, the smaller the ratio of the maximum total anomaly score of all smart sensors to the anomaly score value of each smart sensor, the higher the accuracy of the temperature data, and the smaller the weight when calculating the risk assessment coefficient; in addition, the smaller the average value of the temperature data of all smart sensors in the detection area at the current moment, and the smaller the mean temperature difference of the detection area at the current moment, the smaller the calculated risk assessment coefficient.

[0106] According to the method for obtaining the risk assessment coefficient of each detection area at the current moment, the risk assessment coefficients of all detection areas at multiple moments during the normal working state of the power staff before the current moment are calculated, and the maximum value of the risk assessment coefficients of all detection areas at multiple moments is used as the risk assessment threshold;

[0107] If the risk assessment coefficient of any detection area at the current moment is greater than the risk assessment threshold, there is a thermal radiation risk in the live working scenario at the current moment, and the digital signal processor sends an early warning signal to the buzzer to remind the power workers to evacuate in time; conversely, if the risk assessment coefficient risk assessment threshold of all detection areas at the current moment is less than or equal to the risk assessment threshold, there is no thermal radiation risk in the live working scenario at the current moment.

[0108] Based on the same inventive concept as the above method, the embodiment of the present application further provides a thermal radiation risk warning device in a live working scenario, comprising:

[0109] The intelligent sensor data acquisition module is used to obtain temperature data of each intelligent sensor in different detection areas at the current moment and all acquisition moments before the preset time period in the live working shield suit of the power workers, and divide the preset time period into multiple time periods;

[0110] The smart sensor data analysis module is used to determine all deviation data of each smart sensor in each time period based on the average distribution and dispersion of all temperature data of each smart sensor in each time period; based on the difference between each deviation data of each smart sensor in each time period and the average distribution of all temperature data, as well as the distribution of all deviation data of each smart sensor within a preset time period, determine the temperature deviation of each smart sensor in any detection area in each time period;

[0111] For each smart sensor, all temperature data except deviation data within a preset time period is recorded as temperature correction data. Based on the correlation of all temperature correction data between any two smart sensors, the temperature change similarity between any two smart sensors within the same time period is determined;

[0112] Determine the temperature offset value of each smart sensor in any detection area within the same time period based on the average distribution of all the temperature change similarities in each detection area;

[0113] The temperature deviation and temperature offset value of each smart sensor in each time period are combined into an electromagnetic interference vector, and the total abnormality score of each smart sensor is determined based on the abnormality score of the electromagnetic interference vector;

[0114] The intelligent sensor early warning module is used to determine the risk assessment coefficient of each detection area at the current moment based on the difference between the total abnormal score of each intelligent sensor and the extreme distribution of the total abnormal scores of all intelligent sensors, the average distribution of temperature data of all intelligent sensors at the current moment, and the changing trend of all temperature correction data of each intelligent sensor in each time period, and to issue an early warning of the thermal radiation risk in the live working scenario at the current moment.

[0115] The embodiment of the present application provides a block diagram of a thermal radiation risk warning device in a live working scenario, such as Figure 4 shown.

[0116] Based on the same inventive concept as the above-mentioned method, an embodiment of the present application also provides a thermal radiation risk warning medium in a live working scenario, comprising a memory, a processor, and a computer program stored in the memory and running on the processor. When the processor executes the computer program, the steps of any one of the above-mentioned thermal radiation risk warning methods in a live working scenario are implemented.

[0117] It should be noted that the order in which the embodiments of the present application are presented is for illustrative purposes only and does not necessarily represent the superiority or inferiority of the embodiments. Furthermore, the foregoing descriptions of specific embodiments of this specification are provided. Furthermore, the processes depicted in the accompanying drawings do not necessarily require the specific order or sequential sequence shown to achieve the desired results. In certain embodiments, multitasking and parallel processing are also possible or may be advantageous.

[0118] The various embodiments in this specification are described in a progressive manner, and the same or similar parts between the various embodiments can be referred to each other. Each embodiment focuses on the differences from other embodiments.

[0119] The above description is only a preferred embodiment of the present application and is not intended to limit the present application. Any modifications, equivalent replacements, improvements, etc. made within the principles of the present application shall be included in the scope of protection of the present application.

Claims

1. A method for early warning of thermal radiation risk in live working scenarios, characterized in that: The method comprises the following steps: S1. In the live working shield suits of power workers, the temperature data of each smart sensor in different detection areas is obtained at the current moment and all collection moments before the preset time, and the preset time is divided into multiple time periods; S2. Based on the average distribution and dispersion of all temperature data of each smart sensor in each time period, all deviation data of each smart sensor in each time period are determined; based on the difference between each deviation data of each smart sensor in each time period and the average distribution of all temperature data in any detection area, as well as the distribution of all deviation data of each smart sensor within a preset time period, the temperature deviation of each smart sensor in any detection area in each time period is determined; S3. For each smart sensor, all temperature data except deviation data within a preset time period is recorded as temperature correction data. Based on the correlation of all temperature correction data between any two smart sensors, the temperature change similarity between any two smart sensors within the same time period is determined; S4. Determine the temperature offset value of each smart sensor in any detection area within the same time period based on the average distribution of all the temperature change similarities in each detection area; S5. The temperature deviation and temperature offset value of each smart sensor in each time period are combined into an electromagnetic interference vector, and the total abnormality score of each smart sensor is determined based on the abnormality score of the electromagnetic interference vector; S6. In each detection area, based on the difference between the total abnormality score of each smart sensor and the extreme distribution of the total abnormality scores of all smart sensors, as well as the average distribution of temperature data of all smart sensors at the current moment and the change trend of all temperature correction data of each smart sensor in each time period, the risk assessment coefficient of each detection area at the current moment is determined, and an early warning of the thermal radiation risk in the live working scenario at the current moment is issued.

2. The method for early warning of thermal radiation risk in live working scenarios according to claim 1, characterized in that: The method for determining all deviation data of each intelligent sensor in each time period is as follows: The mean and standard deviation of all temperature data of each smart sensor in each time period are calculated respectively, and the temperature data that differs from the mean by more than three times the standard deviation is used as the deviation data of each smart sensor in each time period.

3. The method for early warning of thermal radiation risk in live working scenarios according to claim 1, characterized in that: The method for determining the temperature deviation of each smart sensor in any detection area within each time period is as follows: In any detection area, calculate the sum of all deviation data of all smart sensors within a preset time period before the current moment, and record it as the first deviation sum value of any detection area; Calculate the cumulative sum of the differences between all deviation data of each smart sensor in each time period and the mean value of all temperature data, and record it as the second deviation sum value of each smart sensor in each time period; The temperature deviation of each smart sensor in any detection area in each time period is the result of the fusion of the first deviation sum value and the second deviation sum value.

4. The method for early warning of thermal radiation risk in live working scenarios according to claim 1, characterized in that: The method for determining the temperature change similarity between any two smart sensors in the same period is: In each detection area, a cross-correlation sequence of all temperature correction data between any two smart sensors in the same period is obtained, and all elements in the cross-correlation sequence are numbered; Calculate the difference between the maximum value in the cross-correlation sequence between any two smart sensors in the same period and the corresponding sequence number of the element value located in the middle of the cross-correlation sequence, and record it as the correlation difference between any two smart sensors in the same period; Calculate the cumulative sum of the ratios of all element values ​​in the cross-correlation sequence between any two smart sensors in the same period to the element value located in the middle of the cross-correlation sequence, and record it as the correlation sum value between any two smart sensors in the same period; The temperature change similarity between any two smart sensors in the same period is the ratio of the correlation sum to the correlation difference between any two smart sensors in the same period.

5. The method for early warning of thermal radiation risk in live working scenarios according to claim 1, characterized in that: The expression of the temperature offset value of each smart sensor in any detection area within the same period is: Where, represents the temperature offset value of smart sensor m in the rth detection area within the time period s; represents the mean value of the temperature change similarity of all smart sensors in the rth detection area within the time period s; represents the mean of the temperature change similarity of all smart sensors in the rth detection area and the ith detection area within the time period s; represents the mean value of the temperature change similarity between the smart sensor m in the rth detection area and the smart sensor j in the ith detection area within the time period s; R represents the number of all detection areas; M i represents the number of all smart sensors in detection area i; ε represents a constant that is preset to be greater than 0.

6. The method for early warning of thermal radiation risk in live working scenarios according to claim 1, characterized in that: The method for determining the total abnormality score of each smart sensor is as follows: The electromagnetic interference vectors of all smart sensors in all detection areas at all time periods are used as input to the isolation forest algorithm, where the number of isolated trees is set to a preset first value, the number of randomly selected samples is set to a preset second value, and the anomaly score of each electromagnetic interference vector under each isolated tree is output; The total abnormal score F of the smart sensor m in the rth detection area r,m The expression is: Where, It represents the time interval between the end time of the period of the wth electromagnetic interference vector of the smart sensor m in the detection area r in the randomly selected sample and the current time; It represents the mean of the abnormal scores of the w-th electromagnetic interference vector of the smart sensor m in the detection area r in the randomly selected samples under all isolated trees; W r,m It represents the number of all electromagnetic interference vectors selected as random samples of the smart sensor m in the detection area r in all time periods; ln() represents the logarithmic function with a natural constant as the base; δ represents a constant preset to be greater than 0.

7. The method for early warning of thermal radiation risk in live working scenarios according to claim 1, characterized in that: The method for determining the risk assessment coefficient of each detection area at the current moment is: Obtain the backward difference sequence of all temperature correction data of each smart sensor in each time period, and calculate the cumulative sum of all elements in the backward difference sequence, which is recorded as the temperature difference sum value of each smart sensor in each time period; Calculate the average of all temperature differences and values ​​of all smart sensors in any detection area within a preset time period, and record it as the average temperature difference of any detection area at the current moment; The risk assessment coefficient E of the detection area r at the current moment r The expression is: Where, represents the maximum value of the total abnormality scores of all smart sensors in the detection area r at the current moment; F r,m represents the total abnormal score of the mth smart sensor in the detection area r at the current moment; Represents the average temperature data of all smart sensors in the detection area r at the current moment; represents the mean temperature difference of the detection area r at the current moment; exp() represents an exponential function with a natural constant as the base.

8. The method for early warning of thermal radiation risk in live working scenarios according to claim 1, characterized in that: The process of early warning of the thermal radiation risk in the live working scenario at the current moment is as follows: According to the method for obtaining the risk assessment coefficient of each detection area at the current moment, the risk assessment coefficients of all detection areas at multiple moments during the normal working state of the power staff before the current moment are calculated, and the maximum value of the risk assessment coefficients of all detection areas at multiple moments is used as the risk assessment threshold; If the risk assessment coefficient of any detection area at the current moment is greater than the risk assessment threshold, there is a risk of thermal radiation in the live working scenario at the current moment; On the contrary, there is no risk of thermal radiation in the live working scenario at the current moment.

9. A device for early warning of thermal radiation risk in live working scenarios, which implements the method for early warning of thermal radiation risk in live working scenarios as claimed in claim 1, characterized in that: The device comprises: The intelligent sensor data acquisition module is used to obtain temperature data of each intelligent sensor in different detection areas at the current moment and all acquisition moments before the preset time period in the live working shield suit of the power workers, and divide the preset time period into multiple time periods; The smart sensor data analysis module is used to determine all deviation data of each smart sensor in each time period based on the average distribution and dispersion of all temperature data of each smart sensor in each time period; based on the difference between each deviation data of each smart sensor in each time period and the average distribution of all temperature data, as well as the distribution of all deviation data of each smart sensor within a preset time period, determine the temperature deviation of each smart sensor in any detection area in each time period; For each smart sensor, all temperature data except deviation data within a preset time period is recorded as temperature correction data. Based on the correlation of all temperature correction data between any two smart sensors, the temperature change similarity between any two smart sensors within the same time period is determined; Determine the temperature offset value of each smart sensor in any detection area within the same time period based on the average distribution of all the temperature change similarities in each detection area; The temperature deviation and temperature offset value of each smart sensor in each time period are combined into an electromagnetic interference vector, and the total abnormality score of each smart sensor is determined based on the abnormality score of the electromagnetic interference vector; The intelligent sensor early warning module is used to determine the risk assessment coefficient of each detection area at the current moment based on the difference between the total abnormal score of each intelligent sensor and the extreme distribution of the total abnormal scores of all intelligent sensors, the average distribution of temperature data of all intelligent sensors at the current moment, and the changing trend of all temperature correction data of each intelligent sensor in each time period, and to issue an early warning of the thermal radiation risk in the live working scenario at the current moment.

10. A thermal radiation risk warning medium in a live working scenario, comprising a memory, a processor, and a computer program stored in the memory and running on the processor, characterized in that: When the processor executes the computer program, the steps of the thermal radiation risk warning method in a live working scenario as described in any one of claims 1 to 8 are implemented.

Images