High-altitude anti-falling comprehensive analysis method based on high-altitude operation monitoring multi-dimensional data and related device

Abstract

The invention provides a high-altitude anti-falling comprehensive analysis method based on high-altitude operation monitoring multi-dimensional data and a related device, and belongs to the field of high-altitude operation safety monitoring. The method comprises the following steps: acquiring position and posture data, physiological data, working environment data and group management data of each operator, and preprocessing to obtain a real-time data set; using the risk evaluation model to obtain a real-time risk score of high-altitude operation; using the risk prediction model to obtain a prediction risk probability of high-altitude operation; and carrying out real-time anti-falling alarm on high-altitude operation based on the real-time risk score and the predicted risk probability. According to the invention, by integrating position, physiology, environment, group management and other multi-dimensional data and through risk evaluation and prediction model comprehensive analysis, the comprehensiveness and accuracy of safety evaluation are significantly improved; and real-time anti-falling alarm is carried out based on a double-model result, so that the early warning accuracy and reliability are effectively improved, group-level security policy management is realized, and modern aerial work management requirements are met.

Description

Technical Field

[0001] This invention belongs to the field of high-altitude operation safety monitoring technology, specifically relating to a comprehensive analysis method and related device for high-altitude fall prevention based on multi-dimensional data from high-altitude operation monitoring. Background Technology

[0002] In today's industrial and high-altitude sector development, working at heights is a fundamental task with extremely wide applications, including building construction, power maintenance, and bridge construction. However, due to the special and complex nature of the working environment, working at heights faces numerous risks, and the safety of workers is constantly threatened.

[0003] To effectively reduce safety hazards in high-altitude operations and protect the lives of workers, protective equipment such as safety belts and safety nets are widely used, and high-altitude operation safety monitoring systems are also gradually developing. These monitoring systems aim to collect and analyze operational data in real time, promptly identify potential hazards and issue early warnings, providing strong support for high-altitude operation safety management.

[0004] However, existing methods for monitoring safety during high-altitude operations still have some shortcomings. Taking common location-based monitoring methods as an example, they mainly manage operations by acquiring the location information of workers, but cannot determine their safety status. While monitoring methods relying on smart wearable devices can effectively monitor the safety status of workers, this method mainly focuses on individual protection and lacks collaborative management capabilities for work teams. Summary of the Invention

[0005] In view of this, the present invention aims to provide a comprehensive analysis method and related device for high-altitude fall prevention based on multi-dimensional data of high-altitude operation monitoring, which aims to at least solve one of the many shortcomings of existing high-altitude operation safety monitoring methods.

[0006] To achieve the above objectives, the technical solution provided by the present invention is as follows:

[0007] In a first aspect, the present invention provides a comprehensive analysis method for fall prevention at heights based on multi-dimensional data from high-altitude operations monitoring, comprising the following steps:

[0008] The system acquires the position and posture data, physiological data, work environment data, and group management data of each worker in the high-altitude work group, and forms a real-time dataset; the group management data is used to evaluate the effectiveness of group management.

[0009] Preprocess the data in the real-time dataset to obtain a real-time dataset with a uniform data scale;

[0010] Based on the preprocessed real-time dataset, a pre-built risk assessment model is used to obtain a real-time risk score for high-altitude operations. The risk assessment model is a model that calculates the risk score by comprehensively considering pre-selected risk factors based on a given dataset. The risk factors include at least worker risk factors, environmental risk factors, and group management risk factors.

[0011] The preprocessed real-time dataset is input into a pre-built risk prediction model to obtain the predicted risk probability of high-altitude operations. The risk prediction model is based on a neural network and is used to output the risk probability based on a given dataset.

[0012] Real-time fall protection alarm for high-altitude operations based on real-time risk scoring and predicted risk probability.

[0013] Furthermore, acquire the position and posture data, physiological data, work environment data, and group management data of each worker in the high-altitude work group, including:

[0014] The system acquires the position, posture, and physiological data of workers by using smart safety belts and smart safety helmets worn by workers at height.

[0015] The operational environment data is obtained through micro-weather stations; the operational environment data includes at least the temperature, humidity and wind speed data at the work site.

[0016] Group management data is obtained through the management system; group management data includes communication stability data of smart seat belts and smart helmets, as well as alarm prompt timeliness data; communication stability data is used to assess whether communication is stable, and alarm prompt timeliness data is used to assess whether alarm prompts are timely.

[0017] Furthermore, the data in the real-time dataset is preprocessed to obtain a real-time dataset with a uniform data scale, including:

[0018] Perform data cleaning on the real-time dataset to remove outliers and fill in missing values.

[0019] After data cleaning, the various types of data are normalized to form a real-time dataset with a unified data scale.

[0020] Furthermore, the risk assessment model is a computational model constructed based on the weighting method;

[0021] Based on the preprocessed real-time dataset, a pre-built risk assessment model is used to obtain a real-time risk score for high-altitude operations, including:

[0022] Identify risk factors; among them, worker risk factors should include at least the wearing of safety belts and helmets, the protective condition of safety belts and helmets, and the worker's condition; environmental risk factors should include at least the working environment; and group management risk factors should include at least the effectiveness of group management.

[0023] Based on the preprocessed real-time dataset, the values ​​of each risk factor are calculated;

[0024] Determine the weight of each risk factor;

[0025] The values ​​and corresponding weights of each risk factor are substituted into the risk assessment model, and the real-time risk score for high-altitude operations is calculated by weighted summation.

[0026] Furthermore, based on the preprocessed real-time dataset, the values ​​of each risk factor are calculated, including:

[0027] For the wearing status of seat belts and helmets, as well as the protective status of seat belts and helmets, the assignment method is used to take values ​​in [0,1] to determine the corresponding risk factor values;

[0028] If both the wearing condition and the protective status are normal, the value is 0; if both the wearing condition and the protective status are abnormal, the value is 1; and for other cases, the value is taken from [0,1].

[0029] For the status of the workers, the movement status factor value and physiological status factor value of the workers are obtained by quantitative calculation using position and posture data and physiological data respectively.

[0030] For environmental risk factors, the values ​​of environmental risk factors are obtained by quantitative calculation using operational environment data.

[0031] To assess the effectiveness of group management, a value for the group management effectiveness is obtained by quantitatively calculating the group management data.

[0032] In each quantitative calculation process, if each data point is within its preset range, the corresponding factor value is assigned to 0; otherwise, the corresponding factor value is obtained by quantifying the degree to which the data deviates from the preset range.

[0033] Furthermore, the risk prediction model is built on a long short-term memory network and includes an input layer, an LSTM layer, and a fully connected layer. The number of neurons in the input layer is determined based on the number of features in the real-time dataset.

[0034] Furthermore, real-time fall protection alarms for high-altitude operations based on real-time risk scoring and predicted risk probabilities include:

[0035] If the real-time risk score exceeds the preset risk score threshold, a risk alarm will be issued in real time.

[0036] If the predicted risk probability exceeds the preset risk probability threshold, a risk warning will be issued in real time.

[0037] Secondly, the present invention provides a comprehensive analysis device for fall prevention based on multi-dimensional data from high-altitude operation monitoring, comprising:

[0038] The data acquisition module is used to acquire the position and posture data, physiological data, work environment data, and group management data of each worker in the high-altitude work group, and form a real-time dataset; the group management data is used to evaluate the effectiveness of group management.

[0039] The data preprocessing module is used to preprocess the data in the real-time dataset to obtain a real-time dataset under a uniform data scale.

[0040] The risk assessment module is used to obtain a real-time risk score for high-altitude operations based on a pre-processed real-time dataset and a pre-built risk assessment model. The risk assessment model is a model that calculates a risk score by comprehensively considering pre-selected risk factors based on a given dataset. The risk factors include at least worker risk factors, environmental risk factors, and group management risk factors.

[0041] The risk prediction module is used to input the pre-processed real-time dataset into a pre-built risk prediction model to obtain the predicted risk probability of high-altitude operations. The risk prediction model is based on a neural network and is used to output the risk probability according to the given dataset.

[0042] The fall protection alarm module is used for real-time fall protection alarms during high-altitude operations based on real-time risk scoring and predicted risk probabilities.

[0043] Thirdly, the present invention provides a computer device, the device including a processor and a memory:

[0044] The memory is used to store computer programs and send the instructions of the computer programs to the processor;

[0045] The processor executes instructions from the computer program, such as the first aspect of a comprehensive analysis method for high-altitude fall prevention based on multi-dimensional data from high-altitude operation monitoring.

[0046] Fourthly, the present invention provides a computer-readable storage medium storing a computer program, which, when executed by a processor, implements a comprehensive analysis method for high-altitude fall prevention based on multi-dimensional data from high-altitude operation monitoring, as described in the first aspect.

[0047] In summary, this invention provides a comprehensive analysis method for fall prevention based on multi-dimensional data from high-altitude operations monitoring. This method includes acquiring position and posture data, physiological data, operational environment data, and group management data of each worker in a high-altitude operations group, forming a real-time dataset. The group management data is used to evaluate the effectiveness of group management. The data in the real-time dataset is preprocessed to obtain a real-time dataset with a unified data scale. Based on the preprocessed real-time dataset, a pre-constructed risk assessment model is used to obtain a real-time risk score for the high-altitude operations. The risk assessment model is a model that comprehensively calculates the risk score based on pre-selected risk factors according to a given dataset. The risk factors include at least worker risk factors, environmental risk factors, and group management risk factors. The preprocessed real-time dataset is input into a pre-constructed risk prediction model to obtain the predicted risk probability of the high-altitude operations. The risk prediction model is based on a neural network and is used to output the risk probability based on a given dataset. Real-time fall prevention alarms for high-altitude operations are generated based on the real-time risk score and the predicted risk probability. This invention integrates multi-dimensional data such as location, physiology, environment, and group management, and conducts comprehensive analysis through risk assessment and prediction models, significantly improving the comprehensiveness and accuracy of safety assessments. Addressing the shortcomings of existing warning systems, such as frequent false alarms and missed alarms, and the lack of collaborative group management, this invention provides real-time fall protection alarms based on dual-model results and incorporates group management data, effectively improving the accuracy and reliability of warnings, achieving group-level safety strategy management, and meeting the needs of modern high-altitude operations management.

[0048] The present invention also provides a high-altitude fall prevention comprehensive analysis device, computer equipment and computer-readable storage medium based on multi-dimensional data of high-altitude operation monitoring, which has similar effects to the above methods when implemented, and will not be described in detail here. Attached Figure Description

[0049] To more clearly illustrate the technical solutions in the embodiments of the present invention or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0050] Figure 1 A flowchart illustrating a comprehensive analysis method for fall prevention based on multi-dimensional data from high-altitude operations monitoring, provided in an embodiment of the present invention;

[0051] Figure 2 A block diagram of a high-altitude fall prevention comprehensive analysis device based on multi-dimensional data of high-altitude operation monitoring provided in an embodiment of the present invention;

[0052] Figure 3This is a block diagram of a computer device provided in an embodiment of the present invention. Detailed Implementation

[0053] To make the objectives, features, and advantages of this invention more apparent and understandable, the technical solutions of the embodiments of this invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the embodiments described below are only some embodiments of this invention, and not all embodiments. Based on the embodiments of this invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of this invention.

[0054] Please see Figure 1 This embodiment provides a comprehensive analysis method for fall prevention based on multi-dimensional data from high-altitude operations monitoring, including the following steps:

[0055] S11: Acquire the position and posture data, physiological data, work environment data, and group management data of each worker in the high-altitude work group, and form a real-time dataset; the group management data is used to evaluate the effectiveness of group management.

[0056] It should be noted that a high-altitude work group refers to a group of personnel working collaboratively in the same high-altitude work scenario. Position and attitude data refers to the spatial coordinates and body attitude parameters (such as pitch and roll angles) of the workers, acquired through devices such as GNSS (Global Navigation Satellite System) and Inertial Measurement Unit (IMU). Physiological data refers to physiological indicators reflecting the workers' physical condition, such as heart rate, blood oxygen saturation, and body temperature, collected using wearable devices. Work environment data includes environmental parameters such as wind speed, light intensity, temperature, and humidity. Group management data refers to data on group management based on the devices worn by each worker in the same high-altitude work scenario to achieve collaborative work and reduce the risks of high-altitude operations. Examples include data measuring the stability of communication between devices and the timeliness of early warnings.

[0057] This step involves collecting multi-source heterogeneous data in real time through various intelligent protective devices worn by workers (such as smart safety belts / safety helmets) and transmitting it to the management platform or monitoring host. The data is then synchronously integrated according to timestamps to form a real-time dataset containing spatiotemporal dimensions.

[0058] S12: Preprocess the data in the real-time dataset to obtain a real-time dataset with a uniform data scale.

[0059] It should be noted that unifying the data scale refers to mapping data with different dimensions and value ranges to a unified numerical range through standardization (Z-score standardization) or normalization (Min-Max normalization), thereby eliminating the impact of dimensional differences on subsequent analysis.

[0060] This step uses data standardization algorithms in feature engineering to convert heterogeneous data such as location data (meter-level) and physiological data (percentage / numerical value) into dimensionless standardized feature vectors, ensuring data comparability.

[0061] S13: Based on the preprocessed real-time dataset, obtain the real-time risk score for high-altitude operations using a pre-built risk assessment model; the risk assessment model is a model that calculates the risk score by comprehensively considering pre-selected risk factors based on the given dataset; the risk factors include at least worker risk factors, environmental risk factors, and group management risk factors.

[0062] It should be noted that the risk assessment model is a multi-factor comprehensive evaluation model, which can be constructed using the analytic hierarchy process (AHP), fuzzy comprehensive evaluation method, or weighted linear combination model. Risk factors include worker risk factors (such as physiological state, equipment wearing status, and protective status), environmental risk factors (such as wind speed and temperature), and group management risk factors (such as communication stability and the timeliness of early warning data).

[0063] This step constructs a risk factor indicator system based on historical accident data and expert knowledge, assigns weights to each factor (such as determining the weight vector through the analytic hierarchy process), substitutes the preprocessed data into the weighted summation model, outputs a real-time risk score, and quantifies the current operational risk level.

[0064] S14: Input the preprocessed real-time dataset into the pre-built risk prediction model to obtain the predicted risk probability of high-altitude operations; the risk prediction model is a model built on a neural network and is used to output the risk probability based on the given dataset.

[0065] It should be noted that the risk prediction model uses a deep learning neural network architecture, such as LSTM (Long Short-Term Memory Network), TCN (Temporal Convolutional Network), or Transformer model, which is suitable for pattern recognition of time series data.

[0066] This step utilizes labeled historical datasets (containing spatiotemporal data and risk event labels) to supervise the learning of neural networks. The model captures implicit risk patterns in the data through multi-layer nonlinear transformations. After inputting the current real-time dataset, it calculates the risk probability distribution through forward propagation, thereby achieving forward-looking risk prediction.

[0067] S15: Real-time fall protection alarm for high-altitude operations based on real-time risk scoring and predicted risk probability.

[0068] It should be noted that the real-time fall protection alarm is based on a two-dimensional decision-making mechanism of risk score and predicted probability, and triggers multi-level warnings (such as yellow, orange and red warnings) through preset thresholds.

[0069] This step combines real-time risk scoring (reflecting the current risk status) with predicted risk probability (reflecting future risk trends) for a comprehensive judgment. It can further combine alarms, mobile push notifications, and other methods to execute real-time early warnings, thereby realizing fall prevention early warnings for high-altitude operations.

[0070] This embodiment provides a comprehensive analysis method for fall prevention based on multi-dimensional data from high-altitude operations monitoring. This method overcomes the limitations of traditional single-dimensional monitoring by integrating physiological / behavioral data of workers, environmental parameters, and group management data. It adopts a heterogeneous model combination of risk assessment model and neural network prediction model to achieve the dual capabilities of quantifying current risks and predicting future risks. Based on a real-time data stream processing architecture, it constructs a dynamically adaptive fall prevention early warning mechanism, providing a data-driven intelligent solution for high-altitude operations safety management, effectively improving the comprehensiveness of risk identification and the foresight of early warning.

[0071] In one embodiment of the present invention, acquiring the position and posture data, physiological data, work environment data, and group management data of each worker in a high-altitude work group includes:

[0072] S21: Acquire the position, posture, and physiological data of workers by using smart safety belts and smart safety helmets worn by workers at height;

[0073] S22: Obtain operational environment data through micro-weather stations; operational environment data shall include at least temperature and humidity data and wind speed data at the operational site;

[0074] S23: Obtain group management data through the management system; group management data includes communication stability data of smart seat belts and smart helmets, as well as alarm prompt timeliness data; communication stability data is used to assess whether communication is stable, and alarm prompt timeliness data is used to assess whether alarm prompts are timely.

[0075] In this embodiment, for high-altitude work scenarios, multi-dimensional data is acquired through various means to form a real-time dataset. Specifically, the location, posture, and physiological data of the workers are collected using smart safety belts and smart helmets worn by the workers; environmental data, including at least temperature, humidity, and wind speed, is collected at the work site using micro-weather stations; and group management data is acquired through a management system, including communication stability data for evaluating the stability of communication between the smart safety belts and smart helmets, and alarm timeliness data for evaluating the timeliness of alarm prompts. It should be noted that all the devices and management systems involved in this embodiment for real-time data acquisition can be existing devices and systems that meet the relevant real-time data acquisition requirements.

[0076] Furthermore, meaningful features can be extracted from the collected data. For example, motion trajectory features can be extracted from location data; health indicators such as heart rate and respiratory rate can be extracted from physiological data; and environmental parameters such as temperature and humidity can be extracted from environmental data. By analyzing the correlations between different types of features, we can identify which combinations best reflect the current state of human operation and potential risks in the current environment, and use this as the basis for subsequent risk assessment and prediction.

[0077] In one embodiment of the present invention, data in a real-time dataset is preprocessed to obtain a real-time dataset with a uniform data scale, including:

[0078] S31: Perform data cleaning on the real-time dataset, removing outliers and filling in missing values;

[0079] S32: Normalize the various types of data after data cleaning to form a real-time dataset under a unified data scale.

[0080] In this embodiment, for the acquired real-time dataset, data cleaning is first performed on the data, that is, outliers are removed and missing values ​​are filled in using appropriate methods to improve data quality. Then, the cleaned data is normalized to transform data with different dimensions and value ranges to a unified data scale, ultimately forming a real-time dataset with a unified data scale suitable for subsequent analysis and processing. For example, normalization can employ existing max-min normalization methods, Z-score standardization, and nonlinear normalization methods. The specific normalization method is determined based on the characteristics of each data item.

[0081] In one embodiment of the present invention, the risk assessment model is a calculation model constructed based on the weighting method;

[0082] Based on the preprocessed real-time dataset, a pre-built risk assessment model is used to obtain a real-time risk score for high-altitude operations, including:

[0083] S41: Identify risk factors; among which, worker risk factors include at least the wearing of safety belts and helmets, the protective condition of safety belts and helmets, and the condition of the workers; environmental risk factors include at least the working environment; and group management risk factors include at least the effectiveness of group management.

[0084] It should be noted that this step is based on an analysis of the actual scenarios and risk sources of high-altitude operations, identifying potential factors affecting the risks of high-altitude operations from three main aspects: personnel, environment, and group management. Regarding personnel, the wearing and protective status of safety belts and helmets directly affect the safety of workers, and the workers' condition reflects the impact of their physical and behavioral state on risk. Regarding the environment, the working environment conditions include the impact of environmental factors such as temperature, humidity, and wind speed on the operation. Regarding group management, the effectiveness of group management includes the impact of the stability of intelligent device communication and the timeliness of alarms on overall operational safety management. By comprehensively reviewing these aspects, the factors that need to be considered in the risk assessment are clarified.

[0085] S42: Calculate the values ​​of each risk factor based on the preprocessed real-time dataset.

[0086] It should be noted that the wearing and protective status of the seat belt and helmet can be determined based on sensor data fed back from the smart seat belt and helmet to determine whether the wearing is correct and whether the protective status is normal. If both wearing and protection are normal, a value of 0 is assigned; if both are abnormal, a value of 1 is assigned; otherwise, a value in [0,1] is assigned according to the actual situation.

[0087] The status of workers is determined by position and posture data. For example, the movement status factor value is quantified by indicators such as the rate of position change and posture stability. Based on physiological data (such as heart rate and blood pressure), the physiological status factor value is quantified and calculated according to the degree of deviation.

[0088] The calculation of environmental risk factor values ​​is based on operational environment data (such as temperature, humidity, and wind speed). The actual measured values ​​are compared with the preset range suitable for operation, and the environmental risk factor value is quantified according to the degree of deviation from the suitable range. For example, the greater the temperature exceeds the suitable range, the higher the environmental risk factor value.

[0089] The calculation of group management risk factor values ​​is achieved by comparing the communication stability data of smart safety belts and safety helmets (such as data transmission success rate, packet loss rate, etc.) and alarm prompt timeliness data obtained by the management system with preset standards, and quantifying the group management effect value according to the degree of deviation, which is then used as the group management risk factor value.

[0090] S43: Determine the weight of each risk factor.

[0091] It should be noted that methods such as the Analytic Hierarchy Process (AHP) can be used. By inviting domain experts, pairwise comparisons of the relative importance of each risk factor are made to construct a judgment matrix. For example, based on their experience and expertise, experts determine whether worker risk factors or environmental risk factors have a greater impact on the risks of working at heights, and the proportional relationship of their impact. By calculating the eigenvectors of the judgment matrix, the weights of each risk factor are determined to reflect their relative importance in the overall risk assessment.

[0092] S44: Substitute the values ​​and corresponding weights of each risk factor into the risk assessment model, and calculate the real-time risk score for high-altitude operations by weighted summation.

[0093] In this embodiment, the risk assessment model is constructed based on the weighted method. First, using a pre-processed real-time dataset, the pre-built model is used to obtain a real-time risk score for high-altitude operations. The first step in this process is to identify risk factors, covering worker risk factors (at least including the wearing of safety belts and helmets, protective status, and worker condition), environmental risk factors (at least including the working environment conditions), and group management risk factors (at least including the effectiveness of group management). Next, the values ​​of each risk factor are calculated based on the pre-processed real-time dataset. Then, the weight of each risk factor is determined. Finally, the values ​​of each risk factor and their corresponding weights are substituted into the risk assessment model, and a weighted summation method is used to obtain the real-time risk score for the high-altitude operation, thereby assessing the current level of risk. It should be noted that the weighted method is a common multi-factor comprehensive evaluation method in existing technologies and is widely used in many fields.

[0094] In a further embodiment of the present invention, the values ​​of each risk factor are calculated based on the preprocessed real-time dataset, including:

[0095] S51: For the wearing status of seat belts and helmets, as well as the protective status of seat belts and helmets, the value is determined by assigning values ​​in [0,1].

[0096] The value is 0 if both the wearing condition and the protective status are normal, and 1 if both the wearing condition and the protective status are abnormal. The remaining values ​​are in the range of [0,1].

[0097] S52: For the status of the workers, the movement status factor value and physiological status factor value of the workers are quantitatively calculated using position and posture data and physiological data respectively.

[0098] S53: For environmental risk factors, the values ​​of environmental risk factors are obtained by quantitative calculation using operational environment data;

[0099] S54: To assess the effectiveness of group management, the group management effectiveness value is obtained by quantitatively calculating the group management data.

[0100] In each quantitative calculation process, if each data point is within its preset range, the corresponding factor value is assigned to 0; otherwise, the corresponding factor value is obtained by quantifying the degree to which the data deviates from the preset range.

[0101] In this embodiment, the wearing and protective status of safety belts and helmets are determined by data fed back from intelligent devices, and values ​​are assigned in the range [0,1] based on the quality of the status to quantify the risk factor value. For the status of workers, movement-related features are extracted from position and posture data, and physiological data is compared with the normal range to quantify the movement status factor value and physiological status factor value respectively. For environmental risk factors, the work environment data is compared with the preset suitable work environment range, and the environmental risk factor value is obtained according to the degree of deviation. Regarding the group management effect, the relevant indicators in the group management data are compared with the preset standard, and the group management effect value is quantified according to the deviation.

[0102] For example, the wearing and protective status of safety belts and helmets can be directly assigned values ​​using an assignment method. For instance, "safety belt wearing status" can be divided into levels such as "fully worn and correctly," "partially worn or incorrectly worn," and "not worn," and assigned corresponding risk scores. When calculating movement status factor values ​​using position and posture data, if the position changes frequently and the posture is unstable over a period of time, it indicates that the worker's movement status is relatively dangerous. Values ​​can be quantified in the range [0,1] based on a preset relationship between the degree of danger and the numerical value. For physiological status factor values, taking heart rate as an example, if the normal range is set to [60,100] beats / minute, values ​​exceeding this range are calculated in the range [0,1] based on the degree of excess. The greater the excess, the closer the value is to 1. Environmental risk factor values ​​and group management effect values ​​can be obtained similarly (calculated as an example only). Assuming the weights of each factor have been determined using the Analytic Hierarchy Process (AHP): seatbelt wearing status 0.3, seatbelt protection status 0.2, worker status 0.2, work environment 0.15, and group management effectiveness 0.15, and corresponding risk scoring standards have been established for each factor, then for a specific work process, the risk score of each factor can be calculated separately and substituted into the risk assessment function to obtain the total risk value.

[0103] In one embodiment of the present invention, the risk prediction model is constructed based on a long short-term memory network. The risk prediction model includes an input layer, an LSTM layer, and a fully connected layer. The number of neurons in the input layer is determined according to the number of features in the real-time dataset.

[0104] In this embodiment, the risk prediction model is built based on a Long Short-Term Memory (LSTM) network. This model consists of an input layer, an LSTM layer, and a fully connected layer. The input layer receives the real-time dataset, and the number of neurons is determined by the number of features in the real-time dataset to ensure complete reception of data features from all dimensions. The LSTM layer utilizes a gating mechanism to effectively handle long-term dependencies in sequence data and extract features from the input data. The fully connected layer further performs nonlinear transformations and combinations on the features output by the LSTM layer, ultimately enabling the model to predict the probability of risks associated with high-altitude operations based on the real-time dataset. Deep learning techniques can also be introduced at this stage to learn and recognize more complex patterns. For example, advanced architectures such as Convolutional Neural Networks (CNNs) and Recurrent Neural Networks (RNNs) can better understand the underlying regularities of time-series signals.

[0105] In one embodiment of the present invention, a real-time fall protection alarm for high-altitude operations based on real-time risk scoring and predicted risk probability includes:

[0106] If the real-time risk score exceeds the preset risk score threshold, a risk alarm will be issued in real time.

[0107] If the predicted risk probability exceeds the preset risk probability threshold, a risk warning will be issued in real time.

[0108] In this embodiment, a dual-indicator judgment mechanism is adopted for real-time fall prevention alarms for high-altitude operations. Firstly, the real-time risk score obtained based on a weighted method is compared with a preset risk score threshold. If the real-time risk score exceeds the threshold, a risk alarm is immediately issued. Secondly, the risk probability predicted based on a long short-term memory network model is compared with a preset risk probability threshold. If the predicted risk probability is higher than the threshold, a risk warning is issued in real time. Through these two methods, high-altitude operation risks are monitored and warned of from different angles, enabling timely detection of potential fall risks and ensuring the safety of personnel working at heights.

[0109] The following is an implementation example designed based on the above embodiments and actual operating scenarios:

[0110] 1. Equipment Deployment:

[0111] Seat belt monitoring equipment: It integrates a main monitoring module, an auxiliary monitoring module, and a hook monitoring module, and has built-in positioning module, IMU (Inertial Measurement Unit), attitude sensor, air pressure sensor, micro-motion sensor, etc.

[0112] Smart safety helmets: The smart safety helmets worn by workers integrate sensors such as GPS modules, IMU, body temperature sensors, and blood pressure monitors.

[0113] Environmental monitoring equipment: Temperature sensors, humidity sensors, wind speed sensors and other environmental sensors are installed at the work site.

[0114] 2. Data Collection:

[0115] Position and pose data:

[0116] Location module: Collects the geographical location data of the workers once per second.

[0117] IMU: Collects the worker's posture data (such as tilt angle, acceleration, etc.) once per second.

[0118] Physiological data:

[0119] Heart rate and body temperature sensors: Collect heart rate data of workers every minute.

[0120] Blood pressure monitor: Collects workers' blood pressure data every 10 minutes.

[0121] Environmental data:

[0122] Temperature sensor: Collects temperature data of the work site once per minute.

[0123] Humidity sensor: Collects humidity data at the work site once per minute.

[0124] Wind speed sensor: Collects wind speed data at the work site once per minute.

[0125] Group management data:

[0126] Data transmission success rate: This is obtained by statistically analyzing the ratio of the amount of data successfully transmitted to the management system by smart safety belts and smart safety helmets within a certain time period (e.g., every 5 minutes as a statistical cycle) to the total amount of data sent.

[0127] Alarm response time: The time interval is recorded from the moment an event that may cause a risk of falling from a height is detected until the moment the smart device issues an alarm.

[0128] 3. Data preprocessing:

[0129] Filtering:

[0130] A low-pass filter is used to filter the IMU data to remove high-frequency noise.

[0131] A high-pass filter is used to filter the heart rate data to remove low-frequency noise.

[0132] Noise reduction:

[0133] The Kalman filter algorithm is applied to denoise the location data, thereby improving the accuracy of the data.

[0134] Wavelet transform is used to denoise physiological data and reduce data fluctuations.

[0135] Normalization:

[0136] Normalize all data to the range of 0-1 to facilitate subsequent analysis and processing.

[0137] 4. Data Analysis:

[0138] Data analysis steps:

[0139] Position and attitude analysis:

[0140] The Support Vector Machine (SVM) algorithm is used to analyze position and orientation data to determine whether the worker is in a dangerous situation (such as being suspended or tilted).

[0141] Physiological state analysis:

[0142] Using neural network algorithms to analyze physiological data, assess the physical condition of workers and determine whether they are suitable to continue working (e.g., abnormal heart rate, high blood pressure, etc.).

[0143] Environmental impact analysis:

[0144] By combining environmental data, regression analysis is used to assess the impact of on-site environmental conditions on work safety (such as high temperature, high humidity, strong wind, etc.).

[0145] 5. Comprehensive Assessment:

[0146] Safety risk index calculation:

[0147] By combining location, posture, physiological state, environmental data, and group management data, a multi-factor comprehensive assessment model is used to calculate the safety risk index of workers.

[0148] 6. Early Warning and Response:

[0149] Early warning and response steps:

[0150] Early warning mechanism:

[0151] When the safety risk index exceeds a preset threshold, the system automatically triggers an alarm. The alarm information is then sent to the mobile terminal of the on-site management personnel and the smart safety helmet of the workers via a wireless communication module.

[0152] The system uses a risk warning model to calculate the probability of a risk. When the probability exceeds a preset threshold, the system automatically triggers a warning. Warning information can be delivered through various means such as voice, vibration, and light.

[0153] Response strategy:

[0154] Different response strategies are provided based on different risk levels, such as suspending operations and evacuating the site.

[0155] Based on the early warning information, on-site management personnel promptly take corresponding safety measures to ensure the safety of workers.

[0156] As described above, this invention integrates multiple sensors to comprehensively collect data on the position, posture, physiological state, and environment of workers. It then utilizes machine learning algorithms to intelligently analyze this multi-dimensional data, achieving a comprehensive assessment of worker safety. In addition to real-time early warning capabilities, it incorporates an automated response mechanism to improve system reliability and efficiency. Furthermore, automated data collection and analysis reduce reliance on manual inspection and intervention. By employing a multi-factor comprehensive assessment model, it quantifies worker safety risks, providing a basis for scientific risk assessment and management. From data collection and analysis to response and risk quantification, this invention enhances the level of safety management for high-altitude operations.

[0157] Based on the same inventive concept, this application also provides a high-altitude fall prevention comprehensive analysis device based on high-altitude operation monitoring multi-dimensional data for implementing the high-altitude operation monitoring multi-dimensional data comprehensive analysis method mentioned above. The solution provided by this device is similar to the solution described in the above method. Therefore, the specific limitations in the embodiments of the high-altitude fall prevention comprehensive analysis device based on high-altitude operation monitoring multi-dimensional data provided below can be found in the limitations of the high-altitude fall prevention comprehensive analysis method based on high-altitude operation monitoring multi-dimensional data described above, and will not be repeated here.

[0158] Please see Figure 2 This invention also provides a comprehensive analysis device for fall prevention based on multi-dimensional data from high-altitude operations monitoring, comprising:

[0159] The data acquisition module is used to acquire the position and posture data, physiological data, work environment data, and group management data of each worker in the high-altitude work group, and form a real-time dataset; the group management data is used to evaluate the effectiveness of group management.

[0160] The data preprocessing module is used to preprocess the data in the real-time dataset to obtain a real-time dataset under a uniform data scale.

[0161] The risk assessment module is used to obtain a real-time risk score for high-altitude operations based on a pre-processed real-time dataset and a pre-built risk assessment model. The risk assessment model is a model that calculates a risk score by comprehensively considering pre-selected risk factors based on a given dataset. The risk factors include at least worker risk factors, environmental risk factors, and group management risk factors.

[0162] The risk prediction module is used to input the pre-processed real-time dataset into a pre-built risk prediction model to obtain the predicted risk probability of high-altitude operations. The risk prediction model is based on a neural network and is used to output the risk probability according to the given dataset.

[0163] The fall protection alarm module is used for real-time fall protection alarms during high-altitude operations based on real-time risk scoring and predicted risk probabilities.

[0164] Those skilled in the art will clearly understand that, for the sake of convenience and brevity, the above-described division of functional units and modules is merely an example. In practical applications, the above functions can be assigned to different functional units and modules as needed, that is, the internal structure of the system can be divided into different functional units or modules to complete all or part of the functions described above. The functional units and modules in the embodiments can be integrated into one processing unit, or each unit can exist physically separately, or two or more units can be integrated into one unit. The integrated unit can be implemented in hardware or as a software functional unit. Furthermore, the specific names of the functional units and modules are only for easy differentiation and are not intended to limit the scope of protection of this application. The specific working process of the units and modules in the above system can be referred to the corresponding process in the foregoing method embodiments, and will not be repeated here.

[0165] Reference Figure 3 The present invention also provides a computer device, including: a memory and a processor, and a computer program stored in the memory. When the computer program is executed on the processor, it implements the high-altitude fall prevention comprehensive analysis method based on multi-dimensional data of high-altitude operation monitoring as described in any of the above methods.

[0166] The computer device may be a desktop computer, laptop, handheld computer, or cloud server, etc. This computer device may include, but is not limited to, a processor and memory. Those skilled in the art will understand that... Figure 3 The examples of computer devices are merely examples and do not constitute a limitation on computer devices. They may include more or fewer components than shown in the illustration, or combinations of certain components, or different components. For example, they may also include input / output devices, network access devices, etc.

[0167] The processor referred to can be a Central Processing Unit (CPU), but it can also be other general-purpose processors, digital signal processors (DSPs), application-specific integrated circuits (ASICs), field-programmable gate arrays (FPGAs), or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components, etc. A general-purpose processor can be a microprocessor or any conventional processor.

[0168] In some embodiments, the memory may be an internal storage unit of the computer device, such as a hard drive or RAM. In other embodiments, the memory may be an external storage device of the computer device, such as a plug-in hard drive, Smart Media Card (SMC), Secure Digital (SD) card, or Flash Card. Furthermore, the memory may include both internal and external storage units of the computer device. The memory is used to store the operating system, applications, bootloader, data, and other programs, such as the program code of the computer program. The memory can also be used to temporarily store data that has been output or will be output.

[0169] This invention also provides a computer-readable storage medium storing a computer program thereon. When the computer program is run by a processor, it implements the comprehensive analysis method for high-altitude fall prevention based on multi-dimensional data from high-altitude operation monitoring, as described in any of the above methods.

[0170] In this embodiment, if the integrated unit is implemented as a software functional unit and sold or used as an independent product, it can be stored in a computer-readable storage medium. Based on this understanding, all or part of the processes in the methods of the above embodiments can be implemented by a computer program instructing related hardware. The computer program can be stored in a computer-readable storage medium, and when executed by a processor, it can implement the steps of the various method embodiments described above. The computer program includes computer program code, which can be in the form of source code, object code, executable files, or certain intermediate forms. The computer-readable medium can include at least: any entity or device capable of carrying computer program code to a photographing device / terminal device, a recording medium, a computer memory, a read-only memory (ROM), a random access memory (RAM), an electrical carrier signal, a telecommunication signal, and a software distribution medium. Examples include USB flash drives, portable hard drives, magnetic disks, or optical disks. In some jurisdictions, according to legislation and patent practice, computer-readable media cannot be electrical carrier signals or telecommunication signals.

[0171] This invention provides a computer program product, including a computer program that, when executed by a processor, implements the comprehensive analysis method for high-altitude fall prevention based on multi-dimensional data from high-altitude operation monitoring, as described in any of the above methods.

[0172] In the above embodiments, the descriptions of each embodiment have different focuses. For parts that are not described in detail or recorded in a certain embodiment, please refer to the relevant descriptions of other embodiments.

[0173] Those skilled in the art will recognize that the units and algorithm steps of the various examples described in conjunction with the embodiments disclosed herein can be implemented in electronic hardware, or a combination of computer software and electronic hardware. Whether these functions are implemented in hardware or software depends on the specific application and design constraints of the technical solution. Those skilled in the art can use different methods to implement the described functions for each specific application, but such implementation should not be considered beyond the scope of this application.

[0174] In the embodiments disclosed in this application, it should be understood that the disclosed devices / terminal equipment and methods can be implemented in other ways. For example, the device / terminal equipment embodiments described above are merely illustrative. For instance, the division of modules or units is only a logical functional division, and in actual implementation, there may be other division methods. For example, multiple units or components may be combined or integrated into another system, or some features may be ignored or not executed. Furthermore, the displayed or discussed mutual coupling or direct coupling or communication connection may be through some interfaces; the indirect coupling or communication connection between devices or units may be electrical, mechanical, or other forms.

[0175] The above embodiments are only used to illustrate the technical solutions of the present invention, and are not intended to limit it. Although the present invention has been described in detail with reference to the foregoing embodiments, those skilled in the art should understand that modifications can still be made to the technical solutions described in the foregoing embodiments, or equivalent substitutions can be made to some of the technical features. Such modifications or substitutions do not cause the essence of the corresponding technical solutions to deviate from the spirit and scope of the technical solutions of the embodiments of the present invention.

Claims

1. A comprehensive analysis method for fall prevention at heights based on multi-dimensional data from high-altitude operations monitoring, characterized in that, Includes the following steps: The system acquires the position and posture data, physiological data, work environment data, and group management data of each worker in the high-altitude work group, and forms a real-time dataset; the group management data is used to evaluate the effectiveness of group management. The data in the real-time dataset is preprocessed to obtain the real-time dataset under a uniform data scale; Based on the preprocessed real-time dataset, a real-time risk score for high-altitude operations is obtained using a pre-built risk assessment model. The risk assessment model is a model that calculates a risk score by comprehensively analyzing pre-selected risk factors based on a given dataset; the risk factors include at least worker risk factors, environmental risk factors, and group management risk factors. The preprocessed real-time dataset is input into a pre-built risk prediction model to obtain the predicted risk probability of high-altitude operations; The risk prediction model is based on a neural network and is used to output risk probabilities based on a given dataset. Real-time fall protection alarm for high-altitude operations based on the real-time risk score and the predicted risk probability.

2. The comprehensive analysis method for fall prevention based on multi-dimensional data from high-altitude operation monitoring according to claim 1, characterized in that, Acquire the position and posture data, physiological data, work environment data, and group management data of each worker in the high-altitude work group, including: The location and posture data, as well as the physiological data of the workers, are obtained by using smart safety belts and smart safety helmets worn by workers at height. The operational environment data is obtained through a micro-weather station; the operational environment data includes at least temperature and humidity data and wind speed data at the operational site; The group management data is obtained through the management system; the group management data includes communication stability data of the smart safety belt and the smart safety helmet, as well as alarm prompt timeliness data; the communication stability data is data to assess whether the communication is stable, and the alarm prompt timeliness data is data to assess whether the alarm prompt is timely.

3. The comprehensive analysis method for fall prevention based on multi-dimensional data from high-altitude operation monitoring as described in claim 1, characterized in that, The data in the real-time dataset is preprocessed to obtain the real-time dataset under a uniform data scale, including: The data in the real-time dataset is cleaned to remove outliers and fill in missing values. After data cleaning, the various types of data are normalized to form the real-time dataset under a unified data scale.

4. The comprehensive analysis method for fall prevention based on multi-dimensional data from high-altitude operation monitoring according to claim 1, characterized in that, The risk assessment model is a calculation model constructed based on the weighting method; Based on the preprocessed real-time dataset, a real-time risk score for high-altitude operations is obtained using a pre-built risk assessment model, including: Identify the risk factors; wherein, the risk factors for workers include at least the wearing status of safety belts and helmets, the protective status of safety belts and helmets, and the status of workers; the risk factors for the environment include at least the working environment conditions; and the risk factors for group management include at least the effectiveness of group management. Based on the preprocessed real-time dataset, the values ​​of each risk factor are calculated; Determine the weight of each of the aforementioned risk factors; The values ​​and corresponding weights of each risk factor are substituted into the risk assessment model, and the real-time risk score for high-altitude operations is calculated by weighted summation.

5. The comprehensive analysis method for fall prevention based on multi-dimensional data from high-altitude operation monitoring according to claim 4, characterized in that, Based on the preprocessed real-time dataset, the values ​​of each risk factor are calculated, including: For the wearing status of the seat belt and helmet, as well as the protective status of the seat belt and helmet, the value is determined by assigning values ​​in [0,1]. If both the wearing condition and the protective status are normal, the value is 0; if both the wearing condition and the protective status are abnormal, the value is 1; and for other cases, the value is taken in [0,1]. For the worker's status, the worker's movement status factor value and physiological status factor value are quantitatively calculated using the position and posture data and the physiological data, respectively. For the aforementioned environmental risk factors, the environmental risk factor values ​​are obtained by quantitative calculation using the aforementioned work environment data; The group management effect is calculated quantitatively using the group management data to obtain the group management effect value. In each quantification calculation process, if each data point is within its preset range, the corresponding factor value is assigned to 0; otherwise, the corresponding factor value is obtained by quantification based on the degree to which the data deviates from the preset range.

6. The comprehensive analysis method for fall prevention based on multi-dimensional data from high-altitude operation monitoring according to claim 1, characterized in that, The risk prediction model is constructed based on a long short-term memory network. The risk prediction model includes an input layer, an LSTM layer, and a fully connected layer. The number of neurons in the input layer is determined according to the number of features in the real-time dataset.

7. The comprehensive analysis method for fall prevention based on multi-dimensional data from high-altitude operation monitoring according to claim 1, characterized in that, Real-time fall protection alarm for high-altitude operations based on the real-time risk score and the predicted risk probability includes: If the real-time risk score exceeds the preset risk score threshold, a risk alarm will be issued in real time. If the predicted risk probability exceeds the preset risk probability threshold, a risk warning will be issued in real time.

8. A comprehensive analysis device for fall prevention based on multi-dimensional data from high-altitude operation monitoring, characterized in that, include: The data acquisition module is used to acquire the position and posture data, physiological data, work environment data, and group management data of each worker in the high-altitude work group, and form a real-time dataset; the group management data is used to evaluate the effectiveness of group management. The data preprocessing module is used to preprocess the data in the real-time dataset to obtain the real-time dataset under a unified data scale; The risk assessment module is used to obtain a real-time risk score for high-altitude operations based on the preprocessed real-time dataset and a pre-built risk assessment model. The risk assessment model is a model that calculates a risk score by comprehensively analyzing pre-selected risk factors based on a given dataset; the risk factors include at least worker risk factors, environmental risk factors, and group management risk factors. The risk prediction module is used to input the preprocessed real-time dataset into a pre-built risk prediction model to obtain the predicted risk probability of high-altitude operations. The risk prediction model is based on a neural network and is used to output risk probabilities based on a given dataset. A fall protection alarm module is used to provide real-time fall protection alarms for high-altitude operations based on the real-time risk score and the predicted risk probability.

9. A computer device, characterized in that, The device includes a processor and a memory: The memory is used to store computer programs and send the instructions of the computer programs to the processor; The processor executes, according to the instructions of the computer program, a comprehensive analysis method for high-altitude fall prevention based on multi-dimensional data of high-altitude operation monitoring as described in any one of claims 1-7.

10. A computer-readable storage medium, characterized in that, The computer-readable storage medium stores a computer program, which, when executed by a processor, implements a comprehensive analysis method for high-altitude fall prevention based on multi-dimensional data from high-altitude operation monitoring as described in any one of claims 1-7.

Images