Method and system for identifying potential safety hazards of hoisting machinery based on AI

Abstract

The invention relates to the technical field of potential safety hazard identification, in particular to an AI-based method and system for identifying potential safety hazards of hoisting machinery, which comprises the following steps of: collecting operation data of the hoisting machinery, including position, speed and load capacity, integrating meteorological conditions and information of near-field obstacles, and combining real-time transmission data to obtain a real-time transmission result of the hoisting machinery; and generating comprehensive real-time data. According to the method, comprehensive assessment and early warning of potential risks are achieved by integrating the operation data and the environment variables of the hoisting machinery, the operation environment can be simulated more accurately, occurrence of risk conditions can be predicted more accurately, the operation strategy is adjusted in time through dynamic risk scoring, the early warning accuracy and the operation safety are effectively improved, and the safety of the hoisting machinery is improved. The state definition and the probability model which are updated in real time can continuously adapt to environment changes, it is ensured that the operation strategy of the hoisting machinery is always kept in the optimal state, the accident occurrence probability is reduced, and meanwhile economic losses caused by equipment faults or misoperation are reduced.

Description

Technical Field

[0001] The present invention relates to the technical field of potential safety hazards identification, and in particular to a method and system for identifying potential safety hazards of lifting machinery based on AI. Background Art

[0002] The field of safety hazard technology involves safety identification, assessment and prevention strategies in various industrial and construction environments. In the application of lifting machinery, this field pays special attention to how to identify and deal with potential dangerous factors that may lead to accidents or damage. Technological development has led to the increasing application of artificial intelligence (AI) in this field, especially in the automated monitoring and prediction of safety hazards. AI technology can predict and warn of potential risks by analyzing historical accident data and real-time monitoring of equipment performance and environmental conditions, thereby improving work efficiency and safety.

[0003] Among them, the AI-based method for identifying safety hazards of lifting machinery refers to the use of artificial intelligence technology to identify and prevent safety problems that occur during the operation of lifting machinery. This method involves machine learning models and algorithms to analyze data collected from sensors, operation records, and environmental monitoring equipment. Through this technology, the operating status of lifting machinery can be monitored in real time, abnormal behavior or potential failures can be discovered in time, and preventive measures can be taken before accidents occur. It not only helps to ensure the safety of operators, but also effectively reduces economic losses caused by equipment failure or operating errors.

[0004] Existing safety hazard technologies rely on traditional data analysis and post-processing models, which can only react after an accident occurs, but lack preventive risk assessment and real-time monitoring. In a rapidly changing industrial environment, this model cannot effectively respond to sudden environmental changes and equipment status changes, resulting in delayed response time for safety warnings and increased risks of accidents. Insensitivity to changes in environmental factors causes lifting machinery to operate under inappropriate conditions, increasing the probability of safety accidents. The shortcomings of existing technologies in data integration and real-time analysis also affect the timeliness and accuracy of operational decisions, and cannot effectively prevent potential safety risks, resulting in unnecessary economic losses and safety hazards. Summary of the invention

[0005] The purpose of the present invention is to solve the shortcomings existing in the prior art and to propose a method and system for identifying safety hazards of lifting machinery based on AI.

[0006] In order to achieve the above object, the present invention adopts the following technical solution, a method for identifying safety hazards of lifting machinery based on AI, comprising the following steps: S1: Collects the operation data of the lifting machinery, including position, speed and load, and integrates the information of meteorological conditions and near-field obstacles, combined with real-time transmission data to generate comprehensive real-time data; S2: Based on the comprehensive real-time data, analyze potential risk factors, identify the combination of conditions that lead to the accident according to the accident data and current operation mode of the lifting machinery, and generate risk identification results; S3: By comparing the risk identification results with normal operation data, using AI, applying weighted calculation of risk factors, updating and calculating real-time risk scores, accumulating the impact of differentiation factors, and obtaining dynamic risk scores; S4: using the dynamic risk score to evaluate the current crane operation risk, and propose adjustment operation measures for potential safety hazards, and generate a safety adjustment strategy by changing the speed and path and triggering an emergency stop; S5: Implement the safety adjustment strategy, update the state definition of the lifting machinery operation, adjust the transition probability between states, match the real-time operation risk and environmental changes, and generate an updated state definition and probability model; S6: According to the updated state definition and probability model, the crane operation data is monitored in real time, the transition probability between differentiated states is calculated, the effectiveness of the safety adjustment strategy is evaluated, and a safety assessment log is obtained.

[0007] As a further solution of the present invention, the comprehensive real-time data includes the position coordinates of the lifting machinery, the current operating speed value and the real-time load, the associated meteorological conditions and the types of nearby obstacles; the risk identification results include the identified risky operation modes, the combination of abnormal load states and adverse meteorological conditions; the dynamic risk score includes the numerical score calculated according to the weights of differentiated risk factors, the real-time update frequency of the score and the differentiated impact; the safety adjustment strategy includes speed reduction, path change measures and emergency stop trigger conditions; the updated state definition and probability model include new operating states, transition conditions and probability values ​​of state transitions; the safety assessment log includes the recorded execution of the safety adjustment strategy, state transition results and risk assessment update results.

[0008] As a further solution of the present invention, the steps of collecting the operation data of the lifting machinery, including the position, speed and load, integrating the information of the meteorological conditions and the near-field obstacles, and combining the real-time transmission data to generate the comprehensive real-time data are as follows: S101: collect operation data of the lifting machinery, use sensors to monitor and record the current position, speed and load of the machinery, and update in real time to generate an operation data set; S102: Based on the operation data set, using environmental monitoring equipment to obtain meteorological condition data, and deploying near-field obstacle detectors to collect obstacle information, thereby generating an environmental data set; S103: Through the environmental data set, the operating data of the lifting machinery is transmitted in real time, the real-time update of the data is verified and the integrity of the data is checked to generate comprehensive real-time data.

[0009] As a further solution of the present invention, based on the comprehensive real-time data, potential risk factors are analyzed, and according to the accident data and current operation mode of the lifting machinery, the combination of conditions leading to the accident is identified, and the steps of generating the risk identification result are specifically as follows: S201: Based on the comprehensive real-time data, by reviewing and analyzing the accident data of the lifting machinery, extracting the operation mode and environmental conditions where the failure occurred, combining the real-time operation data, identifying and recording the data pattern associated with the accident, and generating a pattern comparison result; S202: classifying the data patterns in the pattern comparison result into safety levels, evaluating the probability of accidents under multiple combinations of conditions, verifying the dangerousness of the condition combinations through simulation tests, and generating risk level evaluation results; S203: According to the risk level assessment result, key risk factors are determined, targeted preventive measures are implemented, and the risk of accidents is avoided by adjusting operating parameters and optimizing safety monitoring, thereby generating risk identification results.

[0010] As a further solution of the present invention, by comparing the risk identification results with normal operation data, using AI, applying weighted calculation of risk factors, updating and calculating real-time risk scores, accumulating the influence of differentiation factors, and obtaining dynamic risk scores, the specific steps are: S301: Based on the risk identification results and in combination with normal operation data, AI technology is used to perform data comparison and analysis, extract key differences between risk factors and normal factors, and generate difference analysis results; S302: Based on the difference analysis result, a weighted average method is used to calculate the risk contribution value of multiple factors, and a weighted calculation is performed on multiple risk factors to generate a risk score update record; S303: Using the risk score to update records, implement risk management strategies, dynamically adjust operating guidelines and safety measures, verify the safety of lifting machinery operations in a changing working environment, and obtain a dynamic risk score.

[0011] As a further solution of the present invention, the formula of the weighted average method is as follows: ; in, is the adjusted risk score, For the The weight coefficient of each risk factor is For the The raw risk score of each risk factor, is the risk frequency adjustment factor, It is the square root of the weighted sum.

[0012] As a further solution of the present invention, the dynamic risk score is used to evaluate the current crane operation risk, and adjustment operation measures are proposed for potential safety hazards. The steps of generating a safety adjustment strategy by changing the speed and path and triggering an emergency stop are as follows: S401: Analyze the current crane operation risk by using the dynamic risk score, identify the operation parameters that exceed the safety threshold, determine the operation mode and environmental conditions that need to be adjusted, and generate a risk limit exceeding identification result; S402: According to the risk over-limit identification result, by changing the speed and path of the mechanical operation, optimizing the operation process to match the real-time environment and operating conditions, and generating operation adjustment measures; S403: Implement the operation adjustment measures, trigger an emergency shutdown according to the severity of the risk, interrupt the risk operation by implementing the emergency response procedure, verify the safety of the operator and equipment, and generate a safety adjustment strategy.

[0013] As a further solution of the present invention, the steps of implementing the safety adjustment strategy, updating the state definition of the lifting machinery operation, adjusting the transition probability between states, matching the real-time operation risk and environmental changes, and generating the updated state definition and probability model are as follows: S501: Implement the safety adjustment strategy, redefine the operating state of the lifting machinery, update the parameters of the operating state to match the current risk score and environmental conditions, and generate a state update result; S502: Based on the state update result, adjust the transition probability between states, optimize the state transition model using real-time data feedback, verify the state transition probability according to real-time operational risks and external environment changes, and generate a probability adjustment result; S503: According to the probability adjustment result, the operation guide and risk management of the lifting machinery are updated, the consistency of the operation definition and probability model with the real-time operation environment and risk score is checked, and an updated state definition and probability model are generated.

[0014] As a further solution of the present invention, according to the updated state definition and probability model, the operation data of the lifting machinery is monitored in real time, the transition probability between the differentiated states is calculated, the effectiveness of the safety adjustment strategy is evaluated, and the steps of obtaining the safety assessment log are specifically as follows: S601: Based on the updated state definition and probability model, real-time monitoring and tracking of operation data of the lifting machinery, including position, speed and load, to generate real-time monitoring data; S602: Calculate the transition probability between differentiated states using the real-time monitoring data, update the risk assessment model to match the current operating state and environmental conditions by analyzing the state changes, and generate a state transition analysis result; S603: Evaluate the effectiveness of the safety adjustment strategy based on the state transition analysis result, and generate a safety assessment log by comparing the change in accident rate and the optimization efficiency and speed of operational safety.

[0015] A system for identifying safety hazards of lifting machinery based on AI, the system for identifying safety hazards of lifting machinery based on AI is used to execute the above-mentioned method for identifying safety hazards of lifting machinery based on AI, and the system comprises: The data acquisition module collects the position, speed and load data of the crane, as well as the presence and location data of near-field obstacles to generate a comprehensive real-time data set; The risk analysis module uses the comprehensive real-time data set to detect risk factors in the operation of the lifting machinery, refers to the accident data and operation mode, identifies the conditions for the occurrence of accidents, and outputs risk identification results; The weighted calculation module evaluates the risk identification result, performs weighted calculation on each risk factor, identifies the instant safety level of the lifting machinery operation, and obtains a dynamic risk score; The strategy building module evaluates the safety risk of the current operation according to the dynamic risk score, adjusts the operation parameters, including speed and path, triggers an emergency shutdown when necessary, and obtains a safety adjustment strategy; The state update module implements the safety adjustment strategy, updates the operating state definition of the lifting machinery, adjusts the transition probability between states, matches real-time operating risks and environmental changes, evaluates the effectiveness of the safety adjustment strategy, and generates a safety assessment log.

[0016] Compared with the prior art, the advantages and positive effects of the present invention are: In the present invention, by integrating the operation data of the crane and the environmental variables, a comprehensive assessment and early warning of potential risks are achieved. Data including position, speed and load are collected, and combined with meteorological and obstacle information, the operation environment can be simulated more accurately and the occurrence of risk conditions can be predicted. The operation strategy can be adjusted in time through dynamic risk scoring, which effectively improves the accuracy of early warning and the safety of operation. The state definition and probability model updated in real time can continuously adapt to environmental changes, ensure that the operation strategy of the crane always maintains the optimal state, reduce the probability of accidents, and also reduce the economic losses caused by equipment failure or operation errors. BRIEF DESCRIPTION OF THE DRAWINGS

[0017] Figure 1 It is a schematic diagram of the workflow of the present invention; Figure 2This is a detailed flow chart of S1 of the present invention; Figure 3 This is a detailed flow chart of S2 of the present invention; Figure 4 This is a detailed flow chart of S3 of the present invention; Figure 5 This is a detailed flow chart of S4 of the present invention; Figure 6 This is a detailed flow chart of S5 of the present invention; Figure 7 This is a detailed flow chart of S6 of the present invention; Figure 8 It is a system flow chart of the present invention. DETAILED DESCRIPTION

[0018] In order to make the purpose, technical solution and advantages of the present invention more clearly understood, the present invention is further described in detail below in conjunction with the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are only used to explain the present invention and are not intended to limit the present invention.

[0019] In the description of the present invention, it should be understood that the terms "length", "width", "up", "down", "front", "back", "left", "right", "vertical", "horizontal", "top", "bottom", "inside", "outside" and the like indicate positions or positional relationships based on the positions or positional relationships shown in the drawings, and are only for the convenience of describing the present invention and simplifying the description, rather than indicating or implying that the device or element referred to must have a specific orientation, be constructed and operated in a specific orientation, and therefore cannot be understood as limiting the present invention. In addition, in the description of the present invention, "multiple" means two or more, unless otherwise clearly and specifically defined.

[0020] See also Figure 1 The present invention provides a technical solution, a method for identifying safety hazards of lifting machinery based on AI, comprising the following steps: S1: Collects the operation data of the lifting machinery, including position, speed and load, and integrates the information of meteorological conditions and near-field obstacles, combined with real-time transmission data to generate comprehensive real-time data; S2: Based on comprehensive real-time data, analyze potential risk factors, identify the combination of conditions that lead to accidents based on the accident data and current operation mode of the lifting machinery, and generate risk identification results; S3: By comparing risk identification results and normal operation data, using AI and weighted calculation of risk factors, we update and calculate real-time risk scores, and obtain dynamic risk scores based on the cumulative impact of differentiation factors; S4: Use dynamic risk scoring to assess the current risk of crane operation, propose operational adjustment measures for potential safety hazards, and generate safety adjustment strategies by changing speed and path and triggering emergency shutdown when necessary; S5: Implement safety adjustment strategies, update the state definition of lifting machinery operation, adjust the transition probability between states, match real-time operation risks and environmental changes, and generate updated state definitions and probability models; S6: Based on the updated state definition and probability model, the crane operation data is monitored in real time, the transition probability between differentiated states is calculated, the effectiveness of the safety adjustment strategy is evaluated, the continuous safety supervision is verified, and the safety assessment log is obtained.

[0021] Comprehensive real-time data includes the location coordinates of the lifting machinery, the current operating speed value and real-time load capacity, the associated meteorological conditions and the types of nearby obstacles. The risk identification results include the combination of identified risky operating modes, abnormal load conditions and adverse meteorological conditions. The dynamic risk score includes the numerical score calculated based on the weights of differentiated risk factors, the real-time update frequency of the score and the differentiated impact. The safety adjustment strategy includes speed reduction, path change measures and emergency stop trigger conditions. The updated state definition and probability model include new operating states, transfer conditions and probability values ​​of state transfer. The safety assessment log includes the recorded execution of the safety adjustment strategy, state transfer results and risk assessment update results.

[0022] See also Figure 2 , collects the operation data of the lifting machinery, including position, speed and load, integrates the information of meteorological conditions and near-field obstacles, and combines the real-time transmission data to generate comprehensive real-time data. The specific steps are as follows: S101: Collect the operation data of the lifting machinery, use sensors to monitor and record the current position, speed and load of the machinery, and update it in real time. The execution process of generating the operation data set is as follows; Collecting the operational data of lifting machinery involves using sensors to monitor the current position, speed and load of the machinery. The sensors obtain a continuous data stream through physical detection. These data are converted into digital signals through analog converters. The data acquisition system updates the data set in real time to ensure that the state of the machinery at each moment can be accurately recorded. The data processor in the system will perform preliminary processing and sorting on the received data to remove any errors or outliers, ensuring the reliability and real-time nature of the data. The data is organized and stored through an efficient data management system so that the data can be quickly retrieved when needed. The integrity and real-time nature of the data set provide support for the efficient operation of the lifting machinery. The collected data can be used for subsequent analysis and monitoring to optimize machinery performance and generate operational data sets.

[0023] S102: Based on the operation data set, the environmental monitoring equipment is used to obtain meteorological condition data, and near-field obstacle detectors are deployed to collect obstacle information. The execution process of generating the environmental data set is as follows; Based on the operational data set, meteorological condition data and near-field obstacle information are obtained, and the formula , calculate the comprehensive impact of the environmental data set. Where D represents the obstacle distance, Represents the time from sensor to obstacle, represents wind speed, Represents the area affected by wind direction on crane operation. Detailed explanation of the formula and formula calculation process: Considering the actual impact of meteorological conditions and obstacle information on crane operation, the overall impact of environmental conditions can be evaluated by monitoring obstacle distance and wind speed. At a certain moment, the sensor measures the obstacle distance as 100 meters, the relative time with the obstacle is 5 seconds, and the meteorological monitoring equipment records the wind speed as 15m / s, and the wind direction is facing the crane operation area of ​​50 square meters. According to the formula , this speed indicates the operational risk and complexity of mechanical operation under the current environmental conditions.

[0024] S103: The execution process of transmitting the operation data of the lifting machinery in real time through the environmental data set, verifying the real-time update of the data and checking the integrity of the data, and generating comprehensive real-time data is as follows; The operating data of the lifting machinery is transmitted in real time through the environmental data set. During the real-time transmission process, the data is sent to the control center through the wireless network. The data processor of the center verifies each received data packet to ensure the integrity of the data. High-frequency data transmission technology, such as LTE or 5G networks, is used to ensure low latency and high reliability of data transmission. Every step of data transmission is strictly monitored, and data encryption is implemented to ensure information security. The operating data is updated in real time and the integrity is verified, including data redundancy verification and error detection mechanisms. These mechanisms can detect and correct transmission errors in real time, maintain the integrity and accuracy of the data transmission process, and use real-time transmitted data for instant adjustment and predictive maintenance of the machinery, which is essential to maintaining the optimal operating state of the machinery and generating comprehensive real-time data.

[0025] See also Figure 3 , based on comprehensive real-time data, analyze potential risk factors, identify the combination of conditions that lead to accidents based on the accident data and current operation mode of the lifting machinery, and generate risk identification results in the following steps: S201: Based on the comprehensive real-time data, by reviewing and analyzing the accident data of the lifting machinery, extracting the operation mode and environmental conditions of the failure, combining the real-time operation data, identifying and recording the data pattern associated with the accident, and generating the execution process of the pattern comparison result is as follows; By analyzing the accident data of lifting machinery, extracting the operation mode and environmental conditions, according to the formula , calculate the probability of occurrence of the data pattern associated with the accident. represents the probability of an accident occurring, represents the number of events that failed under certain conditions, Represents the total number of operations. Detailed explanation of the formula and the process of formula calculation: Conduct in-depth analysis of the operation data of the lifting machinery, and extract the operation mode and specific environmental conditions of the failure from the historical accident data. Suppose that in the past 1000 operations, 50 operations have had accidents, and the formula can be used to calculate the number of accidents. , which is 5%. This probability indicates the expected risk level of future operations under similar operating modes and environmental conditions.

[0026] S202: classify the data patterns in the pattern comparison results into safety levels, evaluate the probability of accidents under multiple combinations of conditions, verify the dangerousness of the condition combinations through simulation tests, and generate the risk level evaluation results. The execution process is as follows; The data patterns in the pattern comparison results are classified into safety levels. This process involves evaluating the probability of accidents under multiple combinations of conditions, verifying the degree of danger of various combinations of conditions through simulation tests, and using data analysis and machine learning techniques to classify and probabilistically calculate data patterns. This includes algorithms such as decision trees, random forests, and neural networks to predict and classify risk levels, and optimize algorithm parameters to improve prediction accuracy and reliability. At the same time, actual data is used in simulation tests to test condition combinations. The test results are analyzed through specialized software tools to ensure the accuracy and practicality of the evaluation results, which are used to guide risk management and prevention and control strategies in daily operations, effectively avoid security threats, and generate risk level assessment results.

[0027] S203: According to the risk level assessment results, key risk factors are determined, and targeted preventive measures are implemented to avoid the risk of accidents by adjusting operating parameters and optimizing safety monitoring. The execution process of generating risk identification results is as follows; Based on the risk level assessment results, key risk factors are identified and targeted preventive measures are implemented. This process includes adjusting operating parameters and optimizing the safety monitoring system to avoid accident risks. Data analysts and engineers work together to use the risk assessment results to determine the operating parameters that need to be adjusted, including adjusting speed limits, load limits or operating environmental conditions. The safety monitoring system is optimized by introducing advanced sensors and monitoring technologies to ensure that any abnormal operations or environmental changes that may lead to accidents can be monitored in real time. Real-time data analysis is used to predict and prevent potential accidents, significantly improving operational safety, providing a scientific basis and technical support for avoiding future accidents, and generating risk identification results.

[0028] See also Figure 4 By comparing the risk identification results with normal operation data, using AI, applying the weighted calculation of risk factors, updating and calculating the real-time risk score, and accumulating the impact of differentiation factors, the steps to obtain the dynamic risk score are as follows: S301: Based on the risk identification results, combined with normal operation data, AI technology is used to perform data comparison and analysis, extract the key differences between risk factors and normal factors, and generate the difference analysis results. The execution process is as follows; Use AI technology to compare and analyze data, extract the key differences between risk factors and normal factors, and use the formula , calculate the difference analysis results between risk and normal operation. In the formula, represents the sum of key differences, Represents risk operation data, Represents normal operating data, Represents the number of data points. Formula explanation and formula calculation derivation process: With the help of AI technology, analyze and identify operational factors that are common in risk accidents but uncommon in normal operations. If 100 data points are analyzed, the average load of risk operations is 200 tons, while the average load of normal operations is 150 tons. The difference in load items can be calculated by the formula ,This analysis result helps to identify the risk factors that need to be monitored and adjusted during the operation.

[0029] S302: Based on the difference analysis results, the weighted average method is used to calculate the risk contribution values ​​of multiple factors, and multiple risk factors are weighted to generate the risk score update record. The execution process is as follows; The formula for the weighted average method is as follows: ; in, is the adjusted risk score, For the The weight coefficient of each risk factor is For the The raw risk score of each risk factor, is the risk frequency adjustment factor, It is the square root of the weighted sum.

[0030] Detailed explanation of the formula and the process of formula calculation: represents the adjusted risk score, which is used to quantify the overall risk level, where , and They represent the weight coefficient of risk factors, the original risk score and the risk frequency adjustment factor respectively. The formula obtains a comprehensive risk score by weighting and normalizing the risk factors.

[0031] Weight coefficient : The distribution of each risk factor based on its impact on safe operation is obtained from actual operation data and past accident analysis. If a factor appears in seven of the past ten accidents, it is given a higher weight.

[0032] Raw Risk Score :The original risk level of each risk factor is determined by a professional risk assessment team through field inspections and historical data analysis, and is calculated based on historical accident frequency and severity.

[0033] Risk Frequency Adjustment Factor : Reflects the frequency of occurrence of the risk factor within a certain period of time. It can be obtained through statistical analysis of historical operation data and calculated by the frequency of occurrence of the risk factor in the past year.

[0034] Set specific values ​​for the calculation example: set three risk factors and weights . Raw Risk Score . Risk Frequency .

[0035] 1. Calculate the weighted risk score: ; 2. Calculate the sum of the absolute values ​​of the product of the weight and the risk frequency: ; 3. Calculate the sum of the absolute values ​​of the product of the weight and the risk frequency under the square root: ; 4. Calculate the adjusted risk score: ; The results show that after combining the currently set weights, original scores and frequency factors, the adjusted risk score is about 53.43, which reflects the comprehensive safety risk level of the lifting machinery under the current operating conditions and frequency adjustment of risk factors. This risk score is used to guide safety threshold adjustments and risk management decisions in actual operations.

[0036] S303: Update records using risk scores, implement risk management strategies, dynamically adjust operating guidelines and safety measures, verify the safety of lifting machinery operations in a changing working environment, and obtain the following execution process for dynamic risk scores; Adopt risk scoring to update records and implement risk management strategies. During this process, operating guidelines and safety measures are dynamically adjusted to verify the safety of lifting machinery operation. The risk scoring system adjusts the score based on real-time data updates and historical accident records. Safety experts and data scientists work together to use data analysis to determine the changing trends and key influencing factors of risk scores. The effectiveness of the adjusted measures is verified through comparative analysis of simulation tests and actual operation data, which includes testing mechanical performance and operational safety under different working environments and operating conditions. Through these dynamic scoring and testing, operating manuals and safety guidelines are updated in real time to ensure that the operation of lifting machinery is maintained within the highest safety standards under any working conditions and obtain dynamic risk scores.

[0037] See also Figure 5 ,Use dynamic risk scoring to evaluate the current crane operation risk, and propose ,adjustment operation measures for safety hazards. ,By changing the speed and path and triggering emergency stop, the steps for generating a safety ,adjustment strategy are as follows: S401: Analyze the current crane operation risk by using dynamic risk scoring, identify the operation parameters that exceed the safety threshold, determine the operation mode and environmental conditions that need to be adjusted, and generate the risk limit identification result. The execution process is as follows; Dynamic risk scoring is used to analyze the current operating risks of lifting machinery. The process involves identifying operating parameters that exceed safety thresholds through a real-time monitoring system. Data analysts use statistical and machine learning methods to analyze operating data in real time to determine the key risk factors that cause accidents, such as overweight, excessive speed or harsh environmental conditions. Safety engineers make necessary adjustments to operating modes and environmental conditions based on these data to ensure that all operating parameters are within the safety threshold. In this way, accidents can be prevented, the safety of lifting machinery in various working environments can be ensured, and risk over-limit identification results can be generated.

[0038] S402: Based on the risk over-limit identification result, the operation process is optimized to match the real-time environment and operation conditions by changing the speed and path of the mechanical operation, and the execution process of the operation adjustment measures is generated as follows; According to the risk over-limit identification results, by changing the speed and path of mechanical operation, according to the formula , calculate the speed adjustment of the operation adjustment measure. Where, represents the adjusted speed, represents the initial velocity, represents the identified risk score, Represents the maximum safety risk score. Detailed explanation of the formula and the process of formula calculation: Set the initial speed of the crane operation to 10m / s. According to the risk scoring system, the risk score of the current operation is 50, while the maximum safety risk score is 100. Using the formula This speed adjustment ensures that the safety of machinery operation in identified high-risk conditions is significantly improved.

[0039] S403: Implement operational adjustment measures, trigger emergency shutdown according to the severity of the risk, interrupt risk operations by implementing emergency response procedures, verify the safety of operators and equipment, and generate the execution process of the safety adjustment strategy as follows; Implement operational adjustment measures and trigger emergency shutdowns based on the severity of the risk, including interrupting risky operations by implementing emergency response procedures. The safety system is designed to automatically trigger a shutdown when a high-risk operation is detected to prevent potential accidents. Operators can respond quickly and perform emergency shutdown operations based on the real-time data and system warnings they receive. The emergency response procedure also includes multi-level safety verification to ensure that emergency shutdowns are activated only when they are really needed. This is critical to protecting the safety of operators and equipment, ensuring the controllability and safety of the entire operation process, and generating a safety adjustment strategy.

[0040] See also Figure 6 , implement the safety adjustment strategy, update the state definition of the lifting machinery operation, adjust the transition probability between states, match the real-time operation risk and environmental changes, and generate the updated state definition and probability model in the following steps: S501: Implement the safety adjustment strategy, redefine the operating state of the lifting machinery, update the parameters of the operating state to match the current risk score and environmental conditions, and generate the execution process of the state update result as follows; Implement safety adjustment strategies and redefine the operating status of the lifting machinery. This process adjusts the key parameters of the operating status, including operating speed, lifting weight, mechanical inclination and other data parameters, by analyzing the current risk score and environmental conditions. Real-time monitoring equipment collects key data on the operating status, and these data are classified and optimized through the data analysis platform to ensure that parameter updates match the current environmental conditions and risk levels. The data analysis algorithm will automatically apply the new parameters to the control system, so that the operating mode of the lifting machinery can adapt to real-time risks and changes in external conditions, and generate status update results.

[0041] S502: Based on the state update result, the transition probability between states is adjusted, the state transition model is optimized using real-time data feedback, and the state transition probability is verified according to the real-time operational risk and external environment changes. The execution process of generating the probability adjustment result is as follows; According to the state update result, adjust the transition probability between states according to the formula , calculate the adjustment probability of the state transition model. In the formula, Represents the state Transfer to state The probability of Represents historical data from arrive The transfer frequency, Represents the smoothing coefficient to avoid the denominator being zero. Formula explanation and formula calculation derivation process: Set the historical frequency of transition from state A to state B to 20 times, the total frequency of transition from state A to the remaining states to 50 times, and the smoothing coefficient to 1. Substitute into the formula This probability adjustment ensures that the state transition model can reflect real-time operational risks and environmental changes, further optimizing the operational decisions of the lifting machinery.

[0042] S503: Based on the probability adjustment result, the operation guide and risk management of the lifting machinery are updated, the consistency of the operation definition and probability model with the real-time operation environment and risk score is checked, and the execution process of generating the updated state definition and probability model is as follows; Based on the probability adjustment results, the operating guidelines and risk management of the lifting machinery are updated, and the consistency of the operating definition and probability model with the real-time operating environment and risk score is checked. The process includes importing the probability adjustment results into the risk management system, verifying the accuracy of the probability model through real-time monitoring feedback data, and refining the parameters in the model. The key operating processes in the operating guidelines are revised according to the new model and environmental data, including operating steps, risk monitoring processes, and emergency response strategies, to ensure that every detail of the operating guidelines is consistent with the current risk score and environmental changes, and to generate updated state definitions and probability models.

[0043] See also Figure 7 ,According to the updated state definition and probability model, the operation data of the lifting machinery is monitored in real time, the transition probability between differentiated states is calculated, and the effectiveness of the safety adjustment strategy is evaluated. The specific steps to obtain the safety assessment log are as follows: S601: Based on the updated state definition and probability model, the operation data of the lifting machinery, including position, speed and load, is monitored and tracked in real time, and the execution flow of generating real-time monitoring data is as follows; Based on the updated state definition and probability model, the operation data of the lifting machinery is monitored and tracked in real time. The process obtains the position information, speed and load data of the machinery through high-precision sensors, and uses edge computing devices to process these data in real time. The operation data is transmitted to the monitoring platform through wireless communication technology, and the data points collected every second are analyzed and stored. The monitoring data is also processed by filtering and anomaly detection algorithms to eliminate errors or noise. The data is used to judge the safety and efficiency of the machinery operation, provide comprehensive support for subsequent state transfer analysis and risk assessment, and generate real-time monitoring data.

[0044] S602: Using real-time monitoring data, calculate the transition probability between differentiated states, analyze state changes, update the risk assessment model to match the current operating state and environmental conditions, and generate the state transition analysis results. The execution process is as follows; Use real-time monitoring data to calculate the transition probability between differentiated states, according to the formula , calculate the state transition analysis results. In the formula, Represents the state Transfer to state The probability of Representative moments The state of The number of occurrences of Indicates from the state To status The transfer frequency, Represents the total number of observed times. Formula explanation and formula calculation derivation process: Assume that within a certain period of time, the transition from state A to state B occurs 30 times, and state A appears 100 times in total. Substitute into the formula The transition probability reflects the trend of state change and the dynamic characteristics of the current operating environment, providing a quantitative basis for updating the risk assessment model.

[0045] S603: Based on the state transition analysis results, the effectiveness of the safety adjustment strategy is evaluated, and the execution process of generating a safety assessment log is as follows by comparing the change in the accident rate and the optimization efficiency and speed of the operation safety; According to the results of state transition analysis, the effectiveness of the safety adjustment strategy is evaluated. By collecting and comparing the changes in accident rates with the optimization efficiency and speed of operational safety, data analysis tools are used to compare and analyze historical and real-time data to calculate the reduction in accident rates. The execution time and efficiency of operations before and after the adjustment are compared. The improvement effect of the adjustment strategy is evaluated by calculating the time difference and operation success rate. The risk changes and adaptability of the safety strategy for each operation are recorded in detail. The analysis results are quantified as improvement percentages to provide data support for subsequent optimization and generate safety assessment logs.

[0046] See also Figure 8 , a system for identifying safety hazards of lifting machinery based on AI, the system for identifying safety hazards of lifting machinery based on AI is used to execute the above-mentioned method for identifying safety hazards of lifting machinery based on AI, and the system includes: The data acquisition module collects the position, speed and load data of the crane, as well as the presence and location data of near-field obstacles to generate a comprehensive real-time data set; The risk analysis module uses comprehensive real-time data sets to detect risk factors in crane operation, identifies accident conditions by referring to accident data and operation patterns, and outputs risk identification results; The weighted calculation module evaluates the risk identification results, performs weighted calculation on each risk factor, identifies the immediate safety level of the lifting machinery operation, and obtains a dynamic risk score; The strategy building module evaluates the safety risk of the current operation based on the dynamic risk score, adjusts the operation parameters, including speed and path, triggers emergency shutdown when necessary, and obtains the safety adjustment strategy; The state update module implements the safety adjustment strategy, updates the operating state definition of the lifting machinery, adjusts the transition probability between states, matches real-time operational risks and environmental changes, evaluates the effectiveness of the safety adjustment strategy, and generates a safety assessment log.

[0047] The above are only preferred embodiments of the present invention and are not intended to limit the present invention in other forms. Any technician familiar with the profession may use the technical contents disclosed above to change or modify them into equivalent embodiments with equivalent changes and apply them to other fields. However, any simple modification, equivalent change and modification made to the above embodiments based on the technical essence of the present invention without departing from the technical solution of the present invention still falls within the protection scope of the technical solution of the present invention.

Claims

1. A method for identifying safety hazards of lifting machinery based on AI, characterized in that: The following steps are involved: Collects operational data of lifting machinery, including position, speed and load, while integrating information on meteorological conditions and near-field obstacles, combined with real-time transmission data to generate comprehensive real-time data; Based on the comprehensive real-time data, potential risk factors are analyzed, and according to the accident data and current operation mode of the lifting machinery, the combination of conditions leading to the accident is identified, and a risk identification result is generated; By comparing the risk identification results with normal operation data, using AI, applying weighted calculation of risk factors, updating and calculating real-time risk scores, and accumulating the impact of differentiation factors, a dynamic risk score is obtained; Using the dynamic risk score, the current crane operation risk is evaluated, and adjustment operation measures are proposed for potential safety hazards, and a safety adjustment strategy is generated by changing the speed and path and triggering an emergency stop; Implement the safety adjustment strategy, update the state definition of the lifting machinery operation, adjust the transition probability between states, match the real-time operation risk and environmental changes, and generate an updated state definition and probability model; According to the updated state definition and probability model, the crane operation data is monitored in real time, the transition probability between differentiated states is calculated, the effectiveness of the safety adjustment strategy is evaluated, and a safety assessment log is obtained.

2. The method for identifying safety hazards of lifting machinery based on AI according to claim 1 is characterized in that: The comprehensive real-time data includes the location coordinates of the lifting machinery, the current operating speed value and real-time load capacity, the associated meteorological conditions and the types of nearby obstacles. The risk identification results include the combination of identified risky operating modes, abnormal load states and adverse meteorological conditions. The dynamic risk score includes the numerical score calculated according to the weights of differentiated risk factors, the real-time update frequency of the score and the differentiated impact. The safety adjustment strategy includes speed reduction, path change measures and emergency shutdown trigger conditions. The updated state definition and probability model include new operating states, transfer conditions and probability values ​​of state transfer. The safety assessment log includes the recorded execution of the safety adjustment strategy, state transfer results and risk assessment update results.

3. The method for identifying safety hazards of lifting machinery based on AI according to claim 1 is characterized in that: The steps to collect the operation data of the lifting machinery, including position, speed and load, integrate the information of meteorological conditions and near-field obstacles, and combine with the real-time transmission data to generate comprehensive real-time data are as follows: Collect the operation data of the lifting machinery, use sensors to monitor and record the current position, speed and load of the machinery, and update it in real time to generate an operation data set; Based on the operation data set, environmental monitoring equipment is used to obtain meteorological condition data, and near-field obstacle detectors are deployed to collect obstacle information to generate an environmental data set; Through the environmental data set, the real-time transmission of the operation data of the lifting machinery is carried out, the real-time update of the data is verified and the integrity of the data is checked to generate comprehensive real-time data.

4. The method for identifying safety hazards of lifting machinery based on AI according to claim 1 is characterized in that: Based on the comprehensive real-time data, potential risk factors are analyzed, and according to the accident data and current operation mode of the lifting machinery, the combination of conditions leading to the accident is identified, and the steps of generating the risk identification result are specifically as follows: Based on the comprehensive real-time data, by retrospectively analyzing the accident data of the lifting machinery, extracting the operation mode and environmental conditions where the failure occurred, combining the real-time operation data, identifying and recording the data pattern associated with the accident, and generating a pattern comparison result; Classifying the data patterns in the pattern comparison results into safety levels, evaluating the probability of accidents under multiple combinations of conditions, verifying the dangerousness of the condition combinations through simulation tests, and generating risk level evaluation results; Based on the risk level assessment results, key risk factors are determined, targeted preventive measures are implemented, and the risk of accidents is avoided by adjusting operating parameters and optimizing safety monitoring to generate risk identification results.

5. The method for identifying safety hazards of lifting machinery based on AI according to claim 1 is characterized in that: By comparing the risk identification results with normal operation data, using AI, applying weighted calculation of risk factors, updating and calculating real-time risk scores, and accumulating the impact of differentiation factors, the steps for obtaining dynamic risk scores are as follows: Based on the risk identification results, combined with normal operation data, AI technology is used to perform data comparison and analysis, extract key differences between risk factors and normal factors, and generate difference analysis results; Based on the difference analysis results, the weighted average method is used to calculate the risk contribution values ​​of multiple factors, and multiple risk factors are weighted to generate a risk score update record; The risk score is used to update records, implement risk management strategies, dynamically adjust operating guidelines and safety measures, verify the safety of lifting machinery operations in a changing working environment, and obtain a dynamic risk score.

6. The method for identifying safety hazards of lifting machinery based on AI according to claim 5 is characterized in that: The formula of the weighted average method is as follows: ； in, is the adjusted risk score, For the The weight coefficient of each risk factor is For the The raw risk score of each risk factor, is the risk frequency adjustment factor, It is the square root of the weighted sum.

7. The method for identifying safety hazards of lifting machinery based on AI according to claim 1 is characterized in that: The dynamic risk score is used to evaluate the current crane operation risk, and to propose adjustment operation measures for potential safety hazards. The steps of generating a safety adjustment strategy by changing the speed and path and triggering an emergency stop are as follows: Analyze the current crane operation risk by using the dynamic risk score, identify the operation parameters that exceed the safety threshold, determine the operation mode and environmental conditions that need to be adjusted, and generate a risk limit identification result; According to the risk over-limit identification result, by changing the speed and path of the mechanical operation, the operation process is optimized to match the real-time environment and operating conditions, and operation adjustment measures are generated; Implement the operational adjustment measures, trigger emergency shutdown according to the severity of the risk, interrupt risk operations by implementing emergency response procedures, verify the safety of operators and equipment, and generate a safety adjustment strategy.

8. The method for identifying safety hazards of lifting machinery based on AI according to claim 1 is characterized in that: The steps of implementing the safety adjustment strategy, updating the state definition of the lifting machinery operation, adjusting the transition probability between states, matching the real-time operation risk and environmental changes, and generating the updated state definition and probability model are as follows: Implement the safety adjustment strategy, redefine the operating state of the lifting machinery, update the parameters of the operating state to match the current risk score and environmental conditions, and generate a state update result; Based on the state update results, the transition probability between states is adjusted, the state transition model is optimized using real-time data feedback, the state transition probability is verified according to real-time operational risks and external environment changes, and the probability adjustment result is generated; Based on the probability adjustment results, the operating guidelines and risk management of the lifting machinery are updated, the consistency of the operation definition and probability model with the real-time operating environment and risk score is checked, and an updated state definition and probability model are generated.

9. The method for identifying safety hazards of lifting machinery based on AI according to claim 1 is characterized in that: According to the updated state definition and probability model, the operation data of the lifting machinery is monitored in real time, the transition probability between the differentiated states is calculated, and the effectiveness of the safety adjustment strategy is evaluated. The specific steps of obtaining the safety assessment log are as follows: Based on the updated state definition and probability model, real-time monitoring and tracking of operation data of the lifting machinery, including position, speed and load, generates real-time monitoring data; Using the real-time monitoring data, the transition probability between differentiated states is calculated, and by analyzing the state changes, the risk assessment model is updated to match the current operating state and environmental conditions, and a state transition analysis result is generated; Based on the state transition analysis results, the effectiveness of the safety adjustment strategy is evaluated, and a safety assessment log is generated by comparing the changes in the accident rate and the optimization efficiency and speed of operational safety.

10. A system for identifying safety hazards of lifting machinery based on AI, characterized in that: According to the method for identifying safety hazards of lifting machinery based on AI according to any one of claims 1 to 9, the system comprises: The data acquisition module collects the position, speed and load data of the crane, as well as the presence and location data of near-field obstacles to generate a comprehensive real-time data set; The risk analysis module uses the comprehensive real-time data set to detect risk factors in the operation of the lifting machinery, identifies the conditions for the occurrence of accidents by referring to the accident data and operation mode, and outputs the risk identification results; The weighted calculation module evaluates the risk identification result, performs weighted calculation on each risk factor, identifies the instant safety level of the lifting machinery operation, and obtains a dynamic risk score; The strategy building module evaluates the safety risk of the current operation according to the dynamic risk score, adjusts the operation parameters, including speed and path, triggers an emergency shutdown when necessary, and obtains a safety adjustment strategy; The state update module implements the safety adjustment strategy, updates the operating state definition of the lifting machinery, adjusts the transition probability between states, matches real-time operating risks and environmental changes, evaluates the effectiveness of the safety adjustment strategy, and generates a safety assessment log.

Images