Real-time monitoring and analyzing method and system for safe production of workshop equipment based on Internet of Things

Abstract

The invention relates to the technical field of Internet of Things control, and discloses a workshop equipment safety production real-time monitoring analysis method and system based on the Internet of Things, and the method comprises the steps: carrying out the standardization preprocessing of multi-source sensor data, generating a multi-dimensional safety data stream, and storing the multi-dimensional safety data stream in a database; an equipment characteristic parameter set, an environment characteristic parameter set and a personnel behavior characteristic parameter set are extracted in parallel, a real-time mutual influence relation among the three is analyzed, a system coupling relation network is generated, equipment fault risk prediction is carried out, a basic risk prediction value is obtained, then dynamic correction is carried out, and an equipment dynamic risk index is generated; and finally, risk assessment is carried out, a comprehensive safety risk level is generated, and early warning and visual display are carried out based on the level. According to the invention, the efficiency of real-time monitoring and analysis of safe production of workshop equipment based on the Internet of Things can be improved.

Description

Technical Field

[0001] This invention relates to the field of Internet of Things (IoT) control technology, and in particular to a method and system for real-time monitoring and analysis of safe production of workshop equipment based on IoT. Background Technology

[0002] In the current field of workshop equipment safety production monitoring, most technical solutions only collect data and conduct risk assessments on a single dimension of equipment, environment, and personnel, lacking a systematic consideration of the real-time interaction between the three. Existing methods often process equipment operation data, environmental sensor data, and personnel behavior data in isolation, failing to capture the coupling relationship between equipment failures, environmental anomalies, and operational violations. This results in risk prediction relying solely on historical data from a single dimension, which is prone to misjudgment or omission. At the same time, the risk assessment models of traditional monitoring systems are mostly statically set and cannot adjust the prediction logic according to the dynamic changes of multiple factors in the workshop, making it difficult to adapt to the real-time safety management and control needs of complex production scenarios.

[0003] In terms of data processing and analysis workflows, existing technologies generally suffer from fragmented processes, failing to form a closed-loop mechanism of "correlation analysis - prediction correction." On the one hand, there is a lack of effective means to quantify the correlation strength between multiple safety elements, making it impossible to accurately characterize the degree of mutual influence between equipment, environment, and personnel. On the other hand, risk prediction results are not dynamically optimized in conjunction with real-time correlations, but only output static risk values ​​based on fixed algorithms, resulting in insufficient prediction accuracy and difficulty in providing early warnings of potential complex safety hazards. These deficiencies limit the timeliness of response and the effectiveness of decision-making in existing monitoring systems, making it impossible to provide comprehensive and accurate technical support for safe production in workshops. Therefore, how to adapt to the real-time safety management needs in complex production scenarios and provide comprehensive and accurate technical support has become an urgent problem to be solved. Summary of the Invention

[0004] This invention provides a method and system for real-time monitoring and analysis of workshop equipment safety production based on the Internet of Things, in order to solve the problems mentioned in the background art.

[0005] To achieve the above objectives, the present invention provides a method for real-time monitoring and analysis of workshop equipment safety production based on the Internet of Things, comprising: S1, acquire multi-source sensor data of target equipment, environment and personnel in the workshop in real time, perform spatiotemporal synchronization and standardization preprocessing on the multi-source sensor data, and generate a standardized multi-dimensional safety data stream of workshop equipment; S2, based on the multidimensional security data stream, extract in parallel the set of equipment feature parameters representing the health status of the target equipment, the set of environmental feature parameters representing the environmental hazard level, and the set of personnel behavior feature parameters representing the compliance of personnel operations; S3. Based on the set of equipment characteristic parameters, the set of environmental characteristic parameters, and the set of personnel behavior characteristic parameters, analyze the real-time mutual influence relationship among the target equipment, the environment, and the personnel, and generate a system coupling relationship network. S4. Based on the historical sequence of the equipment feature parameter set, predict the equipment failure risk to obtain a basic risk prediction value. Based on the system coupling relationship network, dynamically correct the basic risk prediction value to generate a corrected equipment dynamic risk index. S5. A risk assessment is performed by combining the equipment dynamic risk index, the environmental characteristic parameter set, and the personnel behavior characteristic parameter set to generate a comprehensive safety risk level. Based on the comprehensive safety risk level, an early warning and visualization display are provided.

[0006] In a preferred embodiment, the step of performing spatiotemporal synchronization and standardization preprocessing on the multi-source sensor data to generate a standardized multidimensional secure data stream includes: The data streams of each dimension in the multi-source sensor data are timestamped to generate a unified time series. Based on a predefined workshop spatial coordinate system, spatial location fusion is performed on each source data stream in the unified time series to generate a spatiotemporal data cube. The original data of each dimension in the spatiotemporal data cube are normalized to generate a standardized multidimensional secure data stream.

[0007] In a preferred embodiment, the step of extracting in parallel, based on the multidimensional security data stream, a set of equipment feature parameters representing the health status of the target equipment, a set of environmental feature parameters representing the environmental hazard level, and a set of personnel behavior feature parameters representing the compliance of personnel operations, includes: The time-domain features, frequency-domain features, and state change features of the multidimensional security data stream are extracted in parallel to generate a set of device feature parameters representing the health status of the target device; The multidimensional safety data stream is subjected to composite environmental index calculation and risk accumulation assessment to generate a set of environmental characteristic parameters representing the environmental hazard level; The multidimensional safety data stream is subjected to danger zone proximity analysis and operational compliance comparison to generate a set of personnel behavior feature parameters representing personnel operational compliance.

[0008] In a preferred embodiment, the step of analyzing the real-time mutual influence relationships among the target equipment, environment, and personnel based on the equipment feature parameter set, the environmental feature parameter set, and the personnel behavior feature parameter set, and generating a system coupling relationship network, includes: Extract multidimensional time series data of the equipment feature parameter set, the environmental feature parameter set, and the personnel behavior feature parameter set within a preset time window; Analyze the pairwise correlation strength between key parameters in the equipment feature parameter set and key parameters in the environmental feature parameter set, and simultaneously analyze the pairwise correlation strength between key parameters in the equipment feature parameter set and key parameters in the personnel behavior feature parameter set to generate an initial correlation strength matrix; Based on the set of significant relationships, a weighted undirected graph is constructed with key security parameters as nodes and the filtered relationship strength values ​​as edge weights, generating a system coupling relationship network.

[0009] In a preferred embodiment, the step of analyzing the pairwise correlation strength between key parameters in the equipment feature parameter set and key parameters in the environmental feature parameter set, and simultaneously analyzing the pairwise correlation strength between key parameters in the equipment feature parameter set and key parameters in the personnel behavior feature parameter set, to generate an initial correlation strength matrix, includes: The mathematical expressions used to analyze the pairwise correlation strength are as follows: ; In the formula, MI(X;Y) represents the mutual information value between random variables X and Y, X and Y represent the discretized time series of the two key security parameters whose correlation strength is to be calculated within the sliding window, x and y are the specific values ​​of variables X and Y in their discretized value range, p(x) and p(y) are the marginal probability distributions of variables X and Y, and p(x,y) is the joint probability distribution of variables X and Y.

[0010] In a preferred embodiment, the step of predicting equipment failure risk based on the historical sequence of the equipment feature parameter set to obtain a basic risk prediction value includes: The standardized time series data of the device feature parameter set over a preset prediction time span are extracted from the time series database to generate a historical sequence of device features; The device feature history sequence is input into a pre-trained long short-term memory network model for feature learning, resulting in a hidden state vector sequence corresponding to each historical time step. Based on the attention mechanism, the attention weight of the current prediction time to each historical time step in the hidden state vector sequence is calculated; Based on the attention weights, the hidden state vector sequence is weighted and summed to obtain a context vector that incorporates key historical information. This context vector is then mapped through a fully connected layer to generate the basic risk prediction value.

[0011] In a preferred embodiment, the step of dynamically correcting the basic risk prediction value based on the system coupling relationship network to generate a corrected equipment dynamic risk index includes: Based on the dynamic system coupling relationship network, identify all environmental feature parameters and personnel behavior feature parameters that have significant coupling edges with the key feature parameters of the target device; Calculate the average mutual information between the key feature parameters of the target device and all identified environmental feature parameters, and the average mutual information between the key feature parameters of the target device and all identified personnel behavior feature parameters, to generate the average environmental coupling strength and the average personnel coupling strength. The average coupling strength of the environment and the average coupling strength of the personnel are weighted and fused to generate a total coupling strength factor; The basic risk prediction value and the total coupling strength factor are dynamically corrected to generate the corrected equipment dynamic risk index.

[0012] In a preferred embodiment, the step of comprehensively assessing the risk based on the equipment dynamic risk index, the environmental characteristic parameter set, and the personnel behavior characteristic parameter set to generate a comprehensive safety risk level includes: The environmental feature parameter set and the personnel behavior feature parameter set are subjected to risk quantification transformation to generate environmental risk quantification value and personnel risk quantification value; By integrating the equipment dynamic risk index, the environmental risk quantification value, and the personnel risk quantification value, a comprehensive safety risk value is generated. Based on a preset risk threshold range, the comprehensive security risk value is mapped to the corresponding discrete level to generate a comprehensive security risk level.

[0013] In a preferred embodiment, the early warning and visualization display based on the comprehensive security risk level includes: Based on a predefined risk level-early warning action mapping table, the comprehensive safety risk level is matched to generate a corresponding multimodal early warning strategy. Based on the aforementioned multimodal early warning strategy, the system executes in parallel the audible and visual alarm drive, the monitoring interface message push, and the linkage control command issuance with the workshop control system. Based on the equipment dynamic risk index, the environmental feature parameter set, the personnel behavior feature parameter set, and the comprehensive safety risk level, visual information is integrated and rendered to generate a workshop safety situation dashboard view. The workshop safety status dashboard view is pushed to a preset monitoring terminal for display.

[0014] To address the above problems, the present invention also provides an Internet of Things-based real-time monitoring and analysis system for safe production of workshop equipment, the system comprising: The data synchronization module is used to acquire multi-source sensor data of target equipment, environment and personnel in the workshop in real time, perform spatiotemporal synchronization and standardization preprocessing on the multi-source sensor data, and generate a standardized multi-dimensional safety data stream of workshop equipment. The multidimensional safety parameter extraction and analysis module is used to extract, in parallel, a set of equipment characteristic parameters representing the health status of the target equipment, a set of environmental characteristic parameters representing the environmental hazard level, and a set of personnel behavior characteristic parameters representing the compliance of personnel operations, based on the multidimensional safety data stream. The coupling strength analysis module is used to analyze the real-time mutual influence relationship between the target equipment, the environment and personnel based on the set of equipment characteristic parameters, the set of environmental characteristic parameters and the set of personnel behavior characteristic parameters, and to generate a system coupling relationship network. The dynamic risk prediction and correction module is used to predict equipment failure risk based on the historical sequence of the equipment feature parameter set, obtain a basic risk prediction value, and dynamically correct the basic risk prediction value based on the system coupling relationship network to generate a corrected equipment dynamic risk index. The comprehensive assessment and intelligent early warning module is used to conduct risk assessment by comprehensively considering the equipment dynamic risk index, the environmental characteristic parameter set, and the personnel behavior characteristic parameter set, generate a comprehensive safety risk level, and provide early warning and visualization based on the comprehensive safety risk level.

[0015] Compared with the prior art, the present invention has the following beneficial effects: 1. This invention innovatively introduces a sequential analysis process of "first coupling analysis, then dynamic correction prediction," which completely changes the problem of the disconnect between risk prediction and multi-factor correlation analysis in traditional monitoring methods. This process first systematically sorts out the characteristic parameters of the three major safety elements—equipment, environment, and personnel—to establish a comprehensive data analysis foundation. Then, it deeply analyzes the real-time mutual influence relationship among the three. Finally, it dynamically corrects the basic risk prediction value of equipment failure based on the coupling correlation results. This logical closed loop not only allows risk prediction to move beyond the limitations of historical data of a single device but also fully integrates real-time influencing factors such as environmental fluctuations and personnel operation compliance, significantly improving the accuracy of the dynamic risk index of equipment. At the same time, it allows the risk assessment results to truly reflect the overall operating status of the workshop's safety production system, providing a more realistic scientific basis for safety decisions and effectively reducing the risk of misjudgment and omission caused by isolated analysis.

[0016] 2. This invention proposes a "dynamic coupling network based on mutual information" construction method, providing a standardized and operable technical solution for quantifying the correlation strength between safety elements. This method accurately calculates the pairwise correlation strength between key parameters of equipment, environment, and personnel through the mutual information algorithm, transforming the abstract mutual influence relationship into a concrete correlation strength matrix. Then, by constructing a weighted undirected graph to form a dynamic coupling network, it clearly presents the correlation topology and strength of each safety element. This quantification method breaks through the fuzziness limitations of traditional correlation analysis. It can not only capture the dynamic changes in the correlation relationship between elements in real time, providing accurate data support for risk correction in the sequential analysis process, but also help staff intuitively identify key correlation nodes and potential risk transmission paths in the system, and predict the evolution trend of complex safety hazards in advance. It provides core technical support for the precise control and risk warning of workshop safety production, and greatly improves the systematicness and foresight of monitoring and analysis. Attached Figure Description

[0017] Figure 1 A flowchart illustrating a method for real-time monitoring and analysis of workshop equipment safety production based on the Internet of Things, provided in an embodiment of the present invention; Figure 2 A functional block diagram of a real-time monitoring and analysis system for safe production of workshop equipment based on the Internet of Things provided in an embodiment of the present invention; Figure 3 A flowchart illustrating the process of extracting a set of equipment characteristic parameters representing the health status of a target device, a set of environmental characteristic parameters representing the environmental hazard level, and a set of personnel behavioral characteristic parameters representing the compliance of personnel operations, as provided in an embodiment of the present invention. The realization of the objective, functional features and advantages of the present invention will be further explained in conjunction with the embodiments and with reference to the accompanying drawings. Detailed Implementation

[0018] It should be understood that the specific embodiments described herein are merely illustrative of the invention and are not intended to limit the invention.

[0019] This application provides a method for real-time monitoring and analysis of workshop equipment safety production based on the Internet of Things (IoT). The executing entity of this IoT-based method includes, but is not limited to, at least one of the following electronic devices that can be configured to execute the method provided in this application: a server, a terminal, etc. In other words, the IoT-based method for real-time monitoring and analysis of workshop equipment safety production can be executed by software or hardware installed on a terminal device or a server device. The server includes, but is not limited to, a single server, a server cluster, a cloud server, or a cloud server cluster. The server can be an independent server or a cloud server that provides basic cloud computing services such as cloud services, cloud databases, cloud computing, cloud functions, cloud storage, network services, cloud communication, middleware services, domain name services, security services, content delivery networks (CDN), and big data and artificial intelligence platforms.

[0020] Reference Figure 1 The diagram shown is a flowchart illustrating a real-time monitoring and analysis method for safe production of workshop equipment based on the Internet of Things (IoT) according to an embodiment of the present invention. In this embodiment, the real-time monitoring and analysis method for safe production of workshop equipment based on the IoT includes: S1, acquire multi-source sensor data of target equipment, environment and personnel in the workshop in real time, perform spatiotemporal synchronization and standardization preprocessing on the multi-source sensor data, and generate a standardized multi-dimensional safety data stream of workshop equipment; In this embodiment of the invention, the step of performing spatiotemporal synchronization and standardization preprocessing on the multi-source sensor data to generate a standardized multidimensional secure data stream includes: The data streams of each dimension in the multi-source sensor data are timestamped to generate a unified time series. Based on a predefined workshop spatial coordinate system, spatial location fusion is performed on each source data stream in the unified time series to generate a spatiotemporal data cube. The original data of each dimension in the spatiotemporal data cube are normalized to generate a standardized multidimensional secure data stream.

[0021] It should be noted that the target equipment sensor data is used to monitor the equipment's operating status, including but not limited to the three-dimensional acceleration signal collected by the vibration sensor, the temperature of the equipment surface or key components collected by the temperature sensor, and the motor drive current collected by the current sensor. Environmental sensor data is used to monitor workshop environmental conditions, including but not limited to ambient temperature and relative humidity collected by temperature and humidity sensors, concentration monitoring data of specific gases such as carbon monoxide and hydrogen sulfide collected by harmful gas sensors, and smoke particle concentration monitored by smoke sensors. Personnel sensor data is used to monitor personnel location and behavior, including but not limited to real-time three-dimensional coordinates of personnel collected by ultra-wideband positioning modules, personnel identity, activity status and some physiological parameters collected by smart badges and wearable devices.

[0022] It should be noted that timestamp alignment processing refers to unifying the raw time-series data from different IoT sensing nodes, each with its own local timestamp, into the same time coordinate system. Specifically, this includes parsing, calibrating, and sorting the timestamps of each data stream, and interpolating the average value of the data from two consecutive time points to address time misalignment issues caused by different sampling frequencies.

[0023] It should be noted that spatial location fusion is based on a predefined workshop 3D coordinate system, which associates the data values ​​of various time-aligned sensor data streams with specific locations in the spatial grid according to their physical deployment locations.

[0024] It should be noted that the spatiotemporal data cube is a three-dimensional data structure, with its three dimensions being time, spatial location, and sensor type, respectively.

[0025] It should be noted that normalization is a process of scaling data from each dimension to a uniform dimensionless numerical range to eliminate the influence of differences in physical dimensions and magnitudes on subsequent analysis. First, the original data value is calculated by subtracting the minimum value of the data in that dimension; then, the difference between the maximum and minimum values ​​of the data in that dimension is calculated; finally, the first calculation result is divided by the second calculation result to obtain the normalized value.

[0026] S2, based on the multidimensional security data stream, extract in parallel the set of equipment feature parameters representing the health status of the target equipment, the set of environmental feature parameters representing the environmental hazard level, and the set of personnel behavior feature parameters representing the compliance of personnel operations; In this embodiment of the invention, the parallel extraction of a set of equipment feature parameters representing the health status of the target equipment, a set of environmental feature parameters representing the environmental hazard level, and a set of personnel behavior feature parameters representing the compliance of personnel operations, based on the multidimensional security data stream, includes: The time-domain features, frequency-domain features, and state change features of the multidimensional security data stream are extracted in parallel to generate a set of device feature parameters representing the health status of the target device; The multidimensional safety data stream is subjected to composite environmental index calculation and risk accumulation assessment to generate a set of environmental characteristic parameters representing the environmental hazard level; The multidimensional safety data stream is subjected to danger zone proximity analysis and operational compliance comparison to generate a set of personnel behavior feature parameters representing personnel operational compliance.

[0027] It should be noted that extracting time-domain features involves directly performing statistical calculations on the device vibration signals in the multidimensional security data stream within a time window to obtain indicators including root mean square value and peak value.

[0028] Furthermore, the root mean square value is calculated based on the vibration acceleration time series, and its mathematical expression is:

[0029] In the formula, RMS is the root mean square value, N is the total number of sampling points within the time window, and x i Let be the vibration acceleration value at the i-th sampling point, where i is the sampling index.

[0030] It should be noted that extracting frequency domain features is the process of calculating the ratio of energy within a preset characteristic frequency band to the total energy in the vibration spectrum of the equipment after it has undergone Fast Fourier Transform (FFT) processing. This step first applies FFT to the time-domain vibration signal to obtain the spectrum; then, it locates the frequency bands corresponding to the fault characteristic frequencies of the key components of the equipment; finally, it calculates the ratio of the sum of squares of the amplitudes within these characteristic frequency bands to the total energy of the entire frequency band. An increase in this ratio indicates an increase in the proportion of fault energy in the corresponding component, which is an effective basis for diagnosing specific fault types.

[0031] It should be noted that the extraction of state change features is based on the temperature and current signals in the multidimensional security data stream, and the calculation of instantaneous values, first-order differences, and total harmonic distortion (THD) is performed. The mathematical expression for THD is as follows:

[0032] In the formula, THD is the total harmonic distortion rate, and I h I1 is the effective value of the h-th harmonic current, H is the highest harmonic number under consideration, and h is the harmonic number index.

[0033] It should be noted that the device feature parameter set is a multi-dimensional vector, whose elements are composed of the time-domain features, frequency-domain features, and state change features extracted in parallel as described above.

[0034] It should be noted that the calculation of the composite environmental index involves comprehensively evaluating the temperature and humidity readings from the multidimensional security data stream using the temperature and humidity index formula, resulting in the temperature and humidity index, the mathematical expression of which is as follows:

[0035] In the formula, THI (temperature and humidity index) represents the ambient temperature in Celsius, and RH represents the relative humidity. The higher the value of the temperature and humidity index, the greater the risk of heat stress caused by the high temperature and high humidity environment.

[0036] It should be noted that the risk accumulation assessment involves calculating the moving average and cumulative exposure amount of the hazardous gas concentration data in the multidimensional safety data stream over a preset time window. The moving average reflects the recent average level of hazardous gas concentration and is used to determine the short-term risk level. The cumulative exposure amount is obtained by integrating the gas concentration over time and is used to assess the total exposure risk of personnel during the current shift or time period, where the preset time window is 15 minutes.

[0037] Furthermore, the mathematical expression used to calculate cumulative exposure is as follows:

[0038] In the formula, Gas_integral represents the cumulative exposure amount. The start time, For the current time, Let be the gas concentration at time t.

[0039] It should be noted that the environmental characteristic parameter set is a vector consisting of the aforementioned composite environmental index and risk accumulation assessment results.

[0040] It should be noted that the approach analysis of hazardous areas is based on the personnel's UWB positioning coordinates in the multidimensional safety data stream and the predefined workshop hazardous area spatial model to calculate the real-time shortest Euclidean distance between the personnel and the boundaries of all hazardous areas. This distance is the core indicator for measuring the physical location risk of personnel. The shorter the distance, the higher the risk of accidents such as mechanical injury and high temperature burns.

[0041] Furthermore, the formula for calculating the shortest Euclidean distance is as follows:

[0042] In the formula, D k For real-time danger distance, (x) p ,y p ,z p (x) represents the coordinates of the personnel. k0 ,y k0 ,z k0 ) represents the coordinates of the point closest to the personnel on the boundary of the danger zone k.

[0043] It should be noted that the operation compliance comparison is based on the frequency of personnel operation events identified in the multidimensional safety data stream, such as starting or stopping equipment, and compared with the standard operation frequency specified in the pre-stored standard operating procedures. The operation frequency anomaly is calculated. This indicator quantifies the degree of deviation between the actual operation rhythm of personnel and the safety specification requirements, and is a key parameter for assessing the compliance of personnel behavior and the stability of operations.

[0044] Furthermore, the expression for calculating the anomaly degree of operation frequency is as follows:

[0045] In the formula, Freq_abn represents the anomaly degree of the operation frequency, f actual f represents the actual frequency of operations observed within the statistical time window. sop The standard operating frequency specified in the standard operating procedure; The higher the value of the abnormality of the operation frequency, the more hurried or slow the operation is, the greater the degree of deviation from safety regulations, and the greater the potential risk of operational errors.

[0046] It should be noted that the personnel behavior characteristic parameter set is a vector consisting of real-time danger distance and operation frequency anomaly degree, used to determine whether personnel are in dangerous positions and whether their operations comply with safe production procedures.

[0047] S3. Based on the set of equipment characteristic parameters, the set of environmental characteristic parameters, and the set of personnel behavior characteristic parameters, analyze the real-time mutual influence relationship among the target equipment, the environment, and the personnel, and generate a system coupling relationship network. In this embodiment of the invention, the step of analyzing the real-time mutual influence relationship among the target equipment, the environment, and the personnel based on the equipment feature parameter set, the environmental feature parameter set, and the personnel behavior feature parameter set, and generating a system coupling relationship network, includes: Extract multidimensional time series data of the equipment feature parameter set, the environmental feature parameter set, and the personnel behavior feature parameter set within a preset time window; Analyze the pairwise correlation strength between key parameters in the equipment feature parameter set and key parameters in the environmental feature parameter set, and simultaneously analyze the pairwise correlation strength between key parameters in the equipment feature parameter set and key parameters in the personnel behavior feature parameter set to generate an initial correlation strength matrix; Based on the set of significant relationships, a weighted undirected graph is constructed with key security parameters as nodes and the filtered relationship strength values ​​as edge weights, generating a system coupling relationship network.

[0048] It should be noted that the sliding time window extraction of multidimensional time series data involves setting a fixed-length historical time window, which defaults to the past 10 minutes, and extracting the time series of all parameters of the equipment feature parameter set, environmental feature parameter set, and personnel behavior feature parameter set within this window from the time series database.

[0049] It should be noted that the initial correlation strength matrix is ​​a symmetric matrix, whose rows and columns correspond to all the key safety parameters involved in the calculation. The elements in the matrix are the mutual information values ​​between the parameters, which describe the strength of the pairwise mutual influence between the elements in the workshop safety system within the sliding time window.

[0050] It should be noted that, in order to remove weak associations below the association strength threshold, all mutual information values ​​in the initial association strength matrix are first sorted in descending order, and an association strength threshold is set. The value is taken as the 75th percentile of the distribution of all mutual information values ​​in ascending order. Mutual information values ​​below this threshold are then removed from the matrix.

[0051] It should be noted that the process of constructing a weighted undirected graph is as follows: each key security parameter is treated as a network node. For each pair of significantly related parameters that are retained after filtering, an undirected edge is established between their corresponding nodes, and the weight of the edge is set to the mutual information value between them. The resulting graph describes the interaction topology between the elements in the security system at the current moment.

[0052] In this embodiment of the invention, the step of analyzing the pairwise correlation strength between key parameters in the equipment feature parameter set and key parameters in the environmental feature parameter set, and simultaneously analyzing the pairwise correlation strength between key parameters in the equipment feature parameter set and key parameters in the personnel behavior feature parameter set, to generate an initial correlation strength matrix, includes: The mathematical expressions used to analyze the pairwise correlation strength are as follows: ; In the formula, MI(X;Y) represents the mutual information value between random variables X and Y, X and Y represent the discretized time series of the two key security parameters whose correlation strength is to be calculated within the sliding window, x and y are the specific values ​​of variables X and Y in their discretized value range, p(x) and p(y) are the marginal probability distributions of variables X and Y, and p(x,y) is the joint probability distribution of variables X and Y.

[0053] It should be noted that the mutual information algorithm used to calculate the pairwise correlation strength is a nonlinear correlation measurement method based on information theory. It is used to quantify the degree of interdependence between two random variables. The larger the value, the more information one variable contains about the other variable. The marginal probability distributions of variables X and Y are obtained by statistically estimating the frequency of each value of X and Y within the sliding window. The joint probability distribution of variables X and Y is obtained by statistically estimating the frequency of joint events where (X,Y) takes the value (x,y) within the sliding window.

[0054] S4. Based on the historical sequence of the equipment feature parameter set, predict the equipment failure risk to obtain a basic risk prediction value. Based on the system coupling relationship network, dynamically correct the basic risk prediction value to generate a corrected equipment dynamic risk index. In this embodiment of the invention, the step of predicting equipment failure risk based on the historical sequence of the equipment feature parameter set to obtain a basic risk prediction value includes: The standardized time series data of the device feature parameter set over a preset prediction time span are extracted from the time series database to generate a historical sequence of device features; The device feature history sequence is input into a pre-trained long short-term memory network model for feature learning, resulting in a hidden state vector sequence corresponding to each historical time step. Based on the attention mechanism, the attention weight of the current prediction time to each historical time step in the hidden state vector sequence is calculated; Based on the attention weights, the hidden state vector sequence is weighted and summed to obtain a context vector that incorporates key historical information. This context vector is then mapped through a fully connected layer to generate the basic risk prediction value.

[0055] It should be noted that, to extract the historical sequence of device features, a preset prediction time span L is first set, with the default value being the past 100 sampling time points. From the standardized time series data of device feature parameters, all feature parameter values ​​from the current time t back to time t−L+1 are extracted to form a matrix of dimension L×DL, where D is the number of device feature parameters.

[0056] It should be noted that using a pre-trained Long Short-Term Memory (LSTM) network model for temporal encoding leverages the ability of the LSTM model to process sequence data and capture long-term dependencies and dynamic patterns in the historical sequence of device features.

[0057] Furthermore, the pre-trained Long Short-Term Memory network model is described by the following set of formulas:

[0058] In the formula, x τ The device feature vector input is given at time step τ, where τ is the time step and h is the number of steps. τ−1 C is the hidden state vector from the previous time step. τ−1 It is the cell state vector of the previous time step, f τ It is the output vector of the forget gate, i τ It is the input gate output vector, C~ τ It is the candidate cell state vector, C τ It is the cell state vector updated at the current time step, o τ It is the output gate vector, h τW is the hidden state vector at the current time step. f W i W C W o These are the weight matrices for the corresponding gates, b f ,b i ,b C ,b o These are the bias vectors for the corresponding gates, ⊙ represents element-wise vector multiplication, σ is the mapping function used to map values ​​to the interval [0,1], and tanh is the hyperbolic tangent function; Among them, W f W i W C W o These are inherent parameters of the Long Short-Term Memory (LSTM) network model, and their specific values ​​are determined during the model building and training phases. First, an LSTM model with a pre-defined network structure is constructed. Then, a training dataset containing a sequence of historical device feature parameters and corresponding labels indicating whether a fault occurred within a given time period is prepared. Finally, using this dataset, with the aforementioned sequence as input and the fault labels as the supervision target, the model is iteratively trained. After training, the final parameters obtained by model convergence are solidified as the weight matrix W. f W i W C W o W r and b r This also stems from the model training process.

[0059] It should be noted that the hidden state vector sequence is a representation of the input history sequence after deep feature extraction and compression by the pre-trained long short-term memory network model. Each vector encodes the device state history information up to that time step.

[0060] It should be noted that the calculation of attention weights is based on the attention mechanism, which evaluates the importance of the hidden state at each time step in the historical sequence for risk prediction at the current moment. The core idea is that not all historical moments contribute equally to the current prediction, and the evolution of equipment failure is often strongly correlated with the state at certain key time points.

[0061] Furthermore, the calculation process for the attention weights is as follows: First, calculate the relevance score between the hidden state at each historical time step and a learnable context vector:

[0062] In the formula, e τ For the relevance score, u is a trainable parameter vector, Wa is the weight matrix, ba is the bias vector, and tanh is the hyperbolic tangent function. The higher the correlation score, the more relevant the state at that historical moment is to the current prediction task; Subsequently, the relevance scores of all time steps are normalized into a probability distribution to obtain the attention weights:

[0063] In the formula, α τ For attention weights, e τ For the relevance score, e k For the k-th relevance score, the preset prediction time span is given, where k is the index of the time span sampling point.

[0064] It should be noted that the weighted summation to generate the context vector is performed by weighting all the hidden states output by the pre-trained Long Short-Term Memory network model according to the attention weights:

[0065] In the formula, c is the context vector, and α τ For attention weights, h τ It is the hidden state vector at the current time step.

[0066] It should be noted that the mathematical expression used to generate the basic risk prediction value through mapping of the fully connected layer is as follows:

[0067] In the formula, R base (t) is the basic risk forecast value, W r and b r σ represents the weights and bias parameters of the fully connected layer, σ is the mapping function, and c is the context vector.

[0068] It should be noted that the basic risk prediction value is a scalar between 0 and 1, where 0 represents no risk of equipment failure and 1 represents an extremely high risk of equipment failure.

[0069] In this embodiment of the invention, the step of dynamically correcting the basic risk prediction value based on the system coupling relationship network to generate a corrected equipment dynamic risk index includes: Based on the dynamic system coupling relationship network, identify all environmental feature parameters and personnel behavior feature parameters that have significant coupling edges with the key feature parameters of the target device; Calculate the average mutual information between the key feature parameters of the target device and all identified environmental feature parameters, and the average mutual information between the key feature parameters of the target device and all identified personnel behavior feature parameters, to generate the average environmental coupling strength and the average personnel coupling strength. The average coupling strength of the environment and the average coupling strength of the personnel are weighted and fused to generate a total coupling strength factor; The basic risk prediction value and the total coupling strength factor are dynamically corrected to generate the corrected equipment dynamic risk index.

[0070] It should be noted that identifying significantly coupled environmental and personnel parameters involves taking the key characteristic parameters of the target device as the starting node in the current system coupling relationship network, traversing all edges connected to that node in the network, and filtering out connections whose edge weights exceed a preset significance threshold. The other end node corresponding to these connections represents the environmental parameters and personnel behavior parameters that have a significant real-time mutual influence with the device state. The preset significance threshold is the value at the 75th percentile of the ascending order of all mutual information value sequences calculated within the current sliding time window.

[0071] It should be noted that the average mutual information is calculated by taking the arithmetic mean of the mutual information values ​​for each of the selected associations.

[0072] It should be noted that the weighted summation to generate the total coupling strength factor is based on the relative importance of environmental and personnel factors in their potential impact on equipment risk, assigning them different weight coefficients. The formula for the collective weighted summation is as follows:

[0073] In the formula, C factor MI is the total coupling strength factor. avg,env MI represents the average coupling strength in the environment. avg,human β represents the average coupling strength of personnel, and β is the personnel coupling weight coefficient.

[0074] Furthermore, the environmental coupling weight coefficient is implicitly set to 1, which means that the environmental coupling strength is fully included, while the personnel coupling strength is included after being proportionally reduced. The design consideration is that the impact of environmental conditions on equipment is often direct and physical, while the impact of personnel behavior may be more indirect or have greater uncertainty. Therefore, it is adjusted through the personnel coupling weight coefficient. The default value of the personnel coupling weight coefficient is 0, and it is adjusted by the operator as needed.

[0075] It should be noted that the dynamic risk index of equipment is calculated using a dynamic correction formula, which is achieved by amplifying and adjusting the basic risk prediction value using the total coupling strength factor. Its mathematical expression is as follows:

[0076] In the formula, R equip_dynamic (t) represents the equipment dynamic risk index, R base (t) is the basic risk forecast value, Cfactor denoted as the total coupling strength factor, and γ as the overall gain coefficient of the total coupling strength factor, wherein the overall gain coefficient is 0.3.

[0077] S5. A risk assessment is performed by combining the equipment dynamic risk index, the environmental characteristic parameter set, and the personnel behavior characteristic parameter set to generate a comprehensive safety risk level. Based on the comprehensive safety risk level, an early warning and visualization display are provided.

[0078] In this embodiment of the invention, the step of comprehensively assessing the risk based on the equipment dynamic risk index, the environmental characteristic parameter set, and the personnel behavior characteristic parameter set to generate a comprehensive safety risk level includes: The environmental feature parameter set and the personnel behavior feature parameter set are subjected to risk quantification transformation to generate environmental risk quantification value and personnel risk quantification value; By integrating the equipment dynamic risk index, the environmental risk quantification value, and the personnel risk quantification value, a comprehensive safety risk value is generated. Based on a preset risk threshold range, the comprehensive security risk value is mapped to the corresponding discrete level to generate a comprehensive security risk level.

[0079] It should be noted that risk quantification transformation refers to converting environmental characteristics and personnel behavior characteristics into standardized risk scores. Specifically, each characteristic is converted into an initial risk component through a preset mapping function; then, the values ​​of all relevant components are aggregated by weighted averaging; finally, the aggregated results are uniformly scaled to the range of 0 to 1 to obtain the environmental risk quantification value and the personnel risk quantification value, respectively.

[0080] Furthermore, the expression for the quantification value of environmental risk is as follows:

[0081] In the formula, R env (t) represents the quantified environmental risk value, THI(t), C avg S(t) and S(t) represent the temperature and humidity index, moving average concentration of harmful gases, and smoke concentration at time t, respectively. fTHI, f gas and f smoke For the preset mapping function, the corresponding parameters are converted into risk coefficients in the interval [0,1]. λ1, λ2, and λ3 are the preset weights of the corresponding risk components, each taking a value of one-third. M i This is the theoretical maximum value of the mapping function, and its value is 1.

[0082] The expression for the quantitative value of personnel risk is as follows:

[0083] In the formula, Rhuman (t) represents the quantitative value of personnel risk, D risk (t) and Freq_abn(t) are the real-time danger distance and operation frequency anomaly degree at time t, respectively, and g dist and g freq The preset mapping function converts the corresponding parameters into risk coefficients in the range [0,1]. μ1 and μ2 are the preset weights of the corresponding risk components, each taking a value of 1 / 2. N j This is the theoretical maximum value of the mapping function, and its value is 1.

[0084] It should be noted that the integrated security risk value is generated by combining risk indicators from three different dimensions into a single global risk indicator, the mathematical expression of which is as follows:

[0085] In the formula, R total (t) represents the comprehensive safety risk value, R equip_dynamic (t) represents the equipment dynamic risk index, R env (t) and R human (t) represent the quantified values ​​of environmental risk and personnel risk, respectively. quip For equipment risk weights, w env For environmental risk weights, w human The risk weights are for personnel, with the risk weights for equipment, environment, and personnel set to 0.4, 0.3, and 0.3, respectively.

[0086] It should be noted that the risk threshold range mapping level is based on predefined ordered, non-overlapping risk value ranges, with each range assigned a discrete risk level label, R. total If (t)∈[0,0.3), then the overall safety risk level is "low risk"; if R total If (t)∈[0.3,0.7), then the overall safety risk level is "medium risk"; if R total If (t)∈[0.7,1.0], then the overall safety risk level is "high risk".

[0087] It should be noted that the comprehensive safety risk level is a discrete, semantically clear classification label. It transforms continuous comprehensive risk values ​​into qualitative descriptions that are easier for decision-makers to understand and respond to quickly. Different levels correspond to different early warning response strategies and emergency plan activation conditions, providing direct and actionable input for subsequent early warning and visualization.

[0088] In this embodiment of the invention, the early warning and visualization display based on the comprehensive security risk level includes: Based on a predefined risk level-early warning action mapping table, the comprehensive safety risk level is matched to generate a corresponding multimodal early warning strategy. Based on the aforementioned multimodal early warning strategy, the system executes in parallel the audible and visual alarm drive, the monitoring interface message push, and the linkage control command issuance with the workshop control system. Based on the equipment dynamic risk index, the environmental feature parameter set, the personnel behavior feature parameter set, and the comprehensive safety risk level, visual information is integrated and rendered to generate a workshop safety situation dashboard view. The workshop safety status dashboard view is pushed to a preset monitoring terminal for display.

[0089] It should be noted that the multimodal early warning strategy is a set of differentiated response actions pre-configured according to the comprehensive safety risk level. It transforms the abstract risk level into a specific composite instruction that can simultaneously affect personnel's senses and the workshop control system. For example, the strategy corresponding to "high risk" may include triggering a red audible and visual alarm, displaying a full-screen pop-up alarm on the monitoring interface, and sending a speed reduction or shutdown interlock instruction to the target equipment controller.

[0090] It should be noted that the parallel execution of early warning operations aims to achieve immediate and comprehensive risk notification and intervention. The audible and visual alarm drive controls the on-site alarm device through a hardware interface, providing intuitive audio-visual warnings. The monitoring interface message push updates and highlights risk details in real time on the software operation interface. The linkage with the workshop control system is to achieve automated safety protection, automatically executing predefined control logic according to the strategy, forming a rapid closed loop from perception to execution.

[0091] It should be noted that the workshop safety situation dashboard view is an integrated visualization interface that comprehensively displays key information such as equipment dynamic risk index, key environmental parameters, personnel behavior parameters, and comprehensive risk level through graphical elements such as digital twin maps, dashboards, and trend curves. It provides safety management personnel with a global, real-time, and intuitive situational awareness view, making it easy to quickly locate risk sources and understand the overall safety status.

[0092] It should be noted that pushing the view to the monitoring terminal ensures that critical security information can be accessed in real time by the remote monitoring center or mobile inspection terminal, thereby supporting remote monitoring and decision-making.

[0093] like Figure 2 The diagram shown is a functional block diagram of a real-time monitoring and analysis system for safe production of workshop equipment based on the Internet of Things provided in an embodiment of the present invention.

[0094] The IoT-based real-time monitoring and analysis system 100 for safe production of workshop equipment described in this invention can be installed in an electronic device. Depending on the functions implemented, the IoT-based real-time monitoring and analysis system 100 may include a data synchronization module 101, a multi-dimensional safety parameter extraction and analysis module 102, a coupling strength analysis module 103, a dynamic risk prediction and correction module 104, and a comprehensive evaluation and intelligent early warning module 105. The module described in this invention can also be called a unit, which refers to a series of computer program segments that can be executed by the processor of an electronic device and can perform a fixed function, stored in the memory of the electronic device.

[0095] In this embodiment, the functions of each module / unit are as follows: The data synchronization module is used to acquire multi-source sensor data of target equipment, environment and personnel in the workshop in real time, perform spatiotemporal synchronization and standardization preprocessing on the multi-source sensor data, and generate a standardized multi-dimensional safety data stream of workshop equipment. The multidimensional safety parameter extraction and analysis module is used to extract, in parallel, a set of equipment characteristic parameters representing the health status of the target equipment, a set of environmental characteristic parameters representing the environmental hazard level, and a set of personnel behavior characteristic parameters representing the compliance of personnel operations, based on the multidimensional safety data stream. The coupling strength analysis module is used to analyze the real-time mutual influence relationship between the target equipment, the environment and the personnel based on the equipment feature parameter set, the environmental feature parameter set and the personnel behavior feature parameter set, and generate a system coupling relationship network. The dynamic risk prediction and correction module is used to predict equipment failure risk based on the historical sequence of the equipment feature parameter set, obtain a basic risk prediction value, and dynamically correct the basic risk prediction value based on the system coupling relationship network to generate a corrected equipment dynamic risk index. The comprehensive assessment and intelligent early warning module is used to conduct risk assessment by comprehensively considering the equipment dynamic risk index, the environmental characteristic parameter set, and the personnel behavior characteristic parameter set, generate a comprehensive safety risk level, and provide early warning and visualization based on the comprehensive safety risk level.

[0096] In the several embodiments provided by this invention, it should be understood that the disclosed methods and systems can be implemented in other ways. For example, the system embodiments described above are merely illustrative; for instance, the division of modules is only a logical functional division, and other division methods may be used in actual implementation.

[0097] The modules described as separate components may or may not be physically separate. The components shown as modules may or may not be physical units; that is, they may be located in one place or distributed across multiple network units. Some or all of the modules can be selected to achieve the purpose of this embodiment according to actual needs.

[0098] Furthermore, the functional modules in the various embodiments of the present invention 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 in the form of hardware plus software functional modules.

[0099] It will be apparent to those skilled in the art that the present invention is not limited to the details of the exemplary embodiments described above, and that the present invention can be implemented in other specific forms without departing from the spirit or essential characteristics of the present invention.

[0100] The embodiments of this application can acquire and process relevant data based on artificial intelligence technology. Artificial intelligence is the theory, method, technology, and application system that uses digital computers or machines controlled by digital computers to simulate, extend, and expand human intelligence, perceive the environment, acquire knowledge, and use that knowledge to obtain optimal results.

[0101] Finally, it should be noted that 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 preferred embodiments, those skilled in the art should understand that modifications or equivalent substitutions can be made to the technical solutions of the present invention without departing from the spirit and scope of the technical solutions of the present invention.

Claims

1. A method for real-time monitoring and analysis of workshop equipment safety production based on the Internet of Things, characterized in that, The method includes: S1, acquire multi-source sensor data of target equipment, environment and personnel in the workshop in real time, perform spatiotemporal synchronization and standardization preprocessing on the multi-source sensor data, and generate a standardized multi-dimensional safety data stream of workshop equipment; S2, based on the multidimensional security data stream, extract in parallel the set of equipment feature parameters representing the health status of the target equipment, the set of environmental feature parameters representing the environmental hazard level, and the set of personnel behavior feature parameters representing the compliance of personnel operations; S3. Based on the set of equipment characteristic parameters, the set of environmental characteristic parameters, and the set of personnel behavior characteristic parameters, analyze the real-time mutual influence relationship among the target equipment, the environment, and the personnel, and generate a system coupling relationship network. S4. Based on the historical sequence of the equipment feature parameter set, predict the equipment failure risk to obtain a basic risk prediction value. Based on the system coupling relationship network, dynamically correct the basic risk prediction value to generate a corrected equipment dynamic risk index. S5. A risk assessment is performed by combining the equipment dynamic risk index, the environmental characteristic parameter set, and the personnel behavior characteristic parameter set to generate a comprehensive safety risk level. Based on the comprehensive safety risk level, an early warning and visualization display are provided.

2. The method for real-time monitoring and analysis of workshop equipment safety production based on the Internet of Things as described in claim 1, characterized in that, The step of performing spatiotemporal synchronization and standardization preprocessing on the multi-source sensor data to generate a standardized multidimensional secure data stream includes: The data streams of each dimension in the multi-source sensor data are timestamped to generate a unified time series. Based on a predefined workshop spatial coordinate system, spatial location fusion is performed on each source data stream in the unified time series to generate a spatiotemporal data cube. The original data of each dimension in the spatiotemporal data cube are normalized to generate a standardized multidimensional secure data stream.

3. The method for real-time monitoring and analysis of workshop equipment safety production based on the Internet of Things as described in claim 1, characterized in that, The parallel extraction of equipment feature parameter sets representing the health status of the target equipment, environmental feature parameter sets representing the environmental hazard level, and personnel behavior feature parameter sets representing the compliance of personnel operations, based on the multidimensional security data stream, includes: The time-domain features, frequency-domain features, and state change features of the multidimensional security data stream are extracted in parallel to generate a set of device feature parameters representing the health status of the target device; The multidimensional safety data stream is subjected to composite environmental index calculation and risk accumulation assessment to generate a set of environmental characteristic parameters representing the environmental hazard level; The multidimensional safety data stream is subjected to danger zone proximity analysis and operational compliance comparison to generate a set of personnel behavior feature parameters representing personnel operational compliance.

4. The method for real-time monitoring and analysis of workshop equipment safety production based on the Internet of Things as described in claim 1, characterized in that, The process of analyzing the real-time mutual influence relationships among the target equipment, environment, and personnel based on the equipment feature parameter set, the environmental feature parameter set, and the personnel behavior feature parameter set, and generating a system coupling relationship network, includes: Extract multidimensional time series data of the equipment feature parameter set, the environmental feature parameter set, and the personnel behavior feature parameter set within a preset time window; Analyze the pairwise correlation strength between key parameters in the equipment feature parameter set and key parameters in the environmental feature parameter set, and simultaneously analyze the pairwise correlation strength between key parameters in the equipment feature parameter set and key parameters in the personnel behavior feature parameter set to generate an initial correlation strength matrix; Based on the set of significant relationships, a weighted undirected graph is constructed with key security parameters as nodes and the filtered relationship strength values ​​as edge weights, generating a system coupling relationship network.

5. The method for real-time monitoring and analysis of workshop equipment safety production based on the Internet of Things as described in claim 4, characterized in that, The analysis examines the pairwise correlation strength between key parameters in the equipment feature parameter set and key parameters in the environmental feature parameter set, and simultaneously analyzes the pairwise correlation strength between key parameters in the equipment feature parameter set and key parameters in the personnel behavior feature parameter set, generating an initial correlation strength matrix, including: The mathematical expressions used to analyze the pairwise correlation strength are as follows: ; In the formula, MI(X;Y) represents the mutual information value between random variables X and Y, X and Y represent the discretized time series of the two key security parameters whose correlation strength is to be calculated within the sliding window, x and y are the specific values ​​of variables X and Y in their discretized value range, p(x) and p(y) are the marginal probability distributions of variables X and Y, and p(x,y) is the joint probability distribution of variables X and Y.

6. The method for real-time monitoring and analysis of workshop equipment safety production based on the Internet of Things as described in claim 1, characterized in that, The process of predicting equipment failure risk based on the historical sequence of the equipment feature parameter set to obtain a basic risk prediction value includes: The standardized time series data of the device feature parameter set over a preset prediction time span are extracted from the time series database to generate a historical sequence of device features; The device feature history sequence is input into a pre-trained long short-term memory network model for feature learning, resulting in a hidden state vector sequence corresponding to each historical time step. Based on the attention mechanism, the attention weight of the current prediction time to each historical time step in the hidden state vector sequence is calculated; Based on the attention weights, the hidden state vector sequence is weighted and summed to obtain a context vector that incorporates key historical information. This context vector is then mapped through a fully connected layer to generate the basic risk prediction value.

7. The method for real-time monitoring and analysis of workshop equipment safety production based on the Internet of Things as described in claim 1, characterized in that, The step of dynamically correcting the basic risk prediction value based on the system coupling relationship network to generate a corrected equipment dynamic risk index includes: Based on the dynamic system coupling relationship network, identify all environmental feature parameters and personnel behavior feature parameters that have significant coupling edges with the key feature parameters of the target device; Calculate the average mutual information between the key feature parameters of the target device and all identified environmental feature parameters, and the average mutual information between the key feature parameters of the target device and all identified personnel behavior feature parameters, to generate the average environmental coupling strength and the average personnel coupling strength. The average coupling strength of the environment and the average coupling strength of the personnel are weighted and fused to generate a total coupling strength factor; The basic risk prediction value and the total coupling strength factor are dynamically corrected to generate the corrected equipment dynamic risk index.

8. The method for real-time monitoring and analysis of workshop equipment safety production based on the Internet of Things as described in claim 1, characterized in that, The risk assessment is performed by combining the equipment dynamic risk index, the environmental characteristic parameter set, and the personnel behavior characteristic parameter set to generate a comprehensive safety risk level, including: The environmental feature parameter set and the personnel behavior feature parameter set are subjected to risk quantification transformation to generate environmental risk quantification value and personnel risk quantification value; By integrating the equipment dynamic risk index, the environmental risk quantification value, and the personnel risk quantification value, a comprehensive safety risk value is generated. Based on a preset risk threshold range, the comprehensive security risk value is mapped to the corresponding discrete level to generate a comprehensive security risk level.

9. The method for real-time monitoring and analysis of workshop equipment safety production based on the Internet of Things as described in claim 1, characterized in that, The early warning and visualization display based on the comprehensive security risk level includes: Based on a predefined risk level-early warning action mapping table, the comprehensive safety risk level is matched to generate a corresponding multimodal early warning strategy. Based on the aforementioned multimodal early warning strategy, the system executes in parallel the audible and visual alarm drive, the monitoring interface message push, and the linkage control command issuance with the workshop control system. Based on the equipment dynamic risk index, the environmental feature parameter set, the personnel behavior feature parameter set, and the comprehensive safety risk level, visual information is integrated and rendered to generate a workshop safety situation dashboard view. The workshop safety status dashboard view is pushed to a preset monitoring terminal for display.

10. A real-time monitoring and analysis system for safe production of workshop equipment based on the Internet of Things, characterized in that, The system includes: The data synchronization module is used to acquire multi-source sensor data of target equipment, environment and personnel in the workshop in real time, perform spatiotemporal synchronization and standardization preprocessing on the multi-source sensor data, and generate a standardized multi-dimensional safety data stream of workshop equipment. The multidimensional safety parameter extraction and analysis module is used to extract, in parallel, a set of equipment characteristic parameters representing the health status of the target equipment, a set of environmental characteristic parameters representing the environmental hazard level, and a set of personnel behavior characteristic parameters representing the compliance of personnel operations, based on the multidimensional safety data stream. The coupling strength analysis module is used to analyze the real-time mutual influence relationship between the target equipment, the environment and personnel based on the set of equipment characteristic parameters, the set of environmental characteristic parameters and the set of personnel behavior characteristic parameters, and to generate a system coupling relationship network. The dynamic risk prediction and correction module is used to predict equipment failure risk based on the historical sequence of the equipment feature parameter set, obtain a basic risk prediction value, and dynamically correct the basic risk prediction value based on the system coupling relationship network to generate a corrected equipment dynamic risk index. The comprehensive assessment and intelligent early warning module is used to conduct risk assessment by comprehensively considering the equipment dynamic risk index, the environmental characteristic parameter set, and the personnel behavior characteristic parameter set, generate a comprehensive safety risk level, and provide early warning and visualization based on the comprehensive safety risk level.

Images