Electric power engineering field personnel supervision system and method based on artificial intelligence

Abstract

The invention relates to the technical field of personnel supervision based on artificial intelligence, and particularly discloses an electric power engineering field personnel supervision system and method based on artificial intelligence. According to the electromagnetic environment characteristics of an electric power engineering field, an electromagnetic shielding room is built, the electromagnetic interference condition is simulated, and response data of an intelligent safety helmet is tested; the stability of the system in a high electromagnetic interference environment is evaluated; calculating a working frequency abnormal index and a signal intensity abnormal index of the intelligent safety helmet by using algorithms such as fast Fourier transform and wavelet transform, so as to evaluate the communication capability of the intelligent safety helmet; based on test results of the intelligent safety helmets, the intelligent safety helmets are divided into qualified intelligent safety helmets and unqualified intelligent safety helmets, and it is ensured that only qualified equipment is used for supervising constructors; by monitoring the positions, health conditions and environment change data of workers in real time, a risk assessment model is established by using a gradient lifting tree algorithm, the risk coefficient of a construction site is calculated, and corresponding personnel supervision and early warning response are implemented according to the risk level.

Description

Technical Field

[0001] The present invention relates to the technical field of artificial intelligence personnel supervision, and in particular to an artificial intelligence-based power engineering site personnel supervision system and method. Background Art

[0002] With the continuous advancement of Internet of Things technology, sensor technology and artificial intelligence technology, traditional personnel supervision methods are gradually being replaced by modern intelligent supervision systems. By integrating multiple sensors such as location tracking, health monitoring, and environmental change monitoring, the intelligent supervision system can obtain the location information, health status, and environmental conditions of construction personnel in real time, providing important data support for the safety management of the construction site. However, in high-risk environments, such as near substations and high-voltage power lines, the stability and reliability of intelligent devices are often affected by electromagnetic interference, resulting in unstable signal transmission, which in turn affects the accuracy of real-time monitoring and the timeliness of emergency response.

[0003] The prior art has the following deficiencies:

[0004] The existing smart helmets have the problem of insufficient electromagnetic compatibility in the application of power engineering sites. Specifically, the existing smart helmets often fail to fully consider the complex and high-intensity electromagnetic interference in the power engineering environment, which may cause microprocessor operation errors, unstable wireless communication module connection or data transmission interruption in the actual use of the equipment, thereby affecting the effectiveness of workers' safety monitoring and personal protection functions. To make up for this deficiency, by accurately simulating the electromagnetic environment under actual working conditions, the smart helmets are strictly tested for electromagnetic compatibility and evaluated for performance to ensure that they can maintain stable and reliable operation under strong electromagnetic radiation conditions, thereby improving the safety management level and work efficiency of power engineering sites. Summary of the invention

[0005] The purpose of the present invention is to provide an artificial intelligence-based power engineering site personnel supervision system and method to solve the above-mentioned background problems.

[0006] The purpose of the present invention can be achieved through the following technical solutions:

[0007] The artificial intelligence-based power engineering site personnel supervision method includes the following steps:

[0008] S1: According to the electromagnetic environment characteristics of the power engineering site, determine the electromagnetic frequency range and electromagnetic intensity level of the test to simulate the actual electromagnetic interference situation;

[0009] Design electromagnetic shielding room to ensure that electromagnetic interference conditions are adjustable;

[0010] S2: Place the smart helmet in an electromagnetic shielding room, gradually increase the electromagnetic field strength, record the response data of the smart helmet in real time, analyze the real-time response data, and evaluate the stability of data transmission of the smart helmet;

[0011] The response data includes: operating frequency and signal strength, where the signal strength is the strength of the wireless signal received or sent by the smart helmet;

[0012] S3: Based on the evaluation results, the smart helmets are divided into qualified smart helmets and unqualified smart helmets. Qualified helmets are used to supervise the personnel at the power engineering site. By real-time monitoring of the workers' location, health status and environmental changes, it is predicted whether there are risks at the power engineering site;

[0013] S4: Based on the prediction results, the construction risk level of the power engineering site is divided, and the on-site personnel are supervised according to the risk level. An early warning response mechanism is adopted to issue corresponding early warnings for different levels of risks.

[0014] As a further solution of the present invention: the analysis of the real-time response data specifically includes:

[0015] Place the smart helmet in an electromagnetic shielding room, gradually increase the electromagnetic field strength, and record the response data of the smart helmet in real time, including the operating frequency and signal strength;

[0016] The working frequency of the smart helmet is obtained according to the electromagnetic field intensity level, and the working frequency abnormality index is calculated according to the fluctuation of the working frequency. The frequency stability of the smart helmet under electromagnetic interference is evaluated according to the working frequency abnormality index;

[0017] The signal strength of the smart helmet is obtained according to the electromagnetic field strength level. The signal strength anomaly index is calculated according to the fluctuation of the signal strength. The signal stability of the smart helmet under electromagnetic interference is evaluated according to the signal strength anomaly index.

[0018] As a further solution of the present invention: the process of obtaining the abnormal working frequency index is as follows:

[0019] Collect the working frequency data f(t) of the smart helmet in real time, and calculate the working frequency data as follows:

[0020]

[0021] Where, t represents the time point of the working frequency data acquisition, f represents the frequency in the frequency domain, n represents the number of acquisition points, N represents the total number of acquisition points, j represents the imaginary unit, e represents the logarithm of the base of the natural number, X(f) represents the complex amplitude value corresponding to the frequency f, and x(n) is the working frequency f(t) of the nth acquisition point represented by the time domain signal;

[0022] Get the frequency with the largest amplitude in the spectrum, that is, the working frequency of the smart helmet, recorded as f0;

[0023] Obtaining a bandwidth Δf of a spectrum in the spectrum;

[0024] Calculate the harmonic energy, the calculation expression is:

[0025]

[0026] In the formula, E h represents harmonic energy, X(h*f0) represents the amplitude value corresponding to the frequency h*f0, h represents the order of the harmonic, and H represents the maximum harmonic order;

[0027] Calculate the degree of deviation of the operating frequency, recorded as frequency deviation, and the calculation expression is:

[0028]

[0029] In the formula, Δf(t) represents the frequency deviation at time point t, f0(t) represents the actual operating frequency at time point t, Indicates the ideal operating frequency without electromagnetic interference;

[0030] The standard deviation of the frequency deviation is calculated by the standard deviation calculation formula, denoted as σ f ;

[0031] The bandwidth under the maximum electromagnetic field strength is calculated by the difference between the bandwidth under the maximum electromagnetic field strength and the bandwidth under the maximum electromagnetic field strength to obtain the bandwidth extreme difference value, which is recorded as Δf bw ;

[0032] The harmonic energy E h , standard deviation of frequency deviation σ f and bandwidth range Δf bw Perform standardization and calculate the working frequency abnormality index A f .

[0033] As a further solution of the present invention: the process of obtaining the signal strength anomaly index is:

[0034] Collect the wireless signal strength data received or sent by the smart helmet, recorded as S(t);

[0035] The signal strength data is converted from the time domain to the time-frequency domain through wavelet transform, and then from the time domain to the time-frequency domain. The calculation expression is:

[0036]

[0037] Where W(a,b) represents the local characteristics of the signal strength data S(t) at scale a and position b. It is a wavelet function, which represents the result of the mother wavelet at scale a and position b, where a is the scale parameter and b is the position parameter;

[0038] The standard deviation of the wavelet transform coefficient W(a,b) is calculated to reflect the fluctuation degree of the signal strength data. The fluctuation degree of the signal strength data at each scale, that is, the standard deviation of the wavelet transform coefficient W(a,b), is weighted averaged to obtain the comprehensive standard deviation of the wavelet transform coefficient.

[0039] Calculating the expected deviation of the electromagnetic shielding efficiency, including: calculating the difference between the signal strength when there is no electromagnetic interference and the signal strength under electromagnetic interference to obtain the expected deviation of the electromagnetic shielding efficiency;

[0040] The signal strength anomaly index A is obtained by summing the comprehensive standard deviation and the expected deviation of electromagnetic shielding efficiency. s .

[0041] As a further solution of the present invention: the evaluation of the stability of data transmission of the smart helmet specifically includes:

[0042] The frequency stability of the smart helmet under electromagnetic interference and the signal stability of the smart helmet under electromagnetic interference are comprehensively analyzed, and the comprehensive coefficient is calculated. The stability of the data transmission of the smart helmet is judged according to the comprehensive coefficient.

[0043] As a further solution of the present invention: the smart helmets are divided into qualified smart helmets and unqualified smart helmets according to the evaluation results, specifically including:

[0044] The working frequency abnormality index A f and signal strength anomaly index A s Perform normalization processing and calculate the comprehensive coefficient;

[0045] Determine whether the comprehensive coefficient of each smart helmet is greater than or equal to the preset threshold. If so, it is recorded as an unqualified helmet; if not, it is recorded as a qualified helmet.

[0046] As a further solution of the present invention: the use of qualified safety helmets to supervise personnel at the power engineering site, calculating the risk coefficient, and predicting whether there is a risk at the power engineering site based on the risk coefficient, specifically includes:

[0047] Obtain workers’ location data;

[0048] Obtain workers’ health status data including: heart rate and body temperature;

[0049] Obtain data on the workers’ environment changes, including ambient temperature and humidity;

[0050] The worker's location data, health status data and environmental change data are used to construct feature vectors, and all feature vectors are used to establish a feature matrix x;

[0051] Obtain the location data, health status data, and environmental change data of historically normal workers as training data;

[0052] Preprocess the training data, select the gradient boosting tree to build the risk assessment model, and calculate the risk coefficient;

[0053] The real-time location data, health status data and environmental change data of workers are used as input features, and the risk factor of power engineering is used as the output item of the model;

[0054] Set the initial model and initialize it by the mean;

[0055] For the mth tree, calculate the residual of each training set;

[0056] Train a new tree to fit the residuals;

[0057] Train a new tree to fit the residuals;

[0058] Train multiple trees iteratively until the maximum number of trees is reached;

[0059] The final output of the model is the risk factor;

[0060] The calculation expression of the risk coefficient is:

[0061]

[0062] In the formula, R(x) represents the final predicted risk factor, F0(x) represents the initial predicted value, and T m (x) represents the output of the mth tree, and η represents the learning rate.

[0063] As a further solution of the present invention: the construction risk level of the power engineering site is divided according to the prediction results, specifically including:

[0064] It is determined whether the risk coefficient is greater than or equal to a preset threshold. If so, it indicates that the corresponding power engineering worker has a high risk level, which is recorded as high risk. If not, it indicates that the corresponding power engineering worker has a low risk level, which is recorded as low risk.

[0065] The artificial intelligence-based power engineering site personnel supervision system includes:

[0066] Electromagnetic environment simulation and testing module, which simulates the electromagnetic environment of the power engineering site to determine the electromagnetic frequency range and intensity level that need to be tested; designs and builds an electromagnetic shielding room to ensure that electromagnetic interference is adjustable, providing a standardized environment for the electromagnetic compatibility test of smart helmets;

[0067] A smart helmet response data collection and analysis module, wherein the smart helmet response data collection and analysis module places the smart helmet in an electromagnetic shielding room, gradually increases interference under different electromagnetic field strengths, and collects response data of the smart helmet in real time; by analyzing the response data of the smart helmet, the data transmission stability of the smart helmet under electromagnetic interference is evaluated;

[0068] Smart helmet assessment and personnel supervision module, which divides smart helmets into qualified helmets and unqualified helmets. Qualified helmets are applied to personnel supervision at power engineering sites, including real-time monitoring of workers' positions, health conditions and environmental changes, and prediction of potential risks at power engineering sites based on data analysis;

[0069] The risk prediction and early warning response module divides the construction risk level of the power engineering site; according to the different risk levels, corresponding personnel supervision measures are taken, and risk warning and management are carried out through the early warning response mechanism.

[0070] Beneficial effects of the present invention:

[0071] (1) The present invention conducts electromagnetic interference testing of smart helmets in an electromagnetic shielding room, and uses a highly adjustable electromagnetic field environment to accurately simulate the electromagnetic interference conditions that may be encountered at the actual power engineering site; the working frequency and signal strength of the smart helmet under different electromagnetic field strengths are monitored in real time, and the frequency fluctuation and signal strength change are analyzed in combination with advanced technologies such as Fourier transform and spectrum analysis, and the working frequency abnormality index and signal strength abnormality index are calculated; these indicators can comprehensively reflect the anti-interference ability and data transmission stability of the smart helmet under high electromagnetic interference, thereby providing a scientific basis for the electromagnetic compatibility of the smart helmet; in a complex electromagnetic environment, the smart helmet needs to maintain efficient and stable communication and data transmission capabilities, which is crucial to ensuring the safety of personnel at the power engineering site; the present invention ensures that the smart helmet can effectively suppress external electromagnetic interference during the actual construction process by accurately calculating and comparing the abnormal degree of the working frequency and signal strength, and ensures the real-time transmission and processing of key data such as the worker's location and health status; in this way, not only the technical stability of the smart helmet is improved, but also a strong technical guarantee is provided for the safety of personnel at the power engineering site, the anti-interference and reliability of the equipment are enhanced, and the safety protection level of the entire power construction site is improved;

[0072] (2) The present invention comprehensively monitors the location, health status and environmental changes of workers at the power engineering site in real time, and uses the gradient boosting tree algorithm to conduct in-depth analysis of multi-dimensional data to calculate an accurate risk coefficient. This method can not only capture the changes in the workers' environment in real time, but also establish an efficient risk assessment model in combination with historical data, and dynamically predict the risk level in the construction process according to different working environments and workers' health conditions. Through accurate risk prediction, the present invention achieves comprehensive control of construction site risks and provides a scientific basis for subsequent safety decisions. With the help of this accurate risk assessment system, the present invention can automatically identify high-risk levels, dynamically adjust personnel supervision strategies according to risk factors, and formulate corresponding safety management plans. When the system detects potential dangers or abnormal health of workers, it can immediately trigger an early warning response mechanism to ensure that timely and effective emergency measures are taken to minimize the possibility of accidents. BRIEF DESCRIPTION OF THE DRAWINGS

[0073] The present invention will be further described below in conjunction with the accompanying drawings.

[0074] Figure 1 It is a flowchart of the specific steps of the power engineering site personnel supervision method based on artificial intelligence of the present invention;

[0075] Figure 2 It is a flowchart of the artificial intelligence-based power engineering on-site personnel supervision system in the present invention. DETAILED DESCRIPTION

[0076] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0077] See also Figure 1 As shown, the present invention is a method for supervising personnel on-site of power engineering projects based on artificial intelligence, comprising the following steps:

[0078] S1: According to the electromagnetic environment characteristics of the power engineering site, determine the electromagnetic frequency range and electromagnetic intensity level of the test to simulate the actual electromagnetic interference situation;

[0079] Design electromagnetic shielding room to ensure that electromagnetic interference conditions are adjustable;

[0080] S2: Place the smart helmet in an electromagnetic shielding room, gradually increase the electromagnetic field strength, record the response data of the smart helmet in real time, analyze the real-time response data, and evaluate the stability of data transmission of the smart helmet;

[0081] Wherein, the response data includes: operating frequency and signal strength;

[0082] S3: Based on the evaluation results, the smart helmets are divided into qualified smart helmets and unqualified smart helmets, and qualified helmets are used to supervise the personnel on the power engineering site, including: predicting whether there are risks on the power engineering site by real-time monitoring of the workers' location, health status and environmental changes;

[0083] S4: Based on the prediction results, the construction risk level of the power engineering site is divided, and the on-site personnel are supervised according to the risk level. An early warning response mechanism is adopted to issue corresponding early warnings for different levels of risks.

[0084] In S1, the electromagnetic frequency range and electromagnetic intensity level of the test are determined according to the electromagnetic environment characteristics of the power engineering site to simulate the actual electromagnetic interference situation, including:

[0085] According to the actual electromagnetic environment characteristics of the power engineering site, the main interference sources and their frequency ranges (such as 50Hz to 3GHz) and electromagnetic field strength levels (such as 1V / m to 200V / m) are identified, and an electromagnetic shielding room that effectively shields electromagnetic interference is designed and built. The shielding structure is constructed with highly conductive metal materials and absorbing materials to ensure shielding of external interference. At the same time, a built-in adjustable electromagnetic interference generation system is used to generate multi-band and multi-mode electromagnetic fields through signal generators and radiating antennas. The high-precision electromagnetic field detectors and central control systems are used to monitor and adjust the test environment in real time, providing accurate and reliable simulation conditions for the electromagnetic compatibility testing of smart helmets under various actual working conditions.

[0086] In S2, the smart helmet is placed in an electromagnetic shielding room, the electromagnetic field strength is gradually increased, the response data of the smart helmet is recorded in real time, the real-time response data is analyzed, and the stability of the data transmission of the smart helmet is evaluated, including:

[0087] Place the smart helmet in an electromagnetic shielding room, gradually increase the electromagnetic field strength, and record the response data of the smart helmet in real time, including the operating frequency and signal strength;

[0088] According to the electromagnetic field intensity level, the working frequency of the smart helmet is obtained, and according to the fluctuation of the working frequency, the working frequency abnormality index is calculated, and the frequency stability of the smart helmet under electromagnetic interference is evaluated according to the working frequency abnormality index;

[0089] The signal strength of the smart helmet is obtained according to the electromagnetic field strength level, and the signal strength anomaly index is calculated according to the fluctuation of the signal strength. The signal stability of the smart helmet under electromagnetic interference is evaluated according to the signal strength anomaly index;

[0090] Among them, the signal strength is the strength of the wireless signal received or sent by the smart helmet;

[0091] Comprehensively analyze the frequency stability of the smart helmet under electromagnetic interference and the signal stability of the smart helmet under electromagnetic interference, calculate the comprehensive coefficient, and judge the stability of the data transmission of the smart helmet according to the comprehensive coefficient;

[0092] The process of obtaining the abnormal working frequency index is as follows:

[0093] Collect the working frequency data f(t) of the smart helmet in real time, and calculate the working frequency data as follows:

[0094]

[0095] Where, t represents the time point of the working frequency data acquisition, f represents the frequency in the frequency domain, n represents the number of acquisition points, N represents the total number of acquisition points, j represents the imaginary unit, e represents the logarithm of the base of the natural number, X(f) represents the complex amplitude value corresponding to the frequency f, and x(n) is the working frequency f(t) of the nth acquisition point represented by the time domain signal;

[0096] Get the frequency with the largest amplitude in the spectrum, that is, the working frequency of the smart helmet, recorded as f0;

[0097] Obtaining a bandwidth Δf of a spectrum in the spectrum;

[0098] Calculate the harmonic energy, the calculation expression is:

[0099]

[0100] In the formula, E h represents harmonic energy, X(h*f0) represents the amplitude value corresponding to the frequency h*f0, h represents the order of the harmonic, and H represents the maximum harmonic order;

[0101] Calculate the degree of deviation of the operating frequency, recorded as frequency deviation, and the calculation expression is:

[0102]

[0103] In the formula, Δf(t) represents the frequency deviation at time point t, f0(t) represents the actual operating frequency at time point t, Indicates the ideal operating frequency without electromagnetic interference;

[0104] The standard deviation of the frequency deviation is calculated by the standard deviation calculation formula, denoted as σ f ;

[0105] The bandwidth under the maximum electromagnetic field strength is calculated by the difference between the bandwidth under the maximum electromagnetic field strength and the bandwidth under the maximum electromagnetic field strength to obtain the bandwidth extreme difference value, which is recorded as Δf bw ;

[0106] The harmonic energy E h , standard deviation of frequency deviation σ f and bandwidth range Δf bw Perform standardization and calculate the working frequency abnormality index A f , the calculation expression is:

[0107]

[0108] In the formula, A f Indicates the abnormal working frequency index, a1, a2 and a3 are preset proportional coefficients, and a1, a2 and a3 are all greater than 0;

[0109] comparing the operating frequency abnormality index with a preset threshold;

[0110] If the working frequency abnormality index is greater than or equal to the preset threshold, it means that the corresponding smart helmet is greatly affected by electromagnetic interference and the corresponding working frequency is abnormal;

[0111] If the working frequency abnormality index is less than the preset threshold, it means that the corresponding smart helmet is less affected by electromagnetic interference and the corresponding working frequency is normal;

[0112] It should be noted that: by applying Fourier transform to analyze the working frequency of the smart helmet, the deviation degree of the working frequency, bandwidth change and harmonic energy distribution are calculated, and finally the working frequency anomaly index is generated. The larger the value of the working frequency anomaly index is, the higher the degree of electromagnetic interference affecting the corresponding smart helmet is.

[0113] The process of obtaining the signal strength anomaly index is as follows:

[0114] Collect the wireless signal strength data received or sent by the smart helmet, recorded as S(t);

[0115] The signal strength data is converted from the time domain to the time-frequency domain through wavelet transform, and then from the time domain to the time-frequency domain. The calculation expression is:

[0116]

[0117] Where W(a,b) represents the local characteristics of the signal strength data S(t) at scale a and position b. It is a wavelet function, which represents the result of the mother wavelet at scale a and position b, where a is the scale parameter and b is the position parameter;

[0118] The standard deviation of the wavelet transform coefficient W(a,b) is calculated to reflect the fluctuation degree of the signal strength data. The fluctuation degree of the signal strength data at each scale, that is, the standard deviation of the wavelet transform coefficient W(a,b), is weighted averaged to obtain the comprehensive standard deviation of the wavelet transform coefficient.

[0119] Calculating the expected deviation of the electromagnetic shielding efficiency, including: calculating the difference between the signal strength when there is no electromagnetic interference and the signal strength under electromagnetic interference to obtain the expected deviation of the electromagnetic shielding efficiency;

[0120] The signal strength anomaly index A is obtained by summing the comprehensive standard deviation and the expected deviation of electromagnetic shielding efficiency. s ;

[0121] comparing the signal strength anomaly index with a preset threshold;

[0122] If the signal strength abnormality index is greater than or equal to the preset threshold, it means that the wireless signal transmission capability of the smart helmet under electromagnetic interference is low;

[0123] If the signal strength anomaly index is less than the preset threshold, it means that the smart helmet has high wireless signal transmission capability under electromagnetic interference;

[0124] It should be noted that the signal strength anomaly index reflects the wireless signal transmission capability of the smart helmet under electromagnetic interference. The larger the value of the signal strength anomaly index, the weaker the corresponding wireless signal transmission capability.

[0125] The process of obtaining the comprehensive coefficient is as follows:

[0126] The working frequency abnormality index A f and signal strength anomaly index A s After normalization, the comprehensive coefficient is calculated. The calculation expression is:

[0127]

[0128] In the formula, Z sf represents the comprehensive coefficient, b1 and b2 represent the preset proportional coefficients;

[0129] It should be noted that the comprehensive coefficient is used to reflect whether the smart safety helmet is suitable for use on-site in power engineering projects, and the smaller the value of the comprehensive coefficient, the higher the protection level of the smart safety helmet against electromagnetic interference.

[0130] In S3, the smart helmets are divided into qualified smart helmets and unqualified smart helmets according to the evaluation results, including:

[0131] Determine whether the comprehensive coefficient of each smart helmet is greater than or equal to a preset threshold. If so, it is recorded as an unqualified helmet; if not, it is recorded as a qualified helmet;

[0132] By real-time monitoring of workers' location, health status and environmental changes, the risk factor is calculated and, based on the risk factor, it is predicted whether there are risks at the power engineering site, including:

[0133] Obtain workers’ location data;

[0134] Obtain workers’ health status data including: heart rate and body temperature;

[0135] Obtain data on the workers’ environment changes, including ambient temperature and humidity;

[0136] The worker's location data, health status data and environmental change data are used to construct feature vectors, and all feature vectors are used to establish a feature matrix x;

[0137] Obtain the location data, health status data, and environmental change data of historically normal workers as training data;

[0138] Preprocess the training data, select the gradient boosting tree to build the risk assessment model, and calculate the risk coefficient;

[0139] The real-time location data, health status data and environmental change data of workers are used as input features, and the risk factor of power engineering is used as the output item of the model;

[0140] Set the initial model and initialize it by the mean;

[0141] For the mth tree, calculate the residual of each training set;

[0142] Train a new tree to fit the residuals;

[0143] Train a new tree to fit the residuals;

[0144] Train multiple trees iteratively until the maximum number of trees is reached;

[0145] The final output of the model is the risk factor;

[0146] Through the trained gradient boosting tree model, each worker and construction site is monitored in real time, and the risk coefficient is calculated based on the real-time input features (such as location, health status, environmental changes, etc.). The calculation expression is:

[0147]

[0148] In the formula, R(x) represents the final predicted risk factor, F0(x) represents the initial predicted value, and T m(x) represents the output of the mth tree, and η represents the learning rate;

[0149] Compare the risk factor with a preset threshold;

[0150] If the risk factor is greater than or equal to the preset threshold, it means that the corresponding power engineering worker has a high risk level;

[0151] If the risk factor is less than the preset threshold, it means that the corresponding power engineering worker has a low risk level.

[0152] In S4, the construction risk level of the power engineering site is divided according to the prediction results. According to the risk level, the on-site personnel are supervised and an early warning response mechanism is adopted to issue corresponding early warnings for different levels of risks, including:

[0153] Dynamically monitor on-site personnel based on different risk levels;

[0154] For areas with low risk levels, the monitoring system can conduct routine inspections and data collection to ensure personnel safety;

[0155] For high-risk areas, the system needs to strengthen close monitoring of on-site personnel, track their activities and health status in real time, and ensure that emergency measures are taken;

[0156] According to different risk levels, the system will activate the corresponding early warning response mechanism'

[0157] When the risk level is low, the system will provide regular reminders and reports;

[0158] When the risk level is high, the system immediately issues a strong alarm, automatically activates the emergency response mechanism, and notifies all relevant personnel to evacuate quickly or take emergency safety measures;

[0159] Finally, the implementation of early warning response mechanisms includes real-time location tracking, health monitoring, and environmental data feedback to enable timely identification and handling of abnormal behaviors or unsafe conditions;

[0160] The system will also automatically trigger alarms and transmit information to the on-site command center based on real-time data on the workers' location and health status, ensuring that the safety management of the construction site meets the most efficient and safest standards.

[0161] See also Figure 2 As shown, the artificial intelligence-based power engineering site personnel supervision system includes:

[0162] Electromagnetic environment simulation and testing module, which simulates the electromagnetic environment of the power engineering site to determine the electromagnetic frequency range and intensity level that need to be tested; designs and builds an electromagnetic shielding room to ensure that electromagnetic interference is adjustable, providing a standardized environment for the electromagnetic compatibility test of smart helmets;

[0163] A smart helmet response data collection and analysis module, wherein the smart helmet response data collection and analysis module places the smart helmet in an electromagnetic shielding room, gradually increases interference under different electromagnetic field strengths, and collects response data of the smart helmet in real time; by analyzing the response data of the smart helmet, the data transmission stability of the smart helmet under electromagnetic interference is evaluated;

[0164] Smart helmet assessment and personnel supervision module, which divides smart helmets into qualified helmets and unqualified helmets. Qualified helmets are applied to personnel supervision at power engineering sites, including real-time monitoring of workers' positions, health conditions and environmental changes, and prediction of potential risks at power engineering sites based on data analysis;

[0165] The risk prediction and early warning response module divides the construction risk level of the power engineering site; according to the different risk levels, corresponding personnel supervision measures are taken, and risk warning and management are carried out through the early warning response mechanism.

[0166] The working principle of the present invention is as follows: by simulating the electromagnetic environment characteristics of the power engineering site, designing and building an electromagnetic shielding room, accurately adjusting the electromagnetic frequency range and intensity level, so as to truly reproduce the electromagnetic interference of facilities such as substations, and provide reliable simulation conditions for the electromagnetic compatibility test of smart helmets; placing the smart helmet in the shielding room, gradually increasing the electromagnetic field intensity, collecting response data such as working frequency and signal strength in real time, evaluating the stability of the smart helmet under electromagnetic interference by calculating the frequency anomaly index and the signal strength anomaly index, and comprehensively analyzing and obtaining the comprehensive coefficient to evaluate the data transmission stability of the smart helmet; according to the evaluation results, the smart helmet is divided into qualified and unqualified categories, and the position, health status and environmental changes of the workers are monitored in real time, and the risk coefficient is calculated by using the gradient boosting tree model to predict the construction risk of the power engineering site; according to the predicted risk coefficient, the risk level is divided, the on-site personnel are dynamically supervised, the safety control of high-risk areas is strengthened, and the corresponding early warning response mechanism is started to ensure that accurate safety measures are taken at different risk levels; the present invention can improve the safety of the power engineering construction site through intelligent supervision, timely warn and respond to potential risks, minimize safety hazards, and protect the lives of workers.

[0167] The above formulas are all dimensionless and numerical calculations. The formula is a formula for the most recent real situation obtained by collecting a large amount of data and performing software simulation. The preset parameters in the formula are set by technicians in this field according to actual conditions.

[0168] The above embodiments can be implemented in whole or in part by software, hardware, firmware or any other combination. When implemented by software, the above embodiments can be implemented in whole or in part in the form of a computer program product. The computer program product includes one or more computer instructions or computer programs. When the computer instructions or computer programs are loaded or executed on a computer, the process or function described in the embodiment of the present application is generated in whole or in part. The computer can be a general-purpose computer, a special-purpose computer, a computer network, or other programmable device. The computer instructions can be stored in a computer-readable storage medium, or transmitted from one computer-readable storage medium to another computer-readable storage medium. For example, the computer instructions can be transmitted from one website site, computer, server or data center to another website site, computer, server or data center by wired or wireless (e.g., infrared, wireless, microwave, etc.). The computer-readable storage medium can be any available medium that a computer can access or a data storage device such as a server or data center that contains one or more available media sets. The available medium can be a magnetic medium (e.g., a floppy disk, a hard disk, a tape), an optical medium (e.g., a DVD), or a semiconductor medium. The semiconductor medium can be a solid-state hard disk.

[0169] It should be understood that the term "and / or" in this article is only a description of the association relationship of associated objects, indicating that there can be three relationships. For example, A and / or B can represent: A exists alone, A and B exist at the same time, and B exists alone. A and B can be singular or plural. In addition, the character " / " in this article generally indicates that the associated objects before and after are in an "or" relationship, but it may also indicate an "and / or" relationship. Please refer to the context for specific understanding.

[0170] It should be understood that in the various embodiments of the present application, the size of the serial numbers of the above-mentioned processes does not mean the order of execution. The execution order of each process should be determined by its function and internal logic, and should not constitute any limitation on the implementation process of the embodiments of the present application.

[0171] The above is a detailed description of an embodiment of the present invention, but the content is only a preferred embodiment of the present invention and cannot be considered to limit the scope of implementation of the present invention. All equivalent changes and improvements made within the scope of the present invention should still fall within the scope of the patent coverage of the present invention.

Claims

1. A method for supervising personnel on site of power engineering projects based on artificial intelligence, characterized in that: The following steps are involved: S1: According to the electromagnetic environment characteristics of the power engineering site, determine the electromagnetic frequency range and electromagnetic intensity level of the test to simulate the actual electromagnetic interference situation; Design electromagnetic shielding room to ensure that electromagnetic interference conditions are adjustable; S2: Place the smart helmet in an electromagnetic shielding room, gradually increase the electromagnetic field strength, record the response data of the smart helmet in real time, analyze the real-time response data, and evaluate the stability of data transmission of the smart helmet; The response data includes: operating frequency and signal strength, where the signal strength is the strength of the wireless signal received or sent by the smart helmet; S3: Based on the evaluation results, the smart helmets are divided into qualified smart helmets and unqualified smart helmets. Qualified helmets are used to supervise the personnel at the power engineering site. By real-time monitoring of the workers' location, health status and environmental changes, it is predicted whether there are risks at the power engineering site; S4: Based on the prediction results, the construction risk level of the power engineering site is divided, and the on-site personnel are supervised according to the risk level. An early warning response mechanism is adopted to issue corresponding early warnings for different levels of risks.

2. The artificial intelligence-based power engineering on-site personnel supervision method according to claim 1 is characterized in that: The analysis of the real-time response data specifically includes: Place the smart helmet in an electromagnetic shielding room, gradually increase the electromagnetic field strength, and record the response data of the smart helmet in real time, including the operating frequency and signal strength; The working frequency of the smart helmet is obtained according to the electromagnetic field intensity level, and the working frequency abnormality index is calculated according to the fluctuation of the working frequency. The frequency stability of the smart helmet under electromagnetic interference is evaluated according to the working frequency abnormality index; The signal strength of the smart helmet is obtained according to the electromagnetic field strength level. The signal strength anomaly index is calculated according to the fluctuation of the signal strength. The signal stability of the smart helmet under electromagnetic interference is evaluated according to the signal strength anomaly index.

3. The artificial intelligence-based power engineering on-site personnel supervision method according to claim 2 is characterized in that: The process of obtaining the abnormal working frequency index is as follows: Collect the working frequency data f(t) of the smart helmet in real time, and calculate the working frequency data as follows: Where, t represents the time point of the working frequency data acquisition, f represents the frequency in the frequency domain, n represents the number of acquisition points, N represents the total number of acquisition points, j represents the imaginary unit, e represents the logarithm of the base of the natural number, X(f) represents the complex amplitude value corresponding to the frequency f, and x(n) is the working frequency f(t) of the nth acquisition point represented by the time domain signal; Get the frequency with the largest amplitude in the spectrum, that is, the working frequency of the smart helmet, recorded as f0; Obtaining a bandwidth Δf of a spectrum in the spectrum; Calculate the harmonic energy, the calculation expression is: In the formula, E h represents harmonic energy, X(h*f0) represents the amplitude value corresponding to the frequency h*f0, h represents the order of the harmonic, and H represents the maximum harmonic order; Calculate the degree of deviation of the operating frequency, recorded as frequency deviation, and the calculation expression is: In the formula, Δf(t) represents the frequency deviation at time point t, f0(t) represents the actual operating frequency at time point t, Indicates the ideal operating frequency without electromagnetic interference; The standard deviation of the frequency deviation is calculated by the standard deviation calculation formula, denoted as σ f ; The bandwidth under the maximum electromagnetic field strength is calculated by the difference between the bandwidth under the maximum electromagnetic field strength and the bandwidth under the maximum electromagnetic field strength to obtain the bandwidth extreme difference value, which is recorded as Δf bw ; The harmonic energy E h , standard deviation of frequency deviation σ f and bandwidth range Δf bw Perform standardization and calculate the working frequency abnormality index A f .

4. The artificial intelligence-based power engineering on-site personnel supervision method according to claim 2 is characterized in that: The process of obtaining the signal strength anomaly index is as follows: Collect the wireless signal strength data received or sent by the smart helmet, recorded as S(t); The signal strength data is converted from the time domain to the time-frequency domain through wavelet transform, and then from the time domain to the time-frequency domain. The calculation expression is: Where W(a,b) represents the local characteristics of the signal strength data S(t) at scale a and position b. It is a wavelet function, which represents the result of the mother wavelet at scale a and position b, where a is the scale parameter and b is the position parameter; The standard deviation of the wavelet transform coefficient W(a,b) is calculated to reflect the fluctuation degree of the signal strength data. The fluctuation degree of the signal strength data at each scale, that is, the standard deviation of the wavelet transform coefficient W(a,b), is weighted averaged to obtain the comprehensive standard deviation of the wavelet transform coefficient. Calculating the expected deviation of the electromagnetic shielding efficiency, including: calculating the difference between the signal strength when there is no electromagnetic interference and the signal strength under electromagnetic interference to obtain the expected deviation of the electromagnetic shielding efficiency; The signal strength anomaly index A is obtained by summing the comprehensive standard deviation and the expected deviation of electromagnetic shielding efficiency. s .

5. The artificial intelligence-based power engineering on-site personnel supervision method according to claim 1 is characterized in that: The evaluation of the stability of data transmission of the smart helmet specifically includes: The frequency stability of the smart helmet under electromagnetic interference and the signal stability of the smart helmet under electromagnetic interference are comprehensively analyzed, and the comprehensive coefficient is calculated. The stability of the data transmission of the smart helmet is judged according to the comprehensive coefficient.

6. The artificial intelligence-based power engineering on-site personnel supervision method according to claim 1 is characterized in that: According to the evaluation results, the smart helmets are divided into qualified smart helmets and unqualified smart helmets, specifically including: The operating frequency abnormality index A f and signal strength anomaly index A s Perform normalization processing and calculate the comprehensive coefficient; Determine whether the comprehensive coefficient of each smart helmet is greater than or equal to the preset threshold. If so, it is recorded as an unqualified helmet; if not, it is recorded as a qualified helmet.

7. The artificial intelligence-based power engineering on-site personnel supervision method according to claim 1 is characterized in that: The use of qualified safety helmets to supervise personnel at the power engineering site, calculate the risk factor, and predict whether there is a risk at the power engineering site based on the risk factor, specifically includes: Obtain workers’ location data; Obtain workers’ health status data including: heart rate and body temperature; Obtain data on the workers’ environment changes, including ambient temperature and humidity; The worker's location data, health status data and environmental change data are used to construct feature vectors, and all feature vectors are used to establish a feature matrix x; Obtain the location data, health status data, and environmental change data of historically normal workers as training data; Preprocess the training data, select the gradient boosting tree to build the risk assessment model, and calculate the risk coefficient; The real-time location data, health status data and environmental change data of workers are used as input features, and the risk factor of power engineering is used as the output item of the model; Set the initial model and initialize it by the mean; For the mth tree, calculate the residual of each training set; Train a new tree to fit the residuals; Train a new tree to fit the residuals; Train multiple trees iteratively until the maximum number of trees is reached; The final output of the model is the risk factor; The calculation expression of the risk coefficient is: In the formula, R(x) represents the final predicted risk factor, F0(x) represents the initial predicted value, and T m (x) represents the output of the mth tree, and η represents the learning rate.

8. The artificial intelligence-based power engineering on-site personnel supervision method according to claim 1 is characterized in that: According to the prediction results, the construction risk level of the power engineering site is divided into the following: It is determined whether the risk coefficient is greater than or equal to a preset threshold. If so, it indicates that the corresponding power engineering worker has a high risk level, which is recorded as high risk. If not, it indicates that the corresponding power engineering worker has a low risk level, which is recorded as low risk.

9. The artificial intelligence-based power engineering site personnel supervision system is characterized by: The artificial intelligence-based power engineering site personnel supervision method as claimed in any one of claims 1 to 8 comprises: Electromagnetic environment simulation and testing module, which simulates the electromagnetic environment of the power engineering site to determine the electromagnetic frequency range and intensity level that need to be tested; designs and builds an electromagnetic shielding room to ensure that electromagnetic interference is adjustable, providing a standardized environment for the electromagnetic compatibility test of smart helmets; A smart helmet response data collection and analysis module, wherein the smart helmet response data collection and analysis module places the smart helmet in an electromagnetic shielding room, gradually increases interference under different electromagnetic field strengths, and collects response data of the smart helmet in real time; by analyzing the response data of the smart helmet, the data transmission stability of the smart helmet under electromagnetic interference is evaluated; Smart helmet assessment and personnel supervision module, which divides smart helmets into qualified helmets and unqualified helmets. Qualified helmets are applied to personnel supervision at power engineering sites, including real-time monitoring of workers' positions, health conditions and environmental changes, and prediction of potential risks at power engineering sites based on data analysis; The risk prediction and early warning response module divides the construction risk level of the power engineering site; according to the different risk levels, corresponding personnel supervision measures are taken, and risk warning and management are carried out through the early warning response mechanism.

Images