Environmental monitoring and early warning method and system for safety helmets

Abstract

The present invention relates to the technical field of intelligent safety helmets, and in particular to an environment monitoring and early warning method and system for safety helmets, which can timely capture and analyze the trend of environmental changes through continuous dynamic monitoring of environmental parameters in a confined space, evaluate environmental conditions at different heights, provide a scientific basis for predicting changes in harmful gas concentrations, realize personalized early warning and in-depth analysis of environmental parameters, and improve safety protection in different spatial working environments; the method comprises: according to a preset sensor monitoring frequency, continuously acquiring environmental data around the wearer and the height of the safety helmet monitored by an environmental sensor integrated in the safety helmet, the environmental data around the wearer including temperature, humidity and harmful gas concentration, and the height of the safety helmet corresponding to the environmental data around the wearer one by one; based on the height of the safety helmet, clustering analysis is performed on the environmental data around the wearer to obtain a plurality of environmental data sets.

Description

Technical Field

[0001] The invention relates to the technical field of intelligent safety helmets, and in particular to an environment monitoring and early warning method for safety helmets and a system thereof. Background Art

[0002] In confined space working environments, such as mines, underground tunnels, and closed factory workshops, monitoring of environmental conditions is crucial to ensure the safety of workers. Traditional environmental monitoring methods mostly rely on fixed monitoring stations. Although they can cover a large area, in confined spaces, due to the complex spatial structure and frequent flow of personnel, fixed monitoring stations are often difficult to fully and real-time reflect the true conditions of the working environment. In recent years, with the rapid development of Internet of Things technology, sensor-based environmental monitoring technology has gradually received attention. By integrating environmental sensors into wearable devices, such as safety helmets, real-time and continuous monitoring of the environment around workers can be achieved, and safety alarms can be issued based on thresholds.

[0003] However, in a confined space working environment, due to the interaction of multiple factors such as space height, temperature, humidity and harmful gas concentration, the existing safety helmet monitoring method only relies on the sensors integrated on the safety helmet to perform simple threshold judgment alarms, which makes it difficult to accurately predict the trend of environmental changes. Especially when the concentration of harmful gases changes, simple threshold alarms may not be able to detect dangerous situations in time, thus affecting the accuracy and timeliness of the early warning. Summary of the invention

[0004] To solve the above technical problems, the present invention provides an environmental monitoring and early warning method and system for safety helmets, which can continuously and dynamically monitor environmental parameters in a confined space, timely capture and analyze environmental change trends, evaluate environmental conditions at different heights, provide a scientific basis for predicting changes in harmful gas concentrations, achieve personalized early warning and in-depth analysis of environmental parameters, and improve safety in different spatial working environments.

[0005] In a first aspect, the present invention provides an environmental monitoring and early warning method for a safety helmet, the method comprising:

[0006] According to the preset sensor monitoring frequency, continuously obtain the environmental data around the wearer and the height of the helmet monitored by the helmet integrated environmental sensor, the environmental data around the wearer includes temperature, humidity and harmful gas concentration, and the height of the helmet corresponds to the environmental data around the wearer one by one;

[0007] Based on the height of the helmet, cluster analysis is performed on the environmental data around the wearer to obtain several environmental data sets, where each environmental data set includes environmental data around all wearers within the height range of the same helmet;

[0008] Performing trend analysis on each environmental data set to obtain an environmental change feature vector, wherein the environmental change feature vector includes a temperature and humidity change trend feature and a harmful gas concentration change trend feature;

[0009] Inputting the environmental change feature vector into a pre-built gas concentration environmental impact correction model to obtain the actual change trend characteristics of the harmful gas concentration corresponding to the height range of the helmet;

[0010] The actual change trend characteristics of the concentration of harmful gases corresponding to the height range of each helmet are arranged according to the height, so as to obtain the actual change trend sequence of the concentration of harmful gases corresponding to the confined space;

[0011] The actual change trend sequence of the harmful gas concentration is input into the pre-built harmful gas concentration prediction model to obtain the time node when the harmful gas concentration is higher than the preset concentration; and the time node is prompted to the wearer.

[0012] Furthermore, the harmful gases include carbon monoxide, sulfur dioxide and methane.

[0013] Furthermore, the method for obtaining the environmental data set includes:

[0014] Acquire environmental data around the wearer through the helmet's integrated environmental sensor;

[0015] Preprocess the environmental data to remove outliers and missing values. After the removal, standardize the temperature, humidity and harmful gas concentrations.

[0016] Select K-means algorithm to cluster the preprocessed environmental data;

[0017] The collected environmental data is divided into multiple height intervals according to the actual situation according to the height of the helmet, and the environmental data in each height interval is a cluster object;

[0018] Use the K-means algorithm to perform cluster analysis on the data in each height interval, and generate an environmental data set for each height interval, including the surrounding environmental data of all wearers in the height interval;

[0019] After clustering is completed, the silhouette coefficient is used to verify the clustering results and evaluate the quality of clustering.

[0020] Furthermore, the method for acquiring the environmental change feature vector includes:

[0021] Clean the data in each environmental data set and standardize the cleaned data;

[0022] Extract time series data of temperature, humidity and harmful gas concentration from each environmental data set;

[0023] Select the linear regression trend analysis method to fit the time series data and identify the changing trends of temperature, humidity and harmful gas concentration over time;

[0024] According to the linear regression trend recognition algorithm, the changing trend characteristics of temperature, humidity and harmful gas concentration in each environmental data set are calculated;

[0025] The slope of the temperature and humidity change trend characteristics was obtained by linear regression trend analysis method;

[0026] The trend characteristics of harmful gas concentration changes are analyzed by linear regression trend analysis method to obtain the slope of the trend of harmful gas concentration changes;

[0027] The slope of the temperature and humidity change trend and the slope of the harmful gas concentration change trend are combined to obtain the environmental change feature vector.

[0028] Furthermore, the method for constructing the gas concentration environmental impact correction model includes:

[0029] Collect the historical temperature and humidity change trend characteristics and harmful gas concentration change trend characteristics around the wearer monitored by the sensor, as well as the actual change trend characteristics of the historical harmful gas concentration;

[0030] A gas concentration environmental impact correction model was constructed based on a multiple linear regression model. The actual trend characteristics of harmful gas concentration were used as dependent variables, and the trend characteristics of temperature and humidity and harmful gas concentration were used as independent variables. The quantitative relationship between the dependent and independent variables was found through regression analysis.

[0031] The collected environmental change feature vectors are divided into a training set and a test set. The gas concentration environmental impact correction model is trained using the training set data. After the training is completed, the gas concentration environmental impact correction model is tested using the test set.

[0032] The gas concentration environmental impact correction model was verified by cross-validation method.

[0033] Furthermore, the method for obtaining the actual change trend sequence of the harmful gas concentration includes:

[0034] Obtain the actual change trend characteristics of the harmful gas concentration corresponding to the height range of each helmet;

[0035] The actual change trend characteristics of harmful gas concentrations include the actual change trend information of harmful gas concentrations within each altitude range;

[0036] The actual change trend characteristics of harmful gas concentration are classified according to the height of the helmet, and each height corresponds to a complete set of change trend characteristics;

[0037] The sorted trend characteristics of harmful gas concentration changes within different height ranges are combined into a continuous trend sequence.

[0038] Furthermore, the method for constructing a harmful gas concentration prediction model includes:

[0039] Collect historical data on changes in harmful gas concentrations and their corresponding time characteristics, check data quality, and handle missing values, outliers, and duplicate records;

[0040] Analyze the correlation between the changes in harmful gas concentrations and the corresponding time features, and select relevant features for model training;

[0041] Select the time series analysis model to build a harmful gas concentration prediction model;

[0042] The data set is divided into a training set, a validation set and a test set, and the training set, the validation set and the test set are used to train, validate and test the harmful gas concentration prediction model;

[0043] The hyperparameters of the harmful gas concentration prediction model are adjusted through the grid search method to optimize the performance of the harmful gas concentration prediction model;

[0044] The cross-validation technique was used to evaluate the generalization ability of the harmful gas concentration prediction model;

[0045] The prediction performance of the harmful gas concentration prediction model was evaluated using mean square error, root mean square error, and mean absolute error.

[0046] Deploy the validated hazardous gas concentration prediction model into the actual environment for real-time prediction;

[0047] Monitor the prediction performance of the hazardous gas concentration prediction model and update it as needed.

[0048] On the other hand, the present application also provides an environmental monitoring and early warning system for a safety helmet, the system comprising:

[0049] The data acquisition module is used to continuously obtain the environmental data of the wearer and the height information of the helmet monitored by the helmet integrated environmental sensor according to the preset sensor monitoring frequency. These data correspond to the height of the helmet.

[0050] A cluster analysis module is used to perform cluster analysis on the environmental data around the wearer based on the height of the helmet to form multiple environmental data sets, each set containing all environmental data within the same height range;

[0051] The trend analysis module is used to perform trend analysis on each environmental data set, extract the trend characteristics of temperature and humidity changes and the trend characteristics of harmful gas concentration changes, and form an environmental change feature vector;

[0052] The model application module is used to input the environmental change feature vector into the pre-built gas concentration environmental impact correction model to obtain the actual change trend characteristics of the harmful gas concentration corresponding to the height range of the helmet;

[0053] A sequence construction module is used to sort the actual change trend characteristics of the concentration of harmful gases corresponding to different height ranges according to the height, and construct the actual change trend sequence of the concentration of harmful gases in the entire confined space;

[0054] A prediction analysis module is used to input the constructed actual change trend sequence of harmful gas concentration into a pre-constructed harmful gas concentration prediction model to predict the time node when the harmful gas concentration is higher than the preset concentration;

[0055] The early warning notification module is used to promptly send reminder information to the wearer when the concentration of harmful gases is predicted to be higher than the preset concentration.

[0056] In a third aspect, the present application provides an electronic device, comprising a bus, a transceiver, a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the transceiver, the memory, and the processor are connected via the bus, and the computer program, when executed by the processor, implements the steps of any one of the above methods.

[0057] In a fourth aspect, the present application further provides a computer-readable storage medium having a computer program stored thereon, wherein the computer program implements the steps in any one of the above-mentioned methods when executed by a processor.

[0058] Compared with the prior art, the beneficial effects of the present invention are as follows: the environmental data around the wearer is continuously monitored by the sensor integrated in the helmet, the changes of environmental parameters with time and position are captured, and the real-time and continuity of environmental data acquisition are improved; the actual change trend characteristics of the concentration of harmful gases corresponding to the height range of the helmet are arranged to obtain the actual change trend sequence of the concentration of harmful gases in the entire confined space, thereby improving the monitoring and early warning of the global environmental conditions of the confined space and reducing the deficiency of fixed monitoring stations covering all areas; the actual change trend of the concentration of harmful gases is predicted by trend analysis of the environmental data set and combined with the gas concentration environmental impact correction model, so that an early warning is issued before the concentration of harmful gases actually reaches a dangerous level, thereby improving the timeliness and accuracy of the early warning; by The actual change trend characteristics of the harmful gas concentration corresponding to the range are arranged to obtain the actual change trend sequence of the harmful gas concentration in the entire confined space, so as to realize the monitoring and early warning of the global environmental conditions of the confined space, and make up for the deficiency that fixed monitoring stations are difficult to cover all areas; through the integration of environmental sensors, personalized monitoring of the environment around the operators is realized, and the actual change trend sequence of the harmful gas concentration and the harmful gas concentration prediction model are combined to realize intelligent environmental monitoring, and automatically judge and warn of potential dangers based on real-time data; through continuous dynamic monitoring of environmental parameters in confined spaces, environmental change trends are captured and analyzed in time, and environmental conditions at different heights are evaluated to provide a scientific basis for predicting changes in harmful gas concentrations, realize personalized early warning and in-depth analysis of environmental parameters, and improve safety in different spatial working environments. BRIEF DESCRIPTION OF THE DRAWINGS

[0059] Figure 1 is a flow chart of the present invention;

[0060] Figure 2 is a flow chart of a method for constructing a harmful gas concentration prediction model;

[0061] Figure 3 It is a structural diagram of the environmental monitoring and early warning system for safety helmets. DETAILED DESCRIPTION

[0062] In the description of this application, those skilled in the art should know that this application can be implemented as a method, an apparatus, an electronic device, and a computer-readable storage medium. Therefore, this application can be specifically implemented in the following forms: complete hardware, complete software (including firmware, resident software, microcode, etc.), a combination of hardware and software. In addition, in some embodiments, this application can also be implemented in the form of a computer program product in one or more computer-readable storage media, and the computer-readable storage medium contains computer program code.

[0063] The above-mentioned computer-readable storage medium may adopt any combination of one or more computer-readable storage media. Computer-readable storage media include: electrical, magnetic, optical, electromagnetic, infrared or semiconductor systems, devices or devices, or any combination of the above. More specific examples of computer-readable storage media include: portable computer disks, hard disks, random access memories, read-only memories, erasable programmable read-only memories, flash memories, optical fibers, optical disc read-only memories, optical storage devices, magnetic storage devices, or any combination of the above. In the present application, computer-readable storage media can be any tangible medium containing or storing a program that can be used by or in conjunction with an instruction execution system, device, or device.

[0064] The acquisition, storage, use, and processing of data in the technical solution of this application comply with the relevant provisions of national laws.

[0065] The present application describes the provided methods, devices, and electronic devices through flowcharts and / or block diagrams.

[0066] It should be understood that each box in the flowchart and / or block diagram and the combination of boxes in the flowchart and / or block diagram can be implemented by computer-readable program instructions. These computer-readable program instructions can be provided to a processor of a general-purpose computer, a special-purpose computer or other programmable data processing device to produce a machine, and these computer-readable program instructions are executed by a computer or other programmable data processing device to produce a device that implements the functions / operations specified by the boxes in the flowchart and / or block diagram.

[0067] These computer-readable program instructions may also be stored in a computer-readable storage medium that enables a computer or other programmable data processing device to work in a specific manner. In this way, the instructions stored in the computer-readable storage medium produce an instruction device product including functions / operations specified in the blocks in the flowchart and / or block diagram.

[0068] Computer-readable program instructions may also be loaded onto a computer, other programmable data processing apparatus, or other device, so that a series of operational steps are performed on the computer, other programmable data processing apparatus, or other device to produce a computer-implemented process, thereby enabling the instructions executed on the computer or other programmable data processing apparatus to provide a process for implementing the functions / operations specified in the blocks in the flowchart and / or block diagram.

[0069] The present application is described below in conjunction with the drawings in the present application.

[0070] Embodiment 1: Figure 1 to Figure 2 As shown, the environmental monitoring and early warning method for a safety helmet of the present invention specifically comprises the following steps:

[0071] S1. Continuously obtain the environmental data of the wearer and the height of the helmet monitored by the helmet integrated environmental sensor according to the preset sensor monitoring frequency. The environmental data of the wearer includes temperature, humidity and concentration of harmful gases. The height of the helmet corresponds to the environmental data of the wearer. The harmful gases include carbon monoxide, sulfur dioxide and methane.

[0072] Step S1 is how to effectively and continuously obtain the environmental data around the wearer and the height information of the helmet; the details are as follows:

[0073] First, a variety of environmental sensors are integrated on the helmet, including temperature and humidity sensors, harmful gases, concentration sensors, and height sensors. When selecting sensors, for example, the temperature and humidity sensors use DHT11 or DHT22, which can provide reliable temperature and humidity collection readings. The gas sensors use MQ series sensors or TGS series sensors, which can detect the concentration of harmful gases such as carbon monoxide, sulfur dioxide, and methane, and provide fast response and accurate measurement.

[0074] To set the sensor monitoring frequency, follow the steps below:

[0075] The monitoring frequency is initially set according to the characteristics of the operating environment and existing industry standards, and different monitoring frequencies are set according to different environmental data;

[0076] Conduct tests in actual operating environments to observe data quality, system stability, and battery life at different monitoring frequencies to determine the optimal monitoring frequency;

[0077] Adjust according to field test results and operational needs, and gradually optimize monitoring frequency to achieve the best balance;

[0078] Collect feedback from operators to understand their needs and feelings about monitoring frequency and further adjust the monitoring frequency;

[0079] As technology advances and the operating environment changes, regularly review and adjust the monitoring frequency to ensure it always complies with the latest safety requirements and technical standards;

[0080] After the sensor monitoring frequency is set, the sensor collects environmental data around the wearer in real time when monitoring at the set frequency, including temperature and humidity, concentration of harmful gases, and the height of the helmet. The built-in memory in each sensor is used to store the monitoring data to prevent data loss when wireless communication is interrupted. Low-power wireless communication technology is used to transmit data to the central processing unit or cloud server, and encrypted communication protocols are used to ensure the stability and security of transmitted data.

[0081] Software control ensures that each time environmental data is collected, the height information of the helmet is recorded synchronously; the timestamp is used as the index of the data record to ensure the accurate correspondence between the environmental data and the height information; a suitable data record format is designed to ensure that each data record contains environmental data, such as temperature and humidity, harmful gas concentration, the height of the helmet and key information of the timestamp, which is convenient for subsequent data processing and analysis; in a confined space, harmful gases are heavier or lighter than air, so the distribution of harmful gases in space will change with different heights; recording the height information of the helmet facilitates more accurate analysis of the changing trend of harmful gas concentrations; in the data collection stage, ensure the integrity and consistency of the data to lay the foundation for subsequent data processing and analysis; the data records are easy to read and process, which is convenient for subsequent analysis by algorithms and models; each wearer's helmet will independently collect and process data, and conduct personalized monitoring and early warning for each wearer's specific location and surrounding environment, which is convenient for precise safety management.

[0082] By selecting reliable sensors, the accuracy and stability of environmental data are ensured. At the same time, the height sensor is integrated to obtain the precise height information of the helmet, providing a multi-dimensional data source for subsequent data analysis. By comprehensively considering the characteristics of the working environment, industry standards, expert advice, on-site tests and feedback from operators, the sensor monitoring frequency is scientifically set, which not only ensures the real-time nature of the data, but also avoids unnecessary energy consumption, achieving the best balance between monitoring efficiency and energy consumption. The sensor continuously collects environmental data at the set frequency and transmits it to the central processing unit or cloud server in real time through low-power wireless communication technology to ensure the continuity and real-time nature of the data. At the same time, the built-in memory is used as a data backup to prevent data loss when wireless communication is interrupted. The encrypted communication protocol is used to reduce data loss and improve the security and stability of data transmission. Record the height information and timestamp of the helmet to ensure the accurate correspondence between environmental data and height information, and improve the reliability of data analysis; design a suitable data recording format to ensure that each data record contains key information to facilitate subsequent data processing and analysis; each wearer's helmet independently collects and processes data to achieve personalized monitoring and early warning for each wearer's specific location and surrounding environment, and improve the accuracy of safety management; in a confined space, the distribution of harmful gases changes with height, and recording height information facilitates accurate analysis of the changing trend of harmful gas concentrations and improves the accuracy and timeliness of early warnings; by using scientific and reasonable sensor selection and integration, monitoring frequency setting, continuous real-time data collection and transmission, safe and stable data recording, monitoring and early warning, etc. in the data collection stage, the safety monitoring level of the confined space working environment can be improved.

[0083] S2. Based on the height of the helmet, cluster analysis is performed on the environmental data surrounding the wearer to obtain several environmental data sets, wherein each environmental data set includes environmental data surrounding all wearers within the height range of the same helmet;

[0084] In step S2, cluster analysis is performed on the environmental data around the wearer by using the height information of the helmet to identify environmental features within different height ranges; specifically, as follows:

[0085] In step S1, the environmental data around the wearer has been continuously acquired, the environmental data including temperature, humidity, concentration of harmful gases, and height information of the helmet. Based on the height information of the helmet, the data is divided into different height ranges;

[0086] Preprocess the acquired environmental data to remove outliers or missing values. After the outliers or missing values ​​are removed, standardize or normalize the temperature, humidity and harmful gas concentrations to facilitate cluster analysis.

[0087] According to the processed environmental data, select a suitable clustering algorithm, such as K-means and DBSCAN. The clustering algorithm is selected according to the characteristics and requirements of the environmental data. Taking the K-means algorithm as an example, first determine the number of clusters, then select the initial cluster center, and then the algorithm will iteratively assign each data point to the nearest cluster center and update the location of the cluster center until the stop condition is met. DBSCAN is suitable for non-spherical clusters and data sets with different densities, and does not need to know the number of clusters in advance.

[0088] The height information of the helmet is used as the main basis for clustering. The environmental conditions at different heights in a confined space are different, such as uneven distribution of temperature gradient or harmful gas concentration, or the height information is convenient for distinguishing different environmental conditions. The collected environmental data are grouped according to the height of the helmet and divided into multiple height intervals. The division of the height interval is set according to the actual situation, and the environmental data in each height interval will be regarded as a clustering object.

[0089] Perform cluster analysis on the data in each height interval using a selected clustering algorithm, and generate an environmental data set for each height interval, including the surrounding environmental data of all wearers in the height interval;

[0090] After clustering is completed, the clustering results are verified, and the quality of clustering is evaluated using indicators such as silhouette coefficient or Calinski-Harabasz index. If the clustering results are not ideal, such as too large differences within clusters, too small differences between clusters, or difficult to explain clustering results, the clustering algorithm should be adjusted in time, such as changing the number of clusters or optimizing the initialization strategy, and clustering analysis should be repeated.

[0091] By taking the height information of the helmet as the main basis for clustering, the environmental characteristics of different height levels can be effectively distinguished, and the environmental characteristics within different height ranges can be accurately identified; in the data preprocessing stage, outliers or missing values ​​are removed, the data scale is unified, and the accuracy and completeness of cluster analysis are improved; according to the characteristics and needs of environmental data, appropriate clustering algorithms are selected, and the division of height intervals is set according to actual conditions, which has high flexibility and scalability, and the clustering effect is continuously optimized by verifying the clustering results and adjusting the clustering algorithm; the height intervals are divided according to actual conditions, making the cluster analysis more suitable for actual application scenarios; by dividing the data according to the installation The heights of the entire hat are grouped to distinguish the environmental characteristics at different heights, so as to accurately understand the trend of environmental changes; the quality of clustering is evaluated using indicators such as the silhouette coefficient and the Calinski-Harabasz index to ensure the effectiveness and rationality of the clustering results and avoid wrong decisions caused by invalid clustering results; when the clustering results are not ideal, iterative optimization is performed by adjusting algorithm parameters, feature selection and other means to ensure that the final clustering results meet expectations; by performing height-based clustering analysis on the wearer's surrounding environmental data, the environmental characteristics within different height ranges can be accurately identified, thereby improving the overall performance and safety of the environmental monitoring system.

[0092] S3, performing trend analysis on each environmental data set to obtain an environmental change feature vector, wherein the environmental change feature vector includes a temperature and humidity change trend feature and a harmful gas concentration change trend feature;

[0093] Step S3 performs trend analysis on each environmental data set to obtain an environmental change feature vector to identify the change trend of temperature, humidity and harmful gas concentration in the environment; the details are as follows:

[0094] First, the data in each environmental data set is cleaned before trend analysis, and then standardized or normalized after cleaning;

[0095] Extract the time series data of temperature, humidity and harmful gas concentration from each environmental data set; use the time series data of temperature, humidity and harmful gas concentration as the basis for trend analysis to analyze the changes of environmental parameters over time;

[0096] Choose a suitable trend identification algorithm to analyze time series data. Common trend analysis methods include linear regression, moving average, autocorrelation function, seasonal decomposition and exponential smoothing. Linear regression trend analysis method is used to fit time series data to identify the changing trend of temperature, humidity and harmful gas concentration over time. Moving average trend analysis method smoothes data fluctuations by calculating the average value over a period of time to help identify potential trends. Autocorrelation function trend analysis method is used to detect autocorrelation in data to help identify whether there are periodic changes. Seasonal decomposition trend analysis method uses seasonal decomposition technology to separate trend, seasonal and random components for data with obvious seasonal patterns. Exponential smoothing trend analysis method predicts future trends by weighting historical data, which is suitable for smoothing data and identifying long-term trends.

[0097] According to the selected trend recognition algorithm, calculate the change trend characteristics of temperature, humidity and harmful gas concentration in each environmental data set; the change trend characteristics include the slope of the trend, the direction of the trend and the duration of the trend;

[0098] The temperature and humidity change trend characteristics are obtained by the linear regression trend analysis method, the slope and moving average slope; the harmful gas concentration change trend characteristics are obtained by the linear regression trend analysis method, the slope or the percentage of concentration change over time;

[0099] Cross-validation and retaining a portion of the data as a test set are used to verify the results of trend analysis to ensure that the extracted feature vectors can accurately reflect the trend of environmental changes.

[0100] By conducting a detailed trend analysis on the time series data of temperature, humidity and harmful gas concentration, the changing trend of environmental parameters can be accurately identified; it is convenient to issue an early warning before the harmful gas concentration is about to exceed the safety threshold, thereby improving the accuracy and timeliness of the warning; data cleaning and standardization or normalization processing ensures the quality of data and improves the accuracy of trend analysis; multiple trend identification algorithms are selected, and the appropriate method is flexibly selected according to the characteristics and needs of the data; it can be selected and optimized according to the characteristics and data characteristics of different confined space environments to improve the adaptability in different scenarios; by extracting the trend characteristics of temperature and humidity changes and the trend characteristics of harmful gas concentration changes, the trend of environmental changes can be more clearly identified; the results of trend analysis are evaluated using a cross-validation method to ensure that the extracted feature vector can accurately reflect the trend of environmental changes and avoid wrong decisions caused by invalid feature vectors; data collection is carried out based on the integrated environmental sensor on the helmet to achieve real-time monitoring and continuous monitoring, so as to facilitate the timely discovery of potential dangerous situations and take corresponding safety measures; by more accurately identifying environmental characteristics, the accuracy of the early warning system and the safety level of operators can be improved; by performing trend analysis on each environmental data set and constructing environmental change feature vectors, the environmental change trend within different height ranges can be accurately identified, and the overall performance and safety of the environmental monitoring system can be improved.

[0101] S4, inputting the environmental change feature vector into a pre-built gas concentration environmental impact correction model to obtain the actual change trend characteristics of the harmful gas concentration corresponding to the height range of the helmet;

[0102] Step S4 analyzes the environmental change feature vector and uses the pre-built gas concentration environmental impact correction model to obtain the actual change trend characteristics of the harmful gas concentration within the height range of the helmet; the details are as follows:

[0103] Collect historical environmental data around the wearer monitored by sensors, including changes in parameters such as temperature, humidity, and harmful gas concentration over time; remove outliers and missing values ​​from historical environmental data to ensure data quality; extract features related to changes in harmful gas concentrations, such as temperature and humidity change trend features and harmful gas concentration change trend features;

[0104] The temperature and humidity change trend characteristics, harmful gas concentration change trend characteristics and height information are selected as the features of the model; the temperature and humidity change trend characteristics include the change rate and change direction of temperature and humidity over time; the harmful gas concentration change trend characteristics include the change rate and change direction of harmful gas concentration over time; the height information includes the height information of the helmet as an additional feature, taking into account the environmental differences at different heights;

[0105] The gas concentration environmental impact correction model is constructed based on the principles of machine learning or statistics, and the construction methods include:

[0106] Multiple regression analysis: Use multiple linear regression or nonlinear regression models, take the trend of harmful gas concentration as the dependent variable, and the trend of temperature and humidity, height, etc. as independent variables, and find out the quantitative relationship between them through regression analysis;

[0107] Neural network: Build a deep learning neural network model and use a large amount of historical data for training, so that the model can automatically learn and identify the changing patterns of harmful gas concentrations under different environmental conditions;

[0108] Time series analysis: Use time series analysis method to consider the impact of time factors on the trend of harmful gas concentration changes and conduct a comprehensive analysis in combination with other environmental parameters;

[0109] After the gas concentration environmental impact correction model is constructed, the obtained environmental change feature vector is divided into a training set and a test set. The gas concentration environmental impact correction model is trained using the training set data. After the training is completed, the gas concentration environmental impact correction model is tested using the test set and the gas concentration environmental impact correction model parameters are adjusted through grid search, random search and other methods to obtain the best performance.

[0110] The model was verified by cross-validation and holdout methods to ensure the stability and reliability of the gas concentration environmental impact correction model under different environmental conditions. 2 Indicators such as scores are used to evaluate the predictive performance of the model;

[0111] The environmental change feature vector obtained in step S3 is input into a pre-built gas concentration environmental impact correction model, and the gas concentration environmental impact correction model outputs the actual change trend characteristics of the harmful gas concentration within the height range where the helmet is located.

[0112] By comprehensively considering the influence of various environmental factors such as temperature, humidity, and altitude, the gas concentration environmental impact correction model accurately reflects the actual change trend of harmful gas concentrations; multi-dimensional analysis is more precise and reliable than simple threshold judgment; the gas concentration environmental impact correction model takes into account environmental differences at different altitudes, making it better adapted to the complex and changeable environmental conditions in confined spaces; whether it is a mine, underground tunnel or other confined environment, the gas concentration environmental impact correction model can provide accurate prediction results; by real-time analysis of environmental change feature vectors and inputting them into the correction model, potential dangers of harmful gas concentrations can be discovered in a timely manner, and the wearer can be prompted at the first time to reduce The possibility of accidents; constructing and training a gas concentration environmental impact correction model based on historical environmental data, so that the prediction results have stronger data support and scientific basis, which is convenient for decision makers to formulate safety strategies and emergency measures more rationally; by adopting a variety of model construction methods and performing parameter optimization and verification, the gas concentration environmental impact correction model maintains high stability and reliability under different environmental conditions; by constructing a gas concentration environmental impact correction model, the actual change trend characteristics of the harmful gas concentration within the height range of the helmet are predicted based on the environmental change characteristic vector, ensuring that the gas concentration environmental impact correction model accurately predicts the changing trend of the harmful gas concentration, thereby improving the accuracy and effectiveness of the prediction model.

[0113] S5. Arrange the actual change trend characteristics of the concentration of harmful gases corresponding to the height range of each helmet according to the height, and obtain the actual change trend sequence of the concentration of harmful gases corresponding to the confined space;

[0114] Step S5 sorts and integrates the actual change trend characteristics of the concentration of harmful gases corresponding to the height ranges of each helmet, to form an actual change trend sequence of the concentration of harmful gases in the confined space; specifically, as follows:

[0115] Through the gas concentration environmental impact correction model of step S4, the actual change trend characteristics of the harmful gas concentration corresponding to the height range of each helmet are obtained; the actual change trend characteristics of the harmful gas concentration include the actual change trend information of the harmful gas concentration in each height range;

[0116] The actual change trend characteristics of harmful gas concentration are classified according to the height of the helmet, and each height corresponds to a complete set of change trend characteristics;

[0117] The trend characteristics of harmful gas concentration changes within the sorted height ranges are combined into a continuous trend sequence; the trend sequence reflects the overall trend of harmful gas concentration changes with height in the entire confined space;

[0118] Analyze trend sequences, identify trends and abnormal change points of increase or decrease in harmful gas concentrations, and use them to predict future changes in harmful gas concentrations;

[0119] The generated results are generated in the form of a chart or a numerical list according to the actual situation. The chart shows the trend of the concentration of harmful gases in the entire confined space with the height change, and the numerical list shows the concentration change trend within each height range;

[0120] The trend sequence can be provided as input to the subsequent harmful gas concentration prediction model for further prediction analysis.

[0121] By sorting and integrating the actual change trend characteristics of harmful gas concentrations within different height ranges, the environmental conditions in the entire confined space are fully covered; each height area is independently monitored and taken into consideration to ensure the comprehensiveness and real-time nature of the data; considering the impact of height on gas concentration distribution in a confined space, the harmful gas concentration is combined with height information, and sorted by height to more accurately reflect the change trend of gas concentration at different height levels, making it easier to understand the diffusion and distribution of gases in confined spaces; by analyzing the trend sequence, the increase or decrease trend of harmful gas concentrations and abnormal change points can be identified; it is easy to predict future changes in harmful gas concentrations, so that measures can be taken in advance to prevent potential Safety risks; the results are displayed in the form of charts or numerical lists, making the monitoring results more intuitive and easy to understand; the chart form can clearly show the trend of harmful gas concentration changes with height in the entire confined space, while the numerical list provides detailed concentration change trend data for further analysis and processing; the trend sequence is provided as input to the subsequent harmful gas concentration prediction model, providing a basis for subsequent prediction analysis; by utilizing trend data, a more accurate and reliable prediction model is constructed to improve the accuracy and timeliness of early warning; by providing important information on changes in harmful gas concentrations, potential hazards in confined spaces can be better understood and responded to, ensuring the safety of operators and improving environmental monitoring efficiency.

[0122] S6. Input the actual change trend sequence of the harmful gas concentration into the pre-built harmful gas concentration prediction model to obtain the time node when the harmful gas concentration is higher than the preset concentration; and prompt the wearer with the time node;

[0123] Step S6 is to convert the actual trend sequence of harmful gas concentration changes into specific warning information so as to promptly notify the wearer of potential dangers; the details are as follows:

[0124] First, a harmful gas concentration prediction model is constructed. The construction method is as follows:

[0125] S61. Collect historical data on changes in harmful gas concentrations and their corresponding time characteristics; check data quality, handle missing values, outliers, and duplicate records; extract useful features such as timestamps, seasonality, and periodicity;

[0126] S62. Analyze the correlation between the change in harmful gas concentration and time characteristics, and select relevant features for model training; when there are a large number of features, use principal component analysis or other dimensionality reduction techniques to reduce the number of features and reduce model complexity;

[0127] S63. Select appropriate models according to the characteristics of the problem, including time series analysis models, machine learning models, and deep learning models;

[0128] S64, dividing the data set into a training set, a validation set, and a test set; using the training set to train the harmful gas concentration prediction model;

[0129] S65. Adjust the hyperparameters of the harmful gas concentration prediction model by grid search, random search and other methods to optimize the performance of the harmful gas concentration prediction model;

[0130] S66. Use cross-validation techniques to evaluate the generalization ability of the harmful gas concentration prediction model;

[0131] S67. Use appropriate indicators, such as mean square error, root mean square error, and mean absolute error, to evaluate the prediction performance of the harmful gas concentration prediction model;

[0132] S68. Deploy the verified harmful gas concentration prediction model into the actual environment for real-time prediction; monitor the prediction performance of the harmful gas concentration prediction model and update it as needed;

[0133] After the harmful gas concentration prediction model is constructed, the actual change trend sequence of the harmful gas concentration obtained in step S5 is input into the harmful gas concentration prediction model for prediction; the harmful gas concentration prediction model outputs the predicted value of the harmful gas concentration in a certain time period in the future;

[0134] Based on industry standards or safety guidelines, determine the threshold of harmful gas concentration, exceeding which there will be safety risks; continuously monitor the predicted harmful gas concentration value, and trigger the alarm mechanism when the predicted value exceeds the threshold;

[0135] The harmful gas concentration prediction model predicts that the harmful gas concentration will exceed the threshold at a certain future time node, and sends an alarm message to the wearer through the communication module integrated in the helmet; at the same time, the alarm message is sent to the remote monitoring center or the mobile terminal of the responsible personnel to facilitate the adoption of countermeasures.

[0136] Through feature selection and dimensionality reduction technology, the features that affect the change of harmful gas concentration are selected, unnecessary features are reduced, the complexity of the harmful gas concentration prediction model is reduced, and the model efficiency is improved; a variety of model options are provided, and the appropriate model is selected according to the characteristics of the actual problem to achieve the best prediction effect; through cross-validation technology and a variety of evaluation indicators, the generalization ability and prediction performance of the harmful gas concentration prediction model are comprehensively evaluated to ensure the effectiveness of the harmful gas concentration prediction model; by real-time monitoring of the predicted harmful gas concentration value and comparing it with the preset threshold, potential dangerous situations are discovered in time, and alarm information is sent to the wearer to enhance the response speed of the early warning system; the harmful gas concentration prediction model The model is continuously monitored and regularly updated to ensure that the prediction model can adapt to the ever-changing working environment and maintain the accuracy of the prediction; in addition to sending alarms directly to the wearer, alarms are also sent to the remote monitoring center or responsible personnel to ensure a multi-level response mechanism and improve the efficiency of emergency handling; as time changes, the preset thresholds and harmful gas concentration prediction model parameters are adjusted according to actual conditions to ensure the long-term effectiveness and applicability of the method; the sensors integrated in the safety helmet can achieve accurate prediction and timely warning of the harmful gas concentration in the confined space, improve the safety in the confined space working environment, reduce the risks brought by harmful gas leakage, and protect the lives of operators.

[0137] Embodiment 2: Figure 3 As shown, the environmental monitoring and early warning system for a safety helmet of the present invention specifically includes the following modules:

[0138] The data acquisition module is used to continuously obtain the environmental data of the wearer and the height information of the helmet monitored by the environmental sensor integrated in the helmet according to the preset sensor monitoring frequency; these data correspond to the height of the helmet;

[0139] A cluster analysis module is used to perform cluster analysis on the environmental data around the wearer based on the height of the helmet to form multiple environmental data sets, each set containing all environmental data within the same height range;

[0140] The trend analysis module is used to perform trend analysis on each environmental data set, extract the trend characteristics of temperature and humidity changes and the trend characteristics of harmful gas concentration changes, and form an environmental change feature vector;

[0141] The model application module is used to input the environmental change feature vector into the pre-built gas concentration environmental impact correction model to obtain the actual change trend characteristics of the harmful gas concentration corresponding to the height range of the helmet;

[0142] A sequence construction module is used to sort the actual change trend characteristics of the concentration of harmful gases corresponding to different height ranges according to the height, and construct the actual change trend sequence of the concentration of harmful gases in the entire confined space;

[0143] A prediction analysis module is used to input the constructed actual change trend sequence of harmful gas concentration into a pre-constructed harmful gas concentration prediction model to predict the time node when the harmful gas concentration is higher than the preset concentration;

[0144] The early warning notification module is used to promptly send reminder information to the wearer when the concentration of harmful gases is predicted to be higher than the preset concentration.

[0145] By continuously acquiring the environmental data around the wearer and the height information of the helmet monitored by the integrated environmental sensor of the helmet according to the preset sensor monitoring frequency, the real-time and continuity of the monitoring data are ensured; the cluster analysis module performs cluster analysis on the environmental data based on the height of the helmet to form multiple environmental data sets, which is convenient for accurately reflecting the environmental conditions at a specific height; the trend analysis module can identify whether the current state exceeds the standard, and predict future trends by extracting the trend characteristics of temperature and humidity changes and the trend characteristics of harmful gas concentration changes, and take preventive measures in advance; the model application module accurately evaluates the actual concentration change trend of harmful gases by using a special gas concentration environmental impact correction model to reduce false alarms or missed alarms. The prediction analysis module uses a pre-built harmful gas concentration prediction model to predict when the harmful gas concentration will reach an unsafe level, which helps to warn in advance; by considering the interaction of multiple environmental factors, the accuracy of the prediction is improved. The warning notification module performs independent analysis based on the data monitored by each helmet, provides personalized safety alerts for each wearer, and improves the response speed and pertinence; the environmental monitoring and early warning system for helmets overcomes the limitations of traditional fixed monitoring stations and improves the efficiency and reliability of the monitoring system by introducing analytical methods.

[0146] The various variations and specific embodiments of the environmental monitoring and early warning method for safety helmets in the aforementioned embodiment 1 are also applicable to the environmental monitoring and early warning system for safety helmets in this embodiment. Through the aforementioned detailed description of the environmental monitoring and early warning method for safety helmets, those skilled in the art can clearly know the implementation method of the environmental monitoring and early warning system for safety helmets in this embodiment, so for the sake of brevity of the specification, it will not be described in detail here.

[0147] In addition, the present application also provides an electronic device, including a bus, a transceiver, a memory, a processor, and a computer program stored in the memory and executable on the processor. The transceiver, the memory, and the processor are respectively connected via a bus. When the computer program is executed by the processor, each process of the above-mentioned method for controlling output data is implemented, and the same technical effect can be achieved. To avoid repetition, it will not be described here.

[0148] The above is only a preferred embodiment of the present invention. It should be pointed out that for ordinary technicians in this technical field, several improvements and modifications can be made without departing from the technical principles of the present invention. These improvements and modifications should also be regarded as the scope of protection of the present invention.

Claims

1. A method for environmental monitoring and early warning of a safety helmet, characterized in that: The method is used to perform environmental safety monitoring and early warning using a safety helmet equipped with an integrated environmental sensor when working in a confined space, and the method comprises: According to the preset sensor monitoring frequency, continuously obtain the environmental data around the wearer and the height of the helmet monitored by the helmet integrated environmental sensor, the environmental data around the wearer includes temperature, humidity and harmful gas concentration, and the height of the helmet corresponds to the environmental data around the wearer one by one; Based on the height of the helmet, cluster analysis is performed on the environmental data around the wearer to obtain several environmental data sets, where each environmental data set includes environmental data around all wearers within the height range of the same helmet; Performing trend analysis on each environmental data set to obtain an environmental change feature vector, wherein the environmental change feature vector includes a temperature and humidity change trend feature and a harmful gas concentration change trend feature; Inputting the environmental change feature vector into a pre-built gas concentration environmental impact correction model to obtain the actual change trend characteristics of the harmful gas concentration corresponding to the height range of the helmet; The actual change trend characteristics of the concentration of harmful gases corresponding to the height range of each helmet are arranged according to the height, so as to obtain the actual change trend sequence of the concentration of harmful gases corresponding to the confined space; Input the actual trend sequence of harmful gas concentration changes into the pre-built harmful gas concentration prediction model to obtain the time node when the harmful gas concentration is higher than the preset concentration; and prompt the wearer with the time node; The method for constructing the gas concentration environmental impact correction model comprises: Collect the historical temperature and humidity change trend characteristics and harmful gas concentration change trend characteristics around the wearer monitored by the sensor, as well as the actual change trend characteristics of the historical harmful gas concentration; A gas concentration environmental impact correction model was constructed based on a multiple linear regression model. The actual trend characteristics of harmful gas concentration were used as dependent variables, and the trend characteristics of temperature and humidity and harmful gas concentration were used as independent variables. The quantitative relationship between the dependent and independent variables was found through regression analysis. The collected environmental change feature vectors are divided into a training set and a test set. The gas concentration environmental impact correction model is trained using the training set data. After the training is completed, the gas concentration environmental impact correction model is tested using the test set. The gas concentration environmental impact correction model was verified by cross-validation method.

2. The environmental monitoring and early warning method for a safety helmet as claimed in claim 1, characterized in that: The harmful gases include carbon monoxide, sulfur dioxide and methane.

3. The environmental monitoring and early warning method for a safety helmet as claimed in claim 1, characterized in that: The method for obtaining the environmental data set includes: Acquire environmental data around the wearer through the helmet's integrated environmental sensor; Preprocess the environmental data to remove outliers and missing values. After the removal, standardize the temperature, humidity and harmful gas concentrations. Select K-means algorithm to cluster the preprocessed environmental data; The collected environmental data is divided into multiple height intervals according to the actual situation according to the height of the helmet, and the environmental data in each height interval is a cluster object; Use the K-means algorithm to perform cluster analysis on the data in each height interval, and generate an environmental data set for each height interval, including the surrounding environmental data of all wearers in the height interval; After clustering is completed, the silhouette coefficient is used to verify the clustering results and evaluate the quality of clustering.

4. The environmental monitoring and early warning method for a safety helmet as claimed in claim 1, characterized in that: The method for obtaining the environmental change feature vector comprises: Clean the data in each environmental data set and standardize the cleaned data; Extract time series data of temperature, humidity and harmful gas concentration from each environmental data set; Select the linear regression trend analysis method to fit the time series data and identify the changing trends of temperature, humidity and harmful gas concentration over time According to the linear regression trend recognition algorithm, the changing trend characteristics of temperature, humidity and harmful gas concentration in each environmental data set are calculated; The slope of the temperature and humidity change trend characteristics was obtained by linear regression trend analysis method; The trend characteristics of harmful gas concentration changes are analyzed by linear regression trend analysis method to obtain the slope of the trend of harmful gas concentration changes; The slope of the temperature and humidity change trend and the slope of the harmful gas concentration change trend are combined to obtain the environmental change feature vector.

5. The environmental monitoring and early warning method for a safety helmet as claimed in claim 1, characterized in that: The method for obtaining the actual change trend sequence of the harmful gas concentration includes: Obtain the actual change trend characteristics of the harmful gas concentration corresponding to the height range of each helmet; The actual change trend characteristics of harmful gas concentrations include the actual change trend information of harmful gas concentrations within each altitude range; The actual change trend characteristics of harmful gas concentration are classified according to the height of the helmet, and each height corresponds to a complete set of change trend characteristics; The sorted trend characteristics of harmful gas concentration changes within different height ranges are combined into a continuous trend sequence.

6. The environmental monitoring and early warning method for a safety helmet as claimed in claim 1, characterized in that: The method for constructing a harmful gas concentration prediction model comprises: Collect historical data on changes in harmful gas concentrations and their corresponding time characteristics, check data quality, and handle missing values, outliers, and duplicate records; Analyze the correlation between the changes in harmful gas concentrations and the corresponding time features, and select relevant features for model training; Select the time series analysis model to build a harmful gas concentration prediction model; The data set is divided into a training set, a validation set and a test set, and the training set, the validation set and the test set are used to train, validate and test the harmful gas concentration prediction model; The hyperparameters of the harmful gas concentration prediction model are adjusted through the grid search method to optimize the performance of the harmful gas concentration prediction model; The cross-validation technique was used to evaluate the generalization ability of the harmful gas concentration prediction model; The prediction performance of the harmful gas concentration prediction model was evaluated using mean square error, root mean square error, and mean absolute error. Deploy the validated hazardous gas concentration prediction model into the actual environment for real-time prediction; Monitor the prediction performance of the hazardous gas concentration prediction model and update it as needed.

7. An environmental monitoring and early warning system for a safety helmet, characterized in that: The system comprises: The data acquisition module is used to continuously obtain the environmental data of the wearer and the height information of the helmet monitored by the helmet integrated environmental sensor according to the preset sensor monitoring frequency. These data correspond to the height of the helmet. A cluster analysis module is used to perform cluster analysis on the environmental data around the wearer based on the height of the helmet to form multiple environmental data sets, each set containing all environmental data within the same height range; The trend analysis module is used to perform trend analysis on each environmental data set, extract the trend characteristics of temperature and humidity changes and the trend characteristics of harmful gas concentration changes, and form an environmental change feature vector; The model application module is used to input the environmental change feature vector into the pre-built gas concentration environmental impact correction model to obtain the actual change trend characteristics of the harmful gas concentration corresponding to the height range of the helmet; A sequence construction module is used to sort the actual change trend characteristics of the concentration of harmful gases corresponding to different height ranges according to the height, and construct the actual change trend sequence of the concentration of harmful gases in the entire confined space; A prediction analysis module is used to input the constructed actual change trend sequence of harmful gas concentration into a pre-constructed harmful gas concentration prediction model to predict the time node when the harmful gas concentration is higher than the preset concentration; The early warning notification module is used to send prompt information to the wearer in time when the concentration of harmful gases is predicted to be higher than the preset concentration; The method for constructing the gas concentration environmental impact correction model comprises: Collect the historical temperature and humidity change trend characteristics and harmful gas concentration change trend characteristics around the wearer monitored by the sensor, as well as the actual change trend characteristics of the historical harmful gas concentration; A gas concentration environmental impact correction model was constructed based on a multiple linear regression model. The actual trend characteristics of harmful gas concentration were used as dependent variables, and the trend characteristics of temperature and humidity and harmful gas concentration were used as independent variables. The quantitative relationship between the dependent and independent variables was found through regression analysis. The collected environmental change feature vectors are divided into a training set and a test set. The gas concentration environmental impact correction model is trained using the training set data. After the training is completed, the gas concentration environmental impact correction model is tested using the test set. The gas concentration environmental impact correction model was verified by cross-validation method.

8. An environmental monitoring and early warning electronic device for a helmet, comprising a bus, a transceiver, a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the transceiver, the memory, and the processor are connected via the bus, and characterized in that: When the computer program is executed by the processor, the steps in the method according to any one of claims 1 to 6 are implemented.

9. A computer-readable storage medium having a computer program stored thereon, characterized in that: When the computer program is executed by a processor, the steps in the method according to any one of claims 1 to 6 are implemented.

Images